@prometheus-io/lezer-promql 0.37.0-rc.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md ADDED
@@ -0,0 +1,2245 @@
1
+ # Changelog
2
+
3
+ ## 2.37.0-rc.1 / 2022-07-11
4
+
5
+ This release candidate addresses an issue with the publishing of the Prometheus
6
+ UI libraries to NPM registry.
7
+
8
+ ## 2.37.0-rc.0 / 2022-07-05
9
+
10
+ Following data loss by users due to lack of unified buffer cache in OpenBSD, we
11
+ will no longer release Prometheus upstream for OpenBSD until a proper solution is
12
+ found. #8799
13
+
14
+ * [FEATURE] Nomad SD: New service discovery for Nomad built-in service discovery. #10915
15
+ * [ENHANCEMENT] Kubernetes SD: Allow attaching node labels for endpoint role. #10759
16
+ * [ENHANCEMENT] PromQL: Optimise creation of signature with/without labels. #10667
17
+ * [ENHANCEMENT] TSDB: Memory optimizations. #10873 #10874
18
+ * [ENHANCEMENT] TSDB: Reduce sleep time when reading WAL. #10859 #10878
19
+ * [BUGFIX] Alerting: Fix Alertmanager targets not being updated when alerts were queued. #10948
20
+ * [BUGFIX] Hetzner SD: Make authentication files relative to Prometheus config file. #10813
21
+ * [BUGFIX] Promtool: Fix `promtool check config` not erroring properly on failures. #10952
22
+ * [BUGFIX] Scrape: Keep relabeled scrape interval and timeout on reloads. #10916
23
+ * [BUGFIX] TSDB: Don't increment `prometheus_tsdb_compactions_failed_total` when context is canceled. #10772
24
+ * [BUGFIX] TSDB: Fix panic if series is not found when deleting series. #10907
25
+ * [BUGFIX] TSDB: Increase `prometheus_tsdb_mmap_chunk_corruptions_total` on out of sequence errors. #10406
26
+ * [BUGFIX] Uyuni SD: Make authentication files relative to Prometheus configuration file and fix default configuration values. #10813
27
+
28
+ ## 2.36.2 / 2022-06-20
29
+
30
+ * [BUGFIX] Fix serving of static assets like fonts and favicon. #10888
31
+
32
+ ## 2.36.1 / 2022-06-09
33
+
34
+ * [BUGFIX] promtool: Add --lint-fatal option. #10840
35
+
36
+ ## 2.36.0 / 2022-05-30
37
+
38
+ * [FEATURE] Add lowercase and uppercase relabel action. #10641
39
+ * [FEATURE] SD: Add IONOS Cloud integration. #10514
40
+ * [FEATURE] SD: Add Vultr integration. #10714
41
+ * [FEATURE] SD: Add Linode SD failure count metric. #10673
42
+ * [FEATURE] Add prometheus_ready metric. #10682
43
+ * [ENHANCEMENT] Add stripDomain to template function. #10475
44
+ * [ENHANCEMENT] UI: Enable active search through dropped targets. #10668
45
+ * [ENHANCEMENT] promtool: support matchers when querying label values. #10727
46
+ * [ENHANCEMENT] Add agent mode identifier. #9638
47
+ * [BUGFIX] Changing TotalQueryableSamples from int to int64. #10549
48
+ * [BUGFIX] tsdb/agent: Ignore duplicate exemplars. #10595
49
+ * [BUGFIX] TSDB: Fix chunk overflow appending samples at a variable rate. #10607
50
+ * [BUGFIX] Stop rule manager before TSDB is stopped. #10680
51
+
52
+ ## 2.35.0 / 2022-04-21
53
+
54
+ This Prometheus release is built with go1.18, which contains two noticeable changes related to TLS:
55
+
56
+ 1. [TLS 1.0 and 1.1 disabled by default client-side](https://go.dev/doc/go1.18#tls10).
57
+ Prometheus users can override this with the `min_version` parameter of [tls_config](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#tls_config).
58
+ 2. [Certificates signed with the SHA-1 hash function are rejected](https://go.dev/doc/go1.18#sha1). This doesn't apply to self-signed root certificates.
59
+
60
+ * [CHANGE] TSDB: Delete `*.tmp` WAL files when Prometheus starts. #10317
61
+ * [CHANGE] promtool: Add new flag `--lint` (enabled by default) for the commands `check rules` and `check config`, resulting in a new exit code (`3`) for linter errors. #10435
62
+ * [FEATURE] Support for automatically setting the variable `GOMAXPROCS` to the container CPU limit. Enable with the flag `--enable-feature=auto-gomaxprocs`. #10498
63
+ * [FEATURE] PromQL: Extend statistics with total and peak number of samples in a query. Additionally, per-step statistics are available with --enable-feature=promql-per-step-stats and using `stats=all` in the query API.
64
+ Enable with the flag `--enable-feature=per-step-stats`. #10369
65
+ * [ENHANCEMENT] Prometheus is built with Go 1.18. #10501
66
+ * [ENHANCEMENT] TSDB: more efficient sorting of postings read from WAL at startup. #10500
67
+ * [ENHANCEMENT] Azure SD: Add metric to track Azure SD failures. #10476
68
+ * [ENHANCEMENT] Azure SD: Add an optional `resource_group` configuration. #10365
69
+ * [ENHANCEMENT] Kubernetes SD: Support `discovery.k8s.io/v1` `EndpointSlice` (previously only `discovery.k8s.io/v1beta1` `EndpointSlice` was supported). #9570
70
+ * [ENHANCEMENT] Kubernetes SD: Allow attaching node metadata to discovered pods. #10080
71
+ * [ENHANCEMENT] OAuth2: Support for using a proxy URL to fetch OAuth2 tokens. #10492
72
+ * [ENHANCEMENT] Configuration: Add the ability to disable HTTP2. #10492
73
+ * [ENHANCEMENT] Config: Support overriding minimum TLS version. #10610
74
+ * [BUGFIX] Kubernetes SD: Explicitly include gcp auth from k8s.io. #10516
75
+ * [BUGFIX] Fix OpenMetrics parser to sort uppercase labels correctly. #10510
76
+ * [BUGFIX] UI: Fix scrape interval and duration tooltip not showing on target page. #10545
77
+ * [BUGFIX] Tracing/GRPC: Set TLS credentials only when insecure is false. #10592
78
+ * [BUGFIX] Agent: Fix ID collision when loading a WAL with multiple segments. #10587
79
+ * [BUGFIX] Remote-write: Fix a deadlock between Batch and flushing the queue. #10608
80
+
81
+ ## 2.34.0 / 2022-03-15
82
+
83
+ * [CHANGE] UI: Classic UI removed. #10208
84
+ * [CHANGE] Tracing: Migrate from Jaeger to OpenTelemetry based tracing. #9724, #10203, #10276
85
+ * [ENHANCEMENT] TSDB: Disable the chunk write queue by default and allow configuration with the experimental flag `--storage.tsdb.head-chunks-write-queue-size`. #10425
86
+ * [ENHANCEMENT] HTTP SD: Add a failure counter. #10372
87
+ * [ENHANCEMENT] Azure SD: Set Prometheus User-Agent on requests. #10209
88
+ * [ENHANCEMENT] Uyuni SD: Reduce the number of logins to Uyuni. #10072
89
+ * [ENHANCEMENT] Scrape: Log when an invalid media type is encountered during a scrape. #10186
90
+ * [ENHANCEMENT] Scrape: Accept application/openmetrics-text;version=1.0.0 in addition to version=0.0.1. #9431
91
+ * [ENHANCEMENT] Remote-read: Add an option to not use external labels as selectors for remote read. #10254
92
+ * [ENHANCEMENT] UI: Optimize the alerts page and add a search bar. #10142
93
+ * [ENHANCEMENT] UI: Improve graph colors that were hard to see. #10179
94
+ * [ENHANCEMENT] Config: Allow escaping of `$` with `$$` when using environment variables with external labels. #10129
95
+ * [BUGFIX] PromQL: Properly return an error from histogram_quantile when metrics have the same labelset. #10140
96
+ * [BUGFIX] UI: Fix bug that sets the range input to the resolution. #10227
97
+ * [BUGFIX] TSDB: Fix a query panic when `memory-snapshot-on-shutdown` is enabled. #10348
98
+ * [BUGFIX] Parser: Specify type in metadata parser errors. #10269
99
+ * [BUGFIX] Scrape: Fix label limit changes not applying. #10370
100
+
101
+ ## 2.33.5 / 2022-03-08
102
+
103
+ The binaries published with this release are built with Go1.17.8 to avoid [CVE-2022-24921](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-24921).
104
+
105
+ * [BUGFIX] Remote-write: Fix deadlock between adding to queue and getting batch. #10395
106
+
107
+ ## 2.33.4 / 2022-02-22
108
+
109
+ * [BUGFIX] TSDB: Fix panic when m-mapping head chunks onto the disk. #10316
110
+
111
+ ## 2.33.3 / 2022-02-11
112
+
113
+ * [BUGFIX] Azure SD: Fix a regression when public IP Address isn't set. #10289
114
+
115
+ ## 2.33.2 / 2022-02-11
116
+
117
+ * [BUGFIX] Azure SD: Fix panic when public IP Address isn't set. #10280
118
+ * [BUGFIX] Remote-write: Fix deadlock when stopping a shard. #10279
119
+
120
+ ## 2.33.1 / 2022-02-02
121
+
122
+ * [BUGFIX] SD: Fix _no such file or directory_ in K8s SD when not running inside K8s. #10235
123
+
124
+ ## 2.33.0 / 2022-01-29
125
+
126
+ * [CHANGE] PromQL: Promote negative offset and `@` modifer to stable features. #10121
127
+ * [CHANGE] Web: Promote remote-write-receiver to stable. #10119
128
+ * [FEATURE] Config: Add `stripPort` template function. #10002
129
+ * [FEATURE] Promtool: Add cardinality analysis to `check metrics`, enabled by flag `--extended`. #10045
130
+ * [FEATURE] SD: Enable target discovery in own K8s namespace. #9881
131
+ * [FEATURE] SD: Add provider ID label in K8s SD. #9603
132
+ * [FEATURE] Web: Add limit field to the rules API. #10152
133
+ * [ENHANCEMENT] Remote-write: Avoid allocations by buffering concrete structs instead of interfaces. #9934
134
+ * [ENHANCEMENT] Remote-write: Log time series details for out-of-order samples in remote write receiver. #9894
135
+ * [ENHANCEMENT] Remote-write: Shard up more when backlogged. #9274
136
+ * [ENHANCEMENT] TSDB: Use simpler map key to improve exemplar ingest performance. #10111
137
+ * [ENHANCEMENT] TSDB: Avoid allocations when popping from the intersected postings heap. #10092
138
+ * [ENHANCEMENT] TSDB: Make chunk writing non-blocking, avoiding latency spikes in remote-write. #10051
139
+ * [ENHANCEMENT] TSDB: Improve label matching performance. #9907
140
+ * [ENHANCEMENT] UI: Optimize the service discovery page and add a search bar. #10131
141
+ * [ENHANCEMENT] UI: Optimize the target page and add a search bar. #10103
142
+ * [BUGFIX] Promtool: Make exit codes more consistent. #9861
143
+ * [BUGFIX] Promtool: Fix flakiness of rule testing. #8818
144
+ * [BUGFIX] Remote-write: Update `prometheus_remote_storage_queue_highest_sent_timestamp_seconds` metric when write irrecoverably fails. #10102
145
+ * [BUGFIX] Storage: Avoid panic in `BufferedSeriesIterator`. #9945
146
+ * [BUGFIX] TSDB: CompactBlockMetas should produce correct mint/maxt for overlapping blocks. #10108
147
+ * [BUGFIX] TSDB: Fix logging of exemplar storage size. #9938
148
+ * [BUGFIX] UI: Fix overlapping click targets for the alert state checkboxes. #10136
149
+ * [BUGFIX] UI: Fix _Unhealthy_ filter on target page to actually display only _Unhealthy_ targets. #10103
150
+ * [BUGFIX] UI: Fix autocompletion when expression is empty. #10053
151
+ * [BUGFIX] TSDB: Fix deadlock from simultaneous GC and write. #10166
152
+
153
+ ## 2.32.1 / 2021-12-17
154
+
155
+ * [BUGFIX] Scrape: Fix reporting metrics when sample limit is reached during the report. #9996
156
+ * [BUGFIX] Scrape: Ensure that scrape interval and scrape timeout are always set. #10023
157
+ * [BUGFIX] TSDB: Expose and fix bug in iterators' `Seek()` method. #10030
158
+
159
+ ## 2.32.0 / 2021-12-09
160
+
161
+ This release introduces the Prometheus Agent, a new mode of operation for
162
+ Prometheus optimized for remote-write only scenarios. In this mode, Prometheus
163
+ does not generate blocks on the local filesystem and is not queryable locally.
164
+ Enable with `--enable-feature=agent`.
165
+
166
+ Learn more about the Prometheus Agent in our [blog post](https://prometheus.io/blog/2021/11/16/agent/).
167
+
168
+ * [CHANGE] Remote-write: Change default max retry time from 100ms to 5 seconds. #9634
169
+ * [FEATURE] Agent: New mode of operation optimized for remote-write only scenarios, without local storage. Enable with `--enable-feature=agent`. #8785 #9851 #9664 #9939 #9941 #9943
170
+ * [FEATURE] Promtool: Add `promtool check service-discovery` command. #8970
171
+ * [FEATURE] UI: Add search in metrics dropdown. #9629
172
+ * [FEATURE] Templates: Add parseDuration to template functions. #8817
173
+ * [ENHANCEMENT] Promtool: Improve test output. #8064
174
+ * [ENHANCEMENT] PromQL: Use kahan summation for better numerical stability. #9588
175
+ * [ENHANCEMENT] Remote-write: Reuse memory for marshalling. #9412
176
+ * [ENHANCEMENT] Scrape: Add `scrape_body_size_bytes` scrape metric behind the `--enable-feature=extra-scrape-metrics` flag. #9569
177
+ * [ENHANCEMENT] TSDB: Add windows arm64 support. #9703
178
+ * [ENHANCEMENT] TSDB: Optimize query by skipping unneeded sorting in TSDB. #9673
179
+ * [ENHANCEMENT] Templates: Support int and uint as datatypes for template formatting. #9680
180
+ * [ENHANCEMENT] UI: Prefer `rate` over `rad`, `delta` over `deg`, and `count` over `cos` in autocomplete. #9688
181
+ * [ENHANCEMENT] Linode SD: Tune API request page sizes. #9779
182
+ * [BUGFIX] TSDB: Add more size checks when writing individual sections in the index. #9710
183
+ * [BUGFIX] PromQL: Make `deriv()` return zero values for constant series. #9728
184
+ * [BUGFIX] TSDB: Fix panic when checkpoint directory is empty. #9687
185
+ * [BUGFIX] TSDB: Fix panic, out of order chunks, and race warning during WAL replay. #9856
186
+ * [BUGFIX] UI: Correctly render links for targets with IPv6 addresses that contain a Zone ID. #9853
187
+ * [BUGFIX] Promtool: Fix checking of `authorization.credentials_file` and `bearer_token_file` fields. #9883
188
+ * [BUGFIX] Uyuni SD: Fix null pointer exception during initialization. #9924 #9950
189
+ * [BUGFIX] TSDB: Fix queries after a failed snapshot replay. #9980
190
+
191
+ ## 2.31.2 / 2021-12-09
192
+
193
+ * [BUGFIX] TSDB: Fix queries after a failed snapshot replay. #9980
194
+
195
+ ## 2.31.1 / 2021-11-05
196
+
197
+ * [BUGFIX] SD: Fix a panic when the experimental discovery manager receives
198
+ targets during a reload. #9656
199
+
200
+ ## 2.31.0 / 2021-11-02
201
+
202
+ * [CHANGE] UI: Remove standard PromQL editor in favour of the codemirror-based editor. #9452
203
+ * [FEATURE] PromQL: Add trigonometric functions and `atan2` binary operator. #9239 #9248 #9515
204
+ * [FEATURE] Remote: Add support for exemplar in the remote write receiver endpoint. #9319 #9414
205
+ * [FEATURE] SD: Add PuppetDB service discovery. #8883
206
+ * [FEATURE] SD: Add Uyuni service discovery. #8190
207
+ * [FEATURE] Web: Add support for security-related HTTP headers. #9546
208
+ * [ENHANCEMENT] Azure SD: Add `proxy_url`, `follow_redirects`, `tls_config`. #9267
209
+ * [ENHANCEMENT] Backfill: Add `--max-block-duration` in `promtool create-blocks-from rules`. #9511
210
+ * [ENHANCEMENT] Config: Print human-readable sizes with unit instead of raw numbers. #9361
211
+ * [ENHANCEMENT] HTTP: Re-enable HTTP/2. #9398
212
+ * [ENHANCEMENT] Kubernetes SD: Warn user if number of endpoints exceeds limit. #9467
213
+ * [ENHANCEMENT] OAuth2: Add TLS configuration to token requests. #9550
214
+ * [ENHANCEMENT] PromQL: Several optimizations. #9365 #9360 #9362 #9552
215
+ * [ENHANCEMENT] PromQL: Make aggregations deterministic in instant queries. #9459
216
+ * [ENHANCEMENT] Rules: Add the ability to limit number of alerts or series. #9260 #9541
217
+ * [ENHANCEMENT] SD: Experimental discovery manager to avoid restarts upon reload. Disabled by default, enable with flag `--enable-feature=new-service-discovery-manager`. #9349 #9537
218
+ * [ENHANCEMENT] UI: Debounce timerange setting changes. #9359
219
+ * [BUGFIX] Backfill: Apply rule labels after query labels. #9421
220
+ * [BUGFIX] Scrape: Resolve conflicts between multiple exported label prefixes. #9479 #9518
221
+ * [BUGFIX] Scrape: Restart scrape loops when `__scrape_interval__` is changed. #9551
222
+ * [BUGFIX] TSDB: Fix memory leak in samples deletion. #9151
223
+ * [BUGFIX] UI: Use consistent margin-bottom for all alert kinds. #9318
224
+
225
+ ## 2.30.4 / 2021-12-09
226
+
227
+ * [BUGFIX] TSDB: Fix queries after a failed snapshot replay. #9980
228
+
229
+ ## 2.30.3 / 2021-10-05
230
+
231
+ * [BUGFIX] TSDB: Fix panic on failed snapshot replay. #9438
232
+ * [BUGFIX] TSDB: Don't fail snapshot replay with exemplar storage disabled when the snapshot contains exemplars. #9438
233
+
234
+ ## 2.30.2 / 2021-10-01
235
+
236
+ * [BUGFIX] TSDB: Don't error on overlapping m-mapped chunks during WAL replay. #9381
237
+
238
+ ## 2.30.1 / 2021-09-28
239
+
240
+ * [ENHANCEMENT] Remote Write: Redact remote write URL when used for metric label. #9383
241
+ * [ENHANCEMENT] UI: Redact remote write URL and proxy URL passwords in the `/config` page. #9408
242
+ * [BUGFIX] promtool rules backfill: Prevent creation of data before the start time. #9339
243
+ * [BUGFIX] promtool rules backfill: Do not query after the end time. #9340
244
+ * [BUGFIX] Azure SD: Fix panic when no computername is set. #9387
245
+
246
+ ## 2.30.0 / 2021-09-14
247
+
248
+ * [FEATURE] **experimental** TSDB: Snapshot in-memory chunks on shutdown for faster restarts. Behind `--enable-feature=memory-snapshot-on-shutdown` flag. #7229
249
+ * [FEATURE] **experimental** Scrape: Configure scrape interval and scrape timeout via relabeling using `__scrape_interval__` and `__scrape_timeout__` labels respectively. #8911
250
+ * [FEATURE] Scrape: Add `scrape_timeout_seconds` and `scrape_sample_limit` metric. Behind `--enable-feature=extra-scrape-metrics` flag to avoid additional cardinality by default. #9247 #9295
251
+ * [ENHANCEMENT] Scrape: Add `--scrape.timestamp-tolerance` flag to adjust scrape timestamp tolerance when enabled via `--scrape.adjust-timestamps`. #9283
252
+ * [ENHANCEMENT] Remote Write: Improve throughput when sending exemplars. #8921
253
+ * [ENHANCEMENT] TSDB: Optimise WAL loading by removing extra map and caching min-time #9160
254
+ * [ENHANCEMENT] promtool: Speed up checking for duplicate rules. #9262/#9306
255
+ * [ENHANCEMENT] Scrape: Reduce allocations when parsing the metrics. #9299
256
+ * [ENHANCEMENT] docker_sd: Support host network mode #9125
257
+ * [BUGFIX] Exemplars: Fix panic when resizing exemplar storage from 0 to a non-zero size. #9286
258
+ * [BUGFIX] TSDB: Correctly decrement `prometheus_tsdb_head_active_appenders` when the append has no samples. #9230
259
+ * [BUGFIX] promtool rules backfill: Return 1 if backfill was unsuccessful. #9303
260
+ * [BUGFIX] promtool rules backfill: Avoid creation of overlapping blocks. #9324
261
+ * [BUGFIX] config: Fix a panic when reloading configuration with a `null` relabel action. #9224
262
+
263
+ ## 2.29.2 / 2021-08-27
264
+
265
+ * [BUGFIX] Fix Kubernetes SD failing to discover Ingress in Kubernetes v1.22. #9205
266
+ * [BUGFIX] Fix data race in loading write-ahead-log (WAL). #9259
267
+
268
+ ## 2.29.1 / 2021-08-11
269
+
270
+ * [BUGFIX] tsdb: align atomically accessed int64 to prevent panic in 32-bit
271
+ archs. #9192
272
+
273
+ ## 2.29.0 / 2021-08-11
274
+
275
+ Note for macOS users: Due to [changes in the upcoming Go 1.17](https://tip.golang.org/doc/go1.17#darwin),
276
+ this is the last Prometheus release that supports macOS 10.12 Sierra.
277
+
278
+ * [CHANGE] Promote `--storage.tsdb.allow-overlapping-blocks` flag to stable. #9117
279
+ * [CHANGE] Promote `--storage.tsdb.retention.size` flag to stable. #9004
280
+ * [FEATURE] Add Kuma service discovery. #8844
281
+ * [FEATURE] Add `present_over_time` PromQL function. #9097
282
+ * [FEATURE] Allow configuring exemplar storage via file and make it reloadable. #8974
283
+ * [FEATURE] UI: Allow selecting time range with mouse drag. #8977
284
+ * [FEATURE] promtool: Add feature flags flag `--enable-feature`. #8958
285
+ * [FEATURE] promtool: Add file_sd file validation. #8950
286
+ * [ENHANCEMENT] Reduce blocking of outgoing remote write requests from series garbage collection. #9109
287
+ * [ENHANCEMENT] Improve write-ahead-log decoding performance. #9106
288
+ * [ENHANCEMENT] Improve append performance in TSDB by reducing mutexes usage. #9061
289
+ * [ENHANCEMENT] Allow configuring `max_samples_per_send` for remote write metadata. #8959
290
+ * [ENHANCEMENT] Add `__meta_gce_interface_ipv4_<name>` meta label to GCE discovery. #8978
291
+ * [ENHANCEMENT] Add `__meta_ec2_availability_zone_id` meta label to EC2 discovery. #8896
292
+ * [ENHANCEMENT] Add `__meta_azure_machine_computer_name` meta label to Azure discovery. #9112
293
+ * [ENHANCEMENT] Add `__meta_hetzner_hcloud_labelpresent_<labelname>` meta label to Hetzner discovery. #9028
294
+ * [ENHANCEMENT] promtool: Add compaction efficiency to `promtool tsdb analyze` reports. #8940
295
+ * [ENHANCEMENT] promtool: Allow configuring max block duration for backfilling via `--max-block-duration` flag. #8919
296
+ * [ENHANCEMENT] UI: Add sorting and filtering to flags page. #8988
297
+ * [ENHANCEMENT] UI: Improve alerts page rendering performance. #9005
298
+ * [BUGFIX] Log when total symbol size exceeds 2^32 bytes, causing compaction to fail, and skip compaction. #9104
299
+ * [BUGFIX] Fix incorrect `target_limit` reloading of zero value. #9120
300
+ * [BUGFIX] Fix head GC and pending readers race condition. #9081
301
+ * [BUGFIX] Fix timestamp handling in OpenMetrics parser. #9008
302
+ * [BUGFIX] Fix potential duplicate metrics in `/federate` endpoint when specifying multiple matchers. #8885
303
+ * [BUGFIX] Fix server configuration and validation for authentication via client cert. #9123
304
+ * [BUGFIX] Allow `start` and `end` again as label names in PromQL queries. They were disallowed since the introduction of @ timestamp feature. #9119
305
+
306
+ ## 2.28.1 / 2021-07-01
307
+
308
+ * [BUGFIX]: HTTP SD: Allow `charset` specification in `Content-Type` header. #8981
309
+ * [BUGFIX]: HTTP SD: Fix handling of disappeared target groups. #9019
310
+ * [BUGFIX]: Fix incorrect log-level handling after moving to go-kit/log. #9021
311
+
312
+ ## 2.28.0 / 2021-06-21
313
+
314
+ * [CHANGE] UI: Make the new experimental PromQL editor the default. #8925
315
+ * [FEATURE] Linode SD: Add Linode service discovery. #8846
316
+ * [FEATURE] HTTP SD: Add generic HTTP-based service discovery. #8839
317
+ * [FEATURE] Kubernetes SD: Allow configuring API Server access via a kubeconfig file. #8811
318
+ * [FEATURE] UI: Add exemplar display support to the graphing interface. #8832 #8945 #8929
319
+ * [FEATURE] Consul SD: Add namespace support for Consul Enterprise. #8900
320
+ * [ENHANCEMENT] Promtool: Allow silencing output when importing / backfilling data. #8917
321
+ * [ENHANCEMENT] Consul SD: Support reading tokens from file. #8926
322
+ * [ENHANCEMENT] Rules: Add a new `.ExternalURL` alert field templating variable, containing the external URL of the Prometheus server. #8878
323
+ * [ENHANCEMENT] Scrape: Add experimental `body_size_limit` scrape configuration setting to limit the allowed response body size for target scrapes. #8833 #8886
324
+ * [ENHANCEMENT] Kubernetes SD: Add ingress class name label for ingress discovery. #8916
325
+ * [ENHANCEMENT] UI: Show a startup screen with progress bar when the TSDB is not ready yet. #8662 #8908 #8909 #8946
326
+ * [ENHANCEMENT] SD: Add a target creation failure counter `prometheus_target_sync_failed_total` and improve target creation failure handling. #8786
327
+ * [ENHANCEMENT] TSDB: Improve validation of exemplar label set length. #8816
328
+ * [ENHANCEMENT] TSDB: Add a `prometheus_tsdb_clean_start` metric that indicates whether a TSDB lockfile from a previous run still existed upon startup. #8824
329
+ * [BUGFIX] UI: In the experimental PromQL editor, fix autocompletion and parsing for special float values and improve series metadata fetching. #8856
330
+ * [BUGFIX] TSDB: When merging chunks, split resulting chunks if they would contain more than the maximum of 120 samples. #8582
331
+ * [BUGFIX] SD: Fix the computation of the `prometheus_sd_discovered_targets` metric when using multiple service discoveries. #8828
332
+
333
+ ## 2.27.1 / 2021-05-18
334
+
335
+ This release contains a bug fix for a security issue in the API endpoint. An
336
+ attacker can craft a special URL that redirects a user to any endpoint via an
337
+ HTTP 302 response. See the [security advisory][GHSA-vx57-7f4q-fpc7] for more details.
338
+
339
+ [GHSA-vx57-7f4q-fpc7]:https://github.com/prometheus/prometheus/security/advisories/GHSA-vx57-7f4q-fpc7
340
+
341
+ This vulnerability has been reported by Aaron Devaney from MDSec.
342
+
343
+ * [BUGFIX] SECURITY: Fix arbitrary redirects under the /new endpoint (CVE-2021-29622)
344
+
345
+ ## 2.27.0 / 2021-05-12
346
+
347
+ * [CHANGE] Remote write: Metric `prometheus_remote_storage_samples_bytes_total` renamed to `prometheus_remote_storage_bytes_total`. #8296
348
+ * [FEATURE] Promtool: Retroactive rule evaluation functionality. #7675
349
+ * [FEATURE] Configuration: Environment variable expansion for external labels. Behind `--enable-feature=expand-external-labels` flag. #8649
350
+ * [FEATURE] TSDB: Add a flag(`--storage.tsdb.max-block-chunk-segment-size`) to control the max chunks file size of the blocks for small Prometheus instances. #8478
351
+ * [FEATURE] UI: Add a dark theme. #8604
352
+ * [FEATURE] AWS Lightsail Discovery: Add AWS Lightsail Discovery. #8693
353
+ * [FEATURE] Docker Discovery: Add Docker Service Discovery. #8629
354
+ * [FEATURE] OAuth: Allow OAuth 2.0 to be used anywhere an HTTP client is used. #8761
355
+ * [FEATURE] Remote Write: Send exemplars via remote write. Experimental and disabled by default. #8296
356
+ * [ENHANCEMENT] Digital Ocean Discovery: Add `__meta_digitalocean_vpc` label. #8642
357
+ * [ENHANCEMENT] Scaleway Discovery: Read Scaleway secret from a file. #8643
358
+ * [ENHANCEMENT] Scrape: Add configurable limits for label size and count. #8777
359
+ * [ENHANCEMENT] UI: Add 16w and 26w time range steps. #8656
360
+ * [ENHANCEMENT] Templating: Enable parsing strings in `humanize` functions. #8682
361
+ * [BUGFIX] UI: Provide errors instead of blank page on TSDB Status Page. #8654 #8659
362
+ * [BUGFIX] TSDB: Do not panic when writing very large records to the WAL. #8790
363
+ * [BUGFIX] TSDB: Avoid panic when mmaped memory is referenced after the file is closed. #8723
364
+ * [BUGFIX] Scaleway Discovery: Fix nil pointer dereference. #8737
365
+ * [BUGFIX] Consul Discovery: Restart no longer required after config update with no targets. #8766
366
+
367
+ ## 2.26.0 / 2021-03-31
368
+
369
+ Prometheus is now built and supporting Go 1.16 (#8544). This reverts the memory release pattern added in Go 1.12. This makes common RSS usage metrics showing more accurate number for actual memory used by Prometheus. You can read more details [here](https://www.bwplotka.dev/2019/golang-memory-monitoring/).
370
+
371
+ Note that from this release Prometheus is using Alertmanager v2 by default.
372
+
373
+ * [CHANGE] Alerting: Using Alertmanager v2 API by default. #8626
374
+ * [CHANGE] Prometheus/Promtool: As agreed on dev summit, binaries are now printing help and usage to stdout instead of stderr. #8542
375
+ * [FEATURE] Remote: Add support for AWS SigV4 auth method for remote_write. #8509
376
+ * [FEATURE] Scaleway Discovery: Add Scaleway Service Discovery. #8555
377
+ * [FEATURE] PromQL: Allow negative offsets. Behind `--enable-feature=promql-negative-offset` flag. #8487
378
+ * [FEATURE] **experimental** Exemplars: Add in-memory storage for exemplars. Behind `--enable-feature=exemplar-storage` flag. #6635
379
+ * [FEATURE] UI: Add advanced auto-completion, syntax highlighting and linting to graph page query input. #8634
380
+ * [ENHANCEMENT] Digital Ocean Discovery: Add `__meta_digitalocean_image` label. #8497
381
+ * [ENHANCEMENT] PromQL: Add `last_over_time`, `sgn`, `clamp` functions. #8457
382
+ * [ENHANCEMENT] Scrape: Add support for specifying type of Authorization header credentials with Bearer by default. #8512
383
+ * [ENHANCEMENT] Scrape: Add `follow_redirects` option to scrape configuration. #8546
384
+ * [ENHANCEMENT] Remote: Allow retries on HTTP 429 response code for remote_write. Disabled by default. See [configuration docs](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#remote_write) for details. #8237 #8477
385
+ * [ENHANCEMENT] Remote: Allow configuring custom headers for remote_read. See [configuration docs](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#remote_read) for details. #8516
386
+ * [ENHANCEMENT] UI: Hitting Enter now triggers new query. #8581
387
+ * [ENHANCEMENT] UI: Better handling of long rule and names on the `/rules` and `/targets` pages. #8608 #8609
388
+ * [ENHANCEMENT] UI: Add collapse/expand all button on the `/targets` page. #8486
389
+ * [BUGFIX] TSDB: Eager deletion of removable blocks on every compaction, saving disk peak space usage. #8007
390
+ * [BUGFIX] PromQL: Fix parser support for special characters like`炬`. #8517
391
+ * [BUGFIX] Rules: Update rule health for append/commit fails. #8619
392
+
393
+ ## 2.25.2 / 2021-03-16
394
+
395
+ * [BUGFIX] Fix the ingestion of scrapes when the wall clock changes, e.g. on suspend. #8601
396
+
397
+ ## 2.25.1 / 2021-03-14
398
+
399
+ * [BUGFIX] Fix a crash in `promtool` when a subquery with default resolution is used. #8569
400
+ * [BUGFIX] Fix a bug that could return duplicate datapoints in queries. #8591
401
+ * [BUGFIX] Fix crashes with arm64 when compiled with go1.16. #8593
402
+
403
+ ## 2.25.0 / 2021-02-17
404
+
405
+ This release includes a new `--enable-feature=` flag that enables
406
+ experimental features. Such features might be changed or removed in the future.
407
+
408
+ In the next minor release (2.26), Prometheus will use the Alertmanager API v2.
409
+ It will be done by defaulting `alertmanager_config.api_version` to `v2`.
410
+ Alertmanager API v2 was released in Alertmanager v0.16.0 (released in January
411
+ 2019).
412
+
413
+ * [FEATURE] **experimental** API: Accept remote_write requests. Behind the --enable-feature=remote-write-receiver flag. #8424
414
+ * [FEATURE] **experimental** PromQL: Add `@ <timestamp>` modifier. Behind the --enable-feature=promql-at-modifier flag. #8121 #8436 #8425
415
+ * [ENHANCEMENT] Add optional name property to testgroup for better test failure output. #8440
416
+ * [ENHANCEMENT] Add warnings into React Panel on the Graph page. #8427
417
+ * [ENHANCEMENT] TSDB: Increase the number of buckets for the compaction duration metric. #8342
418
+ * [ENHANCEMENT] Remote: Allow passing along custom remote_write HTTP headers. #8416
419
+ * [ENHANCEMENT] Mixins: Scope grafana configuration. #8332
420
+ * [ENHANCEMENT] Kubernetes SD: Add endpoint labels metadata. #8273
421
+ * [ENHANCEMENT] UI: Expose total number of label pairs in head in TSDB stats page. #8343
422
+ * [ENHANCEMENT] TSDB: Reload blocks every minute, to detect new blocks and enforce retention more often. #8340
423
+ * [BUGFIX] API: Fix global URL when external address has no port. #8359
424
+ * [BUGFIX] Backfill: Fix error message handling. #8432
425
+ * [BUGFIX] Backfill: Fix "add sample: out of bounds" error when series span an entire block. #8476
426
+ * [BUGFIX] Deprecate unused flag --alertmanager.timeout. #8407
427
+ * [BUGFIX] Mixins: Support remote-write metrics renamed in v2.23 in alerts. #8423
428
+ * [BUGFIX] Remote: Fix garbage collection of dropped series in remote write. #8387
429
+ * [BUGFIX] Remote: Log recoverable remote write errors as warnings. #8412
430
+ * [BUGFIX] TSDB: Remove pre-2.21 temporary blocks on start. #8353.
431
+ * [BUGFIX] UI: Fix duplicated keys on /targets page. #8456
432
+ * [BUGFIX] UI: Fix label name leak into class name. #8459
433
+
434
+ ## 2.24.1 / 2021-01-20
435
+
436
+ * [ENHANCEMENT] Cache basic authentication results to significantly improve performance of HTTP endpoints (via an update of prometheus/exporter-toolkit).
437
+ * [BUGFIX] Prevent user enumeration by timing requests sent to authenticated HTTP endpoints (via an update of prometheus/exporter-toolkit).
438
+
439
+ ## 2.24.0 / 2021-01-06
440
+
441
+ * [FEATURE] Add TLS and basic authentication to HTTP endpoints. #8316
442
+ * [FEATURE] promtool: Add `check web-config` subcommand to check web config files. #8319
443
+ * [FEATURE] promtool: Add `tsdb create-blocks-from openmetrics` subcommand to backfill metrics data from an OpenMetrics file. #8084
444
+ * [ENHANCEMENT] HTTP API: Fast-fail queries with only empty matchers. #8288
445
+ * [ENHANCEMENT] HTTP API: Support matchers for labels API. #8301
446
+ * [ENHANCEMENT] promtool: Improve checking of URLs passed on the command line. #7956
447
+ * [ENHANCEMENT] SD: Expose IPv6 as a label in EC2 SD. #7086
448
+ * [ENHANCEMENT] SD: Reuse EC2 client, reducing frequency of requesting credentials. #8311
449
+ * [ENHANCEMENT] TSDB: Add logging when compaction takes more than the block time range. #8151
450
+ * [ENHANCEMENT] TSDB: Avoid unnecessary GC runs after compaction. #8276
451
+ * [BUGFIX] HTTP API: Avoid double-closing of channel when quitting multiple times via HTTP. #8242
452
+ * [BUGFIX] SD: Ignore CNAME records in DNS SD to avoid spurious `Invalid SRV record` warnings. #8216
453
+ * [BUGFIX] SD: Avoid config error triggered by valid label selectors in Kubernetes SD. #8285
454
+
455
+ ## 2.23.0 / 2020-11-26
456
+
457
+ * [CHANGE] UI: Make the React UI default. #8142
458
+ * [CHANGE] Remote write: The following metrics were removed/renamed in remote write. #6815
459
+ * `prometheus_remote_storage_succeeded_samples_total` was removed and `prometheus_remote_storage_samples_total` was introduced for all the samples attempted to send.
460
+ * `prometheus_remote_storage_sent_bytes_total` was removed and replaced with `prometheus_remote_storage_samples_bytes_total` and `prometheus_remote_storage_metadata_bytes_total`.
461
+ * `prometheus_remote_storage_failed_samples_total` -> `prometheus_remote_storage_samples_failed_total` .
462
+ * `prometheus_remote_storage_retried_samples_total` -> `prometheus_remote_storage_samples_retried_total`.
463
+ * `prometheus_remote_storage_dropped_samples_total` -> `prometheus_remote_storage_samples_dropped_total`.
464
+ * `prometheus_remote_storage_pending_samples` -> `prometheus_remote_storage_samples_pending`.
465
+ * [CHANGE] Remote: Do not collect non-initialized timestamp metrics. #8060
466
+ * [FEATURE] [EXPERIMENTAL] Remote write: Allow metric metadata to be propagated via remote write. The following new metrics were introduced: `prometheus_remote_storage_metadata_total`, `prometheus_remote_storage_metadata_failed_total`, `prometheus_remote_storage_metadata_retried_total`, `prometheus_remote_storage_metadata_bytes_total`. #6815
467
+ * [ENHANCEMENT] Remote write: Added a metric `prometheus_remote_storage_max_samples_per_send` for remote write. #8102
468
+ * [ENHANCEMENT] TSDB: Make the snapshot directory name always the same length. #8138
469
+ * [ENHANCEMENT] TSDB: Create a checkpoint only once at the end of all head compactions. #8067
470
+ * [ENHANCEMENT] TSDB: Avoid Series API from hitting the chunks. #8050
471
+ * [ENHANCEMENT] TSDB: Cache label name and last value when adding series during compactions making compactions faster. #8192
472
+ * [ENHANCEMENT] PromQL: Improved performance of Hash method making queries a bit faster. #8025
473
+ * [ENHANCEMENT] promtool: `tsdb list` now prints block sizes. #7993
474
+ * [ENHANCEMENT] promtool: Calculate mint and maxt per test avoiding unnecessary calculations. #8096
475
+ * [ENHANCEMENT] SD: Add filtering of services to Docker Swarm SD. #8074
476
+ * [BUGFIX] React UI: Fix button display when there are no panels. #8155
477
+ * [BUGFIX] PromQL: Fix timestamp() method for vector selector inside parenthesis. #8164
478
+ * [BUGFIX] PromQL: Don't include rendered expression on PromQL parse errors. #8177
479
+ * [BUGFIX] web: Fix panic with double close() of channel on calling `/-/quit/`. #8166
480
+ * [BUGFIX] TSDB: Fixed WAL corruption on partial writes within a page causing `invalid checksum` error on WAL replay. #8125
481
+ * [BUGFIX] Update config metrics `prometheus_config_last_reload_successful` and `prometheus_config_last_reload_success_timestamp_seconds` right after initial validation before starting TSDB.
482
+ * [BUGFIX] promtool: Correctly detect duplicate label names in exposition.
483
+
484
+ ## 2.22.2 / 2020-11-16
485
+
486
+ * [BUGFIX] Fix race condition in syncing/stopping/reloading scrapers. #8176
487
+
488
+ ## 2.22.1 / 2020-11-03
489
+
490
+ * [BUGFIX] Fix potential "mmap: invalid argument" errors in loading the head chunks, after an unclean shutdown, by performing read repairs. #8061
491
+ * [BUGFIX] Fix serving metrics and API when reloading scrape config. #8104
492
+ * [BUGFIX] Fix head chunk size calculation for size based retention. #8139
493
+
494
+ ## 2.22.0 / 2020-10-07
495
+
496
+ As announced in the 2.21.0 release notes, the experimental gRPC API v2 has been
497
+ removed.
498
+
499
+ * [CHANGE] web: Remove APIv2. #7935
500
+ * [ENHANCEMENT] React UI: Implement missing TSDB head stats section. #7876
501
+ * [ENHANCEMENT] UI: Add Collapse all button to targets page. #6957
502
+ * [ENHANCEMENT] UI: Clarify alert state toggle via checkbox icon. #7936
503
+ * [ENHANCEMENT] Add `rule_group_last_evaluation_samples` and `prometheus_tsdb_data_replay_duration_seconds` metrics. #7737 #7977
504
+ * [ENHANCEMENT] Gracefully handle unknown WAL record types. #8004
505
+ * [ENHANCEMENT] Issue a warning for 64 bit systems running 32 bit binaries. #8012
506
+ * [BUGFIX] Adjust scrape timestamps to align them to the intended schedule, effectively reducing block size. Workaround for a regression in go1.14+. #7976
507
+ * [BUGFIX] promtool: Ensure alert rules are marked as restored in unit tests. #7661
508
+ * [BUGFIX] Eureka: Fix service discovery when compiled in 32-bit. #7964
509
+ * [BUGFIX] Don't do literal regex matching optimisation when case insensitive. #8013
510
+ * [BUGFIX] Fix classic UI sometimes running queries for instant query when in range query mode. #7984
511
+
512
+ ## 2.21.0 / 2020-09-11
513
+
514
+ This release is built with Go 1.15, which deprecates [X.509 CommonName](https://golang.org/doc/go1.15#commonname)
515
+ in TLS certificates validation.
516
+
517
+ In the unlikely case that you use the gRPC API v2 (which is limited to TSDB
518
+ admin commands), please note that we will remove this experimental API in the
519
+ next minor release 2.22.
520
+
521
+ * [CHANGE] Disable HTTP/2 because of concerns with the Go HTTP/2 client. #7588 #7701
522
+ * [CHANGE] PromQL: `query_log_file` path is now relative to the config file. #7701
523
+ * [CHANGE] Promtool: Replace the tsdb command line tool by a promtool tsdb subcommand. #6088
524
+ * [CHANGE] Rules: Label `rule_group_iterations` metric with group name. #7823
525
+ * [FEATURE] Eureka SD: New service discovery. #3369
526
+ * [FEATURE] Hetzner SD: New service discovery. #7822
527
+ * [FEATURE] Kubernetes SD: Support Kubernetes EndpointSlices. #6838
528
+ * [FEATURE] Scrape: Add per scrape-config targets limit. #7554
529
+ * [ENHANCEMENT] Support composite durations in PromQL, config and UI, e.g. 1h30m. #7713 #7833
530
+ * [ENHANCEMENT] DNS SD: Add SRV record target and port meta labels. #7678
531
+ * [ENHANCEMENT] Docker Swarm SD: Support tasks and service without published ports. #7686
532
+ * [ENHANCEMENT] PromQL: Reduce the amount of data queried by remote read when a subquery has an offset. #7667
533
+ * [ENHANCEMENT] Promtool: Add `--time` option to query instant command. #7829
534
+ * [ENHANCEMENT] UI: Respect the `--web.page-title` parameter in the React UI. #7607
535
+ * [ENHANCEMENT] UI: Add duration, labels, annotations to alerts page in the React UI. #7605
536
+ * [ENHANCEMENT] UI: Add duration on the React UI rules page, hide annotation and labels if empty. #7606
537
+ * [BUGFIX] API: Deduplicate series in /api/v1/series. #7862
538
+ * [BUGFIX] PromQL: Drop metric name in bool comparison between two instant vectors. #7819
539
+ * [BUGFIX] PromQL: Exit with an error when time parameters can't be parsed. #7505
540
+ * [BUGFIX] Remote read: Re-add accidentally removed tracing for remote-read requests. #7916
541
+ * [BUGFIX] Rules: Detect extra fields in rule files. #7767
542
+ * [BUGFIX] Rules: Disallow overwriting the metric name in the `labels` section of recording rules. #7787
543
+ * [BUGFIX] Rules: Keep evaluation timestamp across reloads. #7775
544
+ * [BUGFIX] Scrape: Do not stop scrapes in progress during reload. #7752
545
+ * [BUGFIX] TSDB: Fix `chunks.HeadReadWriter: maxt of the files are not set` error. #7856
546
+ * [BUGFIX] TSDB: Delete blocks atomically to prevent corruption when there is a panic/crash during deletion. #7772
547
+ * [BUGFIX] Triton SD: Fix a panic when triton_sd_config is nil. #7671
548
+ * [BUGFIX] UI: Fix react UI bug with series going on and off. #7804
549
+ * [BUGFIX] UI: Fix styling bug for target labels with special names in React UI. #7902
550
+ * [BUGFIX] Web: Stop CMUX and GRPC servers even with stale connections, preventing the server to stop on SIGTERM. #7810
551
+
552
+ ## 2.20.1 / 2020-08-05
553
+
554
+ * [BUGFIX] SD: Reduce the Consul watch timeout to 2m and adjust the request timeout accordingly. #7724
555
+
556
+ ## 2.20.0 / 2020-07-22
557
+
558
+ This release changes WAL compression from opt-in to default. WAL compression will prevent a downgrade to v2.10 or earlier without deleting the WAL. Disable WAL compression explicitly by setting the command line flag `--no-storage.tsdb.wal-compression` if you require downgrading to v2.10 or earlier.
559
+
560
+ * [CHANGE] promtool: Changed rule numbering from 0-based to 1-based when reporting rule errors. #7495
561
+ * [CHANGE] Remote read: Added `prometheus_remote_storage_read_queries_total` counter and `prometheus_remote_storage_read_request_duration_seconds` histogram, removed `prometheus_remote_storage_remote_read_queries_total` counter.
562
+ * [CHANGE] Remote write: Added buckets for longer durations to `prometheus_remote_storage_sent_batch_duration_seconds` histogram.
563
+ * [CHANGE] TSDB: WAL compression is enabled by default. #7410
564
+ * [FEATURE] PromQL: Added `group()` aggregator. #7480
565
+ * [FEATURE] SD: Added Docker Swarm SD. #7420
566
+ * [FEATURE] SD: Added DigitalOcean SD. #7407
567
+ * [FEATURE] SD: Added Openstack config option to query alternative endpoints. #7494
568
+ * [ENHANCEMENT] Configuration: Exit early on invalid config file and signal it with exit code 2. #7399
569
+ * [ENHANCEMENT] PromQL: `without` is now a valid metric identifier. #7533
570
+ * [ENHANCEMENT] PromQL: Optimized regex label matching for literals within the pattern or as prefix/suffix. #7453 #7503
571
+ * [ENHANCEMENT] promtool: Added time range parameters for labels API in promtool. #7463
572
+ * [ENHANCEMENT] Remote write: Include samples waiting in channel in pending samples metric. Log number of dropped samples on hard shutdown. #7335
573
+ * [ENHANCEMENT] Scrape: Ingest synthetic scrape report metrics atomically with the corresponding scraped metrics. #7562
574
+ * [ENHANCEMENT] SD: Reduce timeouts for Openstack SD. #7507
575
+ * [ENHANCEMENT] SD: Use 10m timeout for Consul watches. #7423
576
+ * [ENHANCEMENT] SD: Added AMI meta label for EC2 SD. #7386
577
+ * [ENHANCEMENT] TSDB: Increment WAL corruption metric also on WAL corruption during checkpointing. #7491
578
+ * [ENHANCEMENT] TSDB: Improved query performance for high-cardinality labels. #7448
579
+ * [ENHANCEMENT] UI: Display dates as well as timestamps in status page. #7544
580
+ * [ENHANCEMENT] UI: Improved scrolling when following hash-fragment links. #7456
581
+ * [ENHANCEMENT] UI: React UI renders numbers in alerts in a more human-readable way. #7426
582
+ * [BUGFIX] API: Fixed error status code in the query API. #7435
583
+ * [BUGFIX] PromQL: Fixed `avg` and `avg_over_time` for NaN, Inf, and float64 overflows. #7346
584
+ * [BUGFIX] PromQL: Fixed off-by-one error in `histogram_quantile`. #7393
585
+ * [BUGFIX] promtool: Support extended durations in rules unit tests. #6297
586
+ * [BUGFIX] Scrape: Fix undercounting for `scrape_samples_post_metric_relabeling` in case of errors. #7342
587
+ * [BUGFIX] TSDB: Don't panic on WAL corruptions. #7550
588
+ * [BUGFIX] TSDB: Avoid leaving behind empty files in `chunks_head`, causing startup failures. #7573
589
+ * [BUGFIX] TSDB: Fixed race between compact (gc, populate) and head append causing unknown symbol error. #7560
590
+ * [BUGFIX] TSDB: Fixed unknown symbol error during head compaction. #7526
591
+ * [BUGFIX] TSDB: Fixed panic during TSDB metric registration. #7501
592
+ * [BUGFIX] TSDB: Fixed `--limit` command line flag in `tsdb` tool. #7430
593
+
594
+ ## 2.19.3 / 2020-07-24
595
+
596
+ * [BUGFIX] TSDB: Don't panic on WAL corruptions. #7550
597
+ * [BUGFIX] TSDB: Avoid leaving behind empty files in chunks_head, causing startup failures. #7573
598
+
599
+ ## 2.19.2 / 2020-06-26
600
+
601
+ * [BUGFIX] Remote Write: Fix panic when reloading config with modified queue parameters. #7452
602
+
603
+ ## 2.19.1 / 2020-06-18
604
+
605
+ * [BUGFIX] TSDB: Fix m-map file truncation leading to unsequential files. #7414
606
+
607
+ ## 2.19.0 / 2020-06-09
608
+
609
+ * [FEATURE] TSDB: Memory-map full chunks of Head (in-memory) block from disk. This reduces memory footprint and makes restarts faster. #6679
610
+ * [ENHANCEMENT] Discovery: Added discovery support for Triton global zones. #7250
611
+ * [ENHANCEMENT] Increased alert resend delay to be more tolerant towards failures. #7228
612
+ * [ENHANCEMENT] Remote Read: Added `prometheus_remote_storage_remote_read_queries_total` counter to count total number of remote read queries. #7328
613
+ * [ENHANCEMEMT] Added time range parameters for label names and label values API. #7288
614
+ * [ENHANCEMENT] TSDB: Reduced contention in isolation for high load. #7332
615
+ * [BUGFIX] PromQL: Eliminated collision while checking for duplicate labels. #7058
616
+ * [BUGFIX] React UI: Don't null out data when clicking on the current tab. #7243
617
+ * [BUGFIX] PromQL: Correctly track number of samples for a query. #7307
618
+ * [BUGFIX] PromQL: Return NaN when histogram buckets have 0 observations. #7318
619
+
620
+ ## 2.18.2 / 2020-06-09
621
+
622
+ * [BUGFIX] TSDB: Fix incorrect query results when using Prometheus with remote reads configured #7361
623
+
624
+ ## 2.18.1 / 2020-05-07
625
+
626
+ * [BUGFIX] TSDB: Fixed snapshot API. #7217
627
+
628
+ ## 2.18.0 / 2020-05-05
629
+
630
+ * [CHANGE] Federation: Only use local TSDB for federation (ignore remote read). #7096
631
+ * [CHANGE] Rules: `rule_evaluations_total` and `rule_evaluation_failures_total` have a `rule_group` label now. #7094
632
+ * [FEATURE] Tracing: Added experimental Jaeger support #7148
633
+ * [ENHANCEMENT] TSDB: Significantly reduce WAL size kept around after a block cut. #7098
634
+ * [ENHANCEMENT] Discovery: Add `architecture` meta label for EC2. #7000
635
+ * [BUGFIX] UI: Fixed wrong MinTime reported by /status. #7182
636
+ * [BUGFIX] React UI: Fixed multiselect legend on OSX. #6880
637
+ * [BUGFIX] Remote Write: Fixed blocked resharding edge case. #7122
638
+ * [BUGFIX] Remote Write: Fixed remote write not updating on relabel configs change. #7073
639
+
640
+ ## 2.17.2 / 2020-04-20
641
+
642
+ * [BUGFIX] Federation: Register federation metrics #7081
643
+ * [BUGFIX] PromQL: Fix panic in parser error handling #7132
644
+ * [BUGFIX] Rules: Fix reloads hanging when deleting a rule group that is being evaluated #7138
645
+ * [BUGFIX] TSDB: Fix a memory leak when prometheus starts with an empty TSDB WAL #7135
646
+ * [BUGFIX] TSDB: Make isolation more robust to panics in web handlers #7129 #7136
647
+
648
+ ## 2.17.1 / 2020-03-26
649
+
650
+ * [BUGFIX] TSDB: Fix query performance regression that increased memory and CPU usage #7051
651
+
652
+ ## 2.17.0 / 2020-03-24
653
+
654
+ This release implements isolation in TSDB. API queries and recording rules are
655
+ guaranteed to only see full scrapes and full recording rules. This comes with a
656
+ certain overhead in resource usage. Depending on the situation, there might be
657
+ some increase in memory usage, CPU usage, or query latency.
658
+
659
+ * [FEATURE] TSDB: Support isolation #6841
660
+ * [ENHANCEMENT] PromQL: Allow more keywords as metric names #6933
661
+ * [ENHANCEMENT] React UI: Add normalization of localhost URLs in targets page #6794
662
+ * [ENHANCEMENT] Remote read: Read from remote storage concurrently #6770
663
+ * [ENHANCEMENT] Rules: Mark deleted rule series as stale after a reload #6745
664
+ * [ENHANCEMENT] Scrape: Log scrape append failures as debug rather than warn #6852
665
+ * [ENHANCEMENT] TSDB: Improve query performance for queries that partially hit the head #6676
666
+ * [ENHANCEMENT] Consul SD: Expose service health as meta label #5313
667
+ * [ENHANCEMENT] EC2 SD: Expose EC2 instance lifecycle as meta label #6914
668
+ * [ENHANCEMENT] Kubernetes SD: Expose service type as meta label for K8s service role #6684
669
+ * [ENHANCEMENT] Kubernetes SD: Expose label_selector and field_selector #6807
670
+ * [ENHANCEMENT] Openstack SD: Expose hypervisor id as meta label #6962
671
+ * [BUGFIX] PromQL: Do not escape HTML-like chars in query log #6834 #6795
672
+ * [BUGFIX] React UI: Fix data table matrix values #6896
673
+ * [BUGFIX] React UI: Fix new targets page not loading when using non-ASCII characters #6892
674
+ * [BUGFIX] Remote read: Fix duplication of metrics read from remote storage with external labels #6967 #7018
675
+ * [BUGFIX] Remote write: Register WAL watcher and live reader metrics for all remotes, not just the first one #6998
676
+ * [BUGFIX] Scrape: Prevent removal of metric names upon relabeling #6891
677
+ * [BUGFIX] Scrape: Fix 'superfluous response.WriteHeader call' errors when scrape fails under some circonstances #6986
678
+ * [BUGFIX] Scrape: Fix crash when reloads are separated by two scrape intervals #7011
679
+
680
+ ## 2.16.0 / 2020-02-13
681
+
682
+ * [FEATURE] React UI: Support local timezone on /graph #6692
683
+ * [FEATURE] PromQL: add absent_over_time query function #6490
684
+ * [FEATURE] Adding optional logging of queries to their own file #6520
685
+ * [ENHANCEMENT] React UI: Add support for rules page and "Xs ago" duration displays #6503
686
+ * [ENHANCEMENT] React UI: alerts page, replace filtering togglers tabs with checkboxes #6543
687
+ * [ENHANCEMENT] TSDB: Export metric for WAL write errors #6647
688
+ * [ENHANCEMENT] TSDB: Improve query performance for queries that only touch the most recent 2h of data. #6651
689
+ * [ENHANCEMENT] PromQL: Refactoring in parser errors to improve error messages #6634
690
+ * [ENHANCEMENT] PromQL: Support trailing commas in grouping opts #6480
691
+ * [ENHANCEMENT] Scrape: Reduce memory usage on reloads by reusing scrape cache #6670
692
+ * [ENHANCEMENT] Scrape: Add metrics to track bytes and entries in the metadata cache #6675
693
+ * [ENHANCEMENT] promtool: Add support for line-column numbers for invalid rules output #6533
694
+ * [ENHANCEMENT] Avoid restarting rule groups when it is unnecessary #6450
695
+ * [BUGFIX] React UI: Send cookies on fetch() on older browsers #6553
696
+ * [BUGFIX] React UI: adopt grafana flot fix for stacked graphs #6603
697
+ * [BUFGIX] React UI: broken graph page browser history so that back button works as expected #6659
698
+ * [BUGFIX] TSDB: ensure compactionsSkipped metric is registered, and log proper error if one is returned from head.Init #6616
699
+ * [BUGFIX] TSDB: return an error on ingesting series with duplicate labels #6664
700
+ * [BUGFIX] PromQL: Fix unary operator precedence #6579
701
+ * [BUGFIX] PromQL: Respect query.timeout even when we reach query.max-concurrency #6712
702
+ * [BUGFIX] PromQL: Fix string and parentheses handling in engine, which affected React UI #6612
703
+ * [BUGFIX] PromQL: Remove output labels returned by absent() if they are produced by multiple identical label matchers #6493
704
+ * [BUGFIX] Scrape: Validate that OpenMetrics input ends with `# EOF` #6505
705
+ * [BUGFIX] Remote read: return the correct error if configs can't be marshal'd to JSON #6622
706
+ * [BUGFIX] Remote write: Make remote client `Store` use passed context, which can affect shutdown timing #6673
707
+ * [BUGFIX] Remote write: Improve sharding calculation in cases where we would always be consistently behind by tracking pendingSamples #6511
708
+ * [BUGFIX] Ensure prometheus_rule_group metrics are deleted when a rule group is removed #6693
709
+
710
+ ## 2.15.2 / 2020-01-06
711
+
712
+ * [BUGFIX] TSDB: Fixed support for TSDB blocks built with Prometheus before 2.1.0. #6564
713
+ * [BUGFIX] TSDB: Fixed block compaction issues on Windows. #6547
714
+
715
+ ## 2.15.1 / 2019-12-25
716
+
717
+ * [BUGFIX] TSDB: Fixed race on concurrent queries against same data. #6512
718
+
719
+ ## 2.15.0 / 2019-12-23
720
+
721
+ * [CHANGE] Discovery: Removed `prometheus_sd_kubernetes_cache_*` metrics. Additionally `prometheus_sd_kubernetes_workqueue_latency_seconds` and `prometheus_sd_kubernetes_workqueue_work_duration_seconds` metrics now show correct values in seconds. #6393
722
+ * [CHANGE] Remote write: Changed `query` label on `prometheus_remote_storage_*` metrics to `remote_name` and `url`. #6043
723
+ * [FEATURE] API: Added new endpoint for exposing per metric metadata `/metadata`. #6420 #6442
724
+ * [ENHANCEMENT] TSDB: Significantly reduced memory footprint of loaded TSDB blocks. #6418 #6461
725
+ * [ENHANCEMENT] TSDB: Significantly optimized what we buffer during compaction which should result in lower memory footprint during compaction. #6422 #6452 #6468 #6475
726
+ * [ENHANCEMENT] TSDB: Improve replay latency. #6230
727
+ * [ENHANCEMENT] TSDB: WAL size is now used for size based retention calculation. #5886
728
+ * [ENHANCEMENT] Remote read: Added query grouping and range hints to the remote read request #6401
729
+ * [ENHANCEMENT] Remote write: Added `prometheus_remote_storage_sent_bytes_total` counter per queue. #6344
730
+ * [ENHANCEMENT] promql: Improved PromQL parser performance. #6356
731
+ * [ENHANCEMENT] React UI: Implemented missing pages like `/targets` #6276, TSDB status page #6281 #6267 and many other fixes and performance improvements.
732
+ * [ENHANCEMENT] promql: Prometheus now accepts spaces between time range and square bracket. e.g `[ 5m]` #6065
733
+ * [BUGFIX] Config: Fixed alertmanager configuration to not miss targets when configurations are similar. #6455
734
+ * [BUGFIX] Remote write: Value of `prometheus_remote_storage_shards_desired` gauge shows raw value of desired shards and it's updated correctly. #6378
735
+ * [BUGFIX] Rules: Prometheus now fails the evaluation of rules and alerts where metric results collide with labels specified in `labels` field. #6469
736
+ * [BUGFIX] API: Targets Metadata API `/targets/metadata` now accepts empty `match_targets` parameter as in the spec. #6303
737
+
738
+ ## 2.14.0 / 2019-11-11
739
+
740
+ * [SECURITY/BUGFIX] UI: Ensure warnings from the API are escaped. #6279
741
+ * [FEATURE] API: `/api/v1/status/runtimeinfo` and `/api/v1/status/buildinfo` endpoints added for use by the React UI. #6243
742
+ * [FEATURE] React UI: implement the new experimental React based UI. #5694 and many more
743
+ * Can be found by under `/new`.
744
+ * Not all pages are implemented yet.
745
+ * [FEATURE] Status: Cardinality statistics added to the Runtime & Build Information page. #6125
746
+ * [ENHANCEMENT/BUGFIX] Remote write: fix delays in remote write after a compaction. #6021
747
+ * [ENHANCEMENT] UI: Alerts can be filtered by state. #5758
748
+ * [BUGFIX] API: lifecycle endpoints return 403 when not enabled. #6057
749
+ * [BUGFIX] Build: Fix Solaris build. #6149
750
+ * [BUGFIX] Promtool: Remove false duplicate rule warnings when checking rule files with alerts. #6270
751
+ * [BUGFIX] Remote write: restore use of deduplicating logger in remote write. #6113
752
+ * [BUGFIX] Remote write: do not reshard when unable to send samples. #6111
753
+ * [BUGFIX] Service discovery: errors are no longer logged on context cancellation. #6116, #6133
754
+ * [BUGFIX] UI: handle null response from API properly. #6071
755
+
756
+ ## 2.13.1 / 2019-10-16
757
+
758
+ * [BUGFIX] Fix panic in ARM builds of Prometheus. #6110
759
+ * [BUGFIX] promql: fix potential panic in the query logger. #6094
760
+ * [BUGFIX] Multiple errors of http: superfluous response.WriteHeader call in the logs. #6145
761
+
762
+ ## 2.13.0 / 2019-10-04
763
+
764
+ * [SECURITY/BUGFIX] UI: Fix a Stored DOM XSS vulnerability with query history [CVE-2019-10215](http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2019-10215). #6098
765
+ * [CHANGE] Metrics: renamed prometheus_sd_configs_failed_total to prometheus_sd_failed_configs and changed to Gauge #5254
766
+ * [ENHANCEMENT] Include the tsdb tool in builds. #6089
767
+ * [ENHANCEMENT] Service discovery: add new node address types for kubernetes. #5902
768
+ * [ENHANCEMENT] UI: show warnings if query have returned some warnings. #5964
769
+ * [ENHANCEMENT] Remote write: reduce memory usage of the series cache. #5849
770
+ * [ENHANCEMENT] Remote read: use remote read streaming to reduce memory usage. #5703
771
+ * [ENHANCEMENT] Metrics: added metrics for remote write max/min/desired shards to queue manager. #5787
772
+ * [ENHANCEMENT] Promtool: show the warnings during label query. #5924
773
+ * [ENHANCEMENT] Promtool: improve error messages when parsing bad rules. #5965
774
+ * [ENHANCEMENT] Promtool: more promlint rules. #5515
775
+ * [BUGFIX] Promtool: fix recording inconsistency due to duplicate labels. #6026
776
+ * [BUGFIX] UI: fixes service-discovery view when accessed from unhealthy targets. #5915
777
+ * [BUGFIX] Metrics format: OpenMetrics parser crashes on short input. #5939
778
+ * [BUGFIX] UI: avoid truncated Y-axis values. #6014
779
+
780
+ ## 2.12.0 / 2019-08-17
781
+
782
+ * [FEATURE] Track currently active PromQL queries in a log file. #5794
783
+ * [FEATURE] Enable and provide binaries for `mips64` / `mips64le` architectures. #5792
784
+ * [ENHANCEMENT] Improve responsiveness of targets web UI and API endpoint. #5740
785
+ * [ENHANCEMENT] Improve remote write desired shards calculation. #5763
786
+ * [ENHANCEMENT] Flush TSDB pages more precisely. tsdb#660
787
+ * [ENHANCEMENT] Add `prometheus_tsdb_retention_limit_bytes` metric. tsdb#667
788
+ * [ENHANCEMENT] Add logging during TSDB WAL replay on startup. tsdb#662
789
+ * [ENHANCEMENT] Improve TSDB memory usage. tsdb#653, tsdb#643, tsdb#654, tsdb#642, tsdb#627
790
+ * [BUGFIX] Check for duplicate label names in remote read. #5829
791
+ * [BUGFIX] Mark deleted rules' series as stale on next evaluation. #5759
792
+ * [BUGFIX] Fix JavaScript error when showing warning about out-of-sync server time. #5833
793
+ * [BUGFIX] Fix `promtool test rules` panic when providing empty `exp_labels`. #5774
794
+ * [BUGFIX] Only check last directory when discovering checkpoint number. #5756
795
+ * [BUGFIX] Fix error propagation in WAL watcher helper functions. #5741
796
+ * [BUGFIX] Correctly handle empty labels from alert templates. #5845
797
+
798
+ ## 2.11.2 / 2019-08-14
799
+
800
+ * [BUGFIX/SECURITY] Fix a Stored DOM XSS vulnerability with query history. #5888
801
+
802
+ ## 2.11.1 / 2019-07-10
803
+
804
+ * [BUGFIX] Fix potential panic when prometheus is watching multiple zookeeper paths. #5749
805
+
806
+ ## 2.11.0 / 2019-07-09
807
+
808
+ * [CHANGE] Remove `max_retries` from queue_config (it has been unused since rewriting remote-write to utilize the write-ahead-log). #5649
809
+ * [CHANGE] The meta file `BlockStats` no longer holds size information. This is now dynamically calculated and kept in memory. It also includes the meta file size which was not included before. tsdb#637
810
+ * [CHANGE] Renamed metric from `prometheus_tsdb_wal_reader_corruption_errors` to `prometheus_tsdb_wal_reader_corruption_errors_total`. tsdb#622
811
+ * [FEATURE] Add option to use Alertmanager API v2. #5482
812
+ * [FEATURE] Added `humanizePercentage` function for templates. #5670
813
+ * [FEATURE] Include InitContainers in Kubernetes Service Discovery. #5598
814
+ * [FEATURE] Provide option to compress WAL records using Snappy. [#609](https://github.com/prometheus/tsdb/pull/609)
815
+ * [ENHANCEMENT] Create new clean segment when starting the WAL. tsdb#608
816
+ * [ENHANCEMENT] Reduce allocations in PromQL aggregations. #5641
817
+ * [ENHANCEMENT] Add storage warnings to LabelValues and LabelNames API results. #5673
818
+ * [ENHANCEMENT] Add `prometheus_http_requests_total` metric. #5640
819
+ * [ENHANCEMENT] Enable openbsd/arm build. #5696
820
+ * [ENHANCEMENT] Remote-write allocation improvements. #5614
821
+ * [ENHANCEMENT] Query performance improvement: Efficient iteration and search in HashForLabels and HashWithoutLabels. #5707
822
+ * [ENHANCEMENT] Allow injection of arbitrary headers in promtool. #4389
823
+ * [ENHANCEMENT] Allow passing `external_labels` in alert unit tests groups. #5608
824
+ * [ENHANCEMENT] Allows globs for rules when unit testing. #5595
825
+ * [ENHANCEMENT] Improved postings intersection matching. tsdb#616
826
+ * [ENHANCEMENT] Reduced disk usage for WAL for small setups. tsdb#605
827
+ * [ENHANCEMENT] Optimize queries using regexp for set lookups. tsdb#602
828
+ * [BUGFIX] resolve race condition in maxGauge. #5647
829
+ * [BUGFIX] Fix ZooKeeper connection leak. #5675
830
+ * [BUGFIX] Improved atomicity of .tmp block replacement during compaction for usual case. tsdb#636
831
+ * [BUGFIX] Fix "unknown series references" after clean shutdown. tsdb#623
832
+ * [BUGFIX] Re-calculate block size when calling `block.Delete`. tsdb#637
833
+ * [BUGFIX] Fix unsafe snapshots with head block. tsdb#641
834
+ * [BUGFIX] `prometheus_tsdb_compactions_failed_total` is now incremented on any compaction failure. tsdb#613
835
+
836
+ ## 2.10.0 / 2019-05-25
837
+
838
+ * [CHANGE/BUGFIX] API: Encode alert values as string to correctly represent Inf/NaN. #5582
839
+ * [FEATURE] Template expansion: Make external labels available as `$externalLabels` in alert and console template expansion. #5463
840
+ * [FEATURE] TSDB: Add `prometheus_tsdb_wal_segment_current` metric for the WAL segment index that TSDB is currently writing to. tsdb#601
841
+ * [FEATURE] Scrape: Add `scrape_series_added` per-scrape metric. #5546
842
+ * [ENHANCEMENT] Discovery/kubernetes: Add labels `__meta_kubernetes_endpoint_node_name` and `__meta_kubernetes_endpoint_hostname`. #5571
843
+ * [ENHANCEMENT] Discovery/azure: Add label `__meta_azure_machine_public_ip`. #5475
844
+ * [ENHANCEMENT] TSDB: Simplify mergedPostings.Seek, resulting in better performance if there are many posting lists. tsdb#595
845
+ * [ENHANCEMENT] Log filesystem type on startup. #5558
846
+ * [ENHANCEMENT] Cmd/promtool: Use POST requests for Query and QueryRange. client_golang#557
847
+ * [ENHANCEMENT] Web: Sort alerts by group name. #5448
848
+ * [ENHANCEMENT] Console templates: Add convenience variables `$rawParams`, `$params`, `$path`. #5463
849
+ * [BUGFIX] TSDB: Don't panic when running out of disk space and recover nicely from the condition. tsdb#582
850
+ * [BUGFIX] TSDB: Correctly handle empty labels. tsdb#594
851
+ * [BUGFIX] TSDB: Don't crash on an unknown tombstone reference. tsdb#604
852
+ * [BUGFIX] Storage/remote: Remove queue-manager specific metrics if queue no longer exists. #5445 #5485 #5555
853
+ * [BUGFIX] PromQL: Correctly display `{__name__="a"}`. #5552
854
+ * [BUGFIX] Discovery/kubernetes: Use `service` rather than `ingress` as the name for the service workqueue. #5520
855
+ * [BUGFIX] Discovery/azure: Don't panic on a VM with a public IP. #5587
856
+ * [BUGFIX] Discovery/triton: Always read HTTP body to completion. #5596
857
+ * [BUGFIX] Web: Fixed Content-Type for js and css instead of using `/etc/mime.types`. #5551
858
+
859
+ ## 2.9.2 / 2019-04-24
860
+
861
+ * [BUGFIX] Make sure subquery range is taken into account for selection #5467
862
+ * [BUGFIX] Exhaust every request body before closing it #5166
863
+ * [BUGFIX] Cmd/promtool: return errors from rule evaluations #5483
864
+ * [BUGFIX] Remote Storage: string interner should not panic in release #5487
865
+ * [BUGFIX] Fix memory allocation regression in mergedPostings.Seek tsdb#586
866
+
867
+ ## 2.9.1 / 2019-04-16
868
+
869
+ * [BUGFIX] Discovery/kubernetes: fix missing label sanitization #5462
870
+ * [BUGFIX] Remote_write: Prevent reshard concurrent with calling stop #5460
871
+
872
+ ## 2.9.0 / 2019-04-15
873
+
874
+ This releases uses Go 1.12, which includes a change in how memory is released
875
+ to Linux. This will cause RSS to be reported as higher, however this is harmless
876
+ and the memory is available to the kernel when it needs it.
877
+
878
+ * [CHANGE/ENHANCEMENT] Update Consul to support catalog.ServiceMultipleTags. #5151
879
+ * [FEATURE] Add honor_timestamps scrape option. #5304
880
+ * [ENHANCEMENT] Discovery/kubernetes: add present labels for labels/annotations. #5443
881
+ * [ENHANCEMENT] OpenStack SD: Add ProjectID and UserID meta labels. #5431
882
+ * [ENHANCEMENT] Add GODEBUG and retention to the runtime page. #5324 #5322
883
+ * [ENHANCEMENT] Add support for POSTing to /series endpoint. #5422
884
+ * [ENHANCEMENT] Support PUT methods for Lifecycle and Admin APIs. #5376
885
+ * [ENHANCEMENT] Scrape: Add global jitter for HA server. #5181
886
+ * [ENHANCEMENT] Check for cancellation on every step of a range evaluation. #5131
887
+ * [ENHANCEMENT] String interning for labels & values in the remote_write path. #5316
888
+ * [ENHANCEMENT] Don't lose the scrape cache on a failed scrape. #5414
889
+ * [ENHANCEMENT] Reload cert files from disk automatically. common#173
890
+ * [ENHANCEMENT] Use fixed length millisecond timestamp format for logs. common#172
891
+ * [ENHANCEMENT] Performance improvements for postings. tsdb#509 tsdb#572
892
+ * [BUGFIX] Remote Write: fix checkpoint reading. #5429
893
+ * [BUGFIX] Check if label value is valid when unmarshaling external labels from YAML. #5316
894
+ * [BUGFIX] Promparse: sort all labels when parsing. #5372
895
+ * [BUGFIX] Reload rules: copy state on both name and labels. #5368
896
+ * [BUGFIX] Exponentiation operator to drop metric name in result of operation. #5329
897
+ * [BUGFIX] Config: resolve more file paths. #5284
898
+ * [BUGFIX] Promtool: resolve relative paths in alert test files. #5336
899
+ * [BUGFIX] Set TLSHandshakeTimeout in HTTP transport. common#179
900
+ * [BUGFIX] Use fsync to be more resilient to machine crashes. tsdb#573 tsdb#578
901
+ * [BUGFIX] Keep series that are still in WAL in checkpoints. tsdb#577
902
+ * [BUGFIX] Fix output sample values for scalar-to-vector comparison operations. #5454
903
+
904
+ ## 2.8.1 / 2019-03-28
905
+
906
+ * [BUGFIX] Display the job labels in `/targets` which was removed accidentally. #5406
907
+
908
+ ## 2.8.0 / 2019-03-12
909
+
910
+ This release uses Write-Ahead Logging (WAL) for the remote_write API. This currently causes a slight increase in memory usage, which will be addressed in future releases.
911
+
912
+ * [CHANGE] Default time retention is used only when no size based retention is specified. These are flags where time retention is specified by the flag `--storage.tsdb.retention` and size retention by `--storage.tsdb.retention.size`. #5216
913
+ * [CHANGE] `prometheus_tsdb_storage_blocks_bytes_total` is now `prometheus_tsdb_storage_blocks_bytes`. prometheus/tsdb#506
914
+ * [FEATURE] [EXPERIMENTAL] Time overlapping blocks are now allowed; vertical compaction and vertical query merge. It is an optional feature which is controlled by the `--storage.tsdb.allow-overlapping-blocks` flag, disabled by default. prometheus/tsdb#370
915
+ * [ENHANCEMENT] Use the WAL for remote_write API. #4588
916
+ * [ENHANCEMENT] Query performance improvements. prometheus/tsdb#531
917
+ * [ENHANCEMENT] UI enhancements with upgrade to Bootstrap 4. #5226
918
+ * [ENHANCEMENT] Reduce time that Alertmanagers are in flux when reloaded. #5126
919
+ * [ENHANCEMENT] Limit number of metrics displayed on UI to 10000. #5139
920
+ * [ENHANCEMENT] (1) Remember All/Unhealthy choice on target-overview when reloading page. (2) Resize text-input area on Graph page on mouseclick. #5201
921
+ * [ENHANCEMENT] In `histogram_quantile` merge buckets with equivalent le values. #5158.
922
+ * [ENHANCEMENT] Show list of offending labels in the error message in many-to-many scenarios. #5189
923
+ * [ENHANCEMENT] Show `Storage Retention` criteria in effect on `/status` page. #5322
924
+ * [BUGFIX] Fix sorting of rule groups. #5260
925
+ * [BUGFIX] Fix support for password_file and bearer_token_file in Kubernetes SD. #5211
926
+ * [BUGFIX] Scrape: catch errors when creating HTTP clients #5182. Adds new metrics:
927
+ * `prometheus_target_scrape_pools_total`
928
+ * `prometheus_target_scrape_pools_failed_total`
929
+ * `prometheus_target_scrape_pool_reloads_total`
930
+ * `prometheus_target_scrape_pool_reloads_failed_total`
931
+ * [BUGFIX] Fix panic when aggregator param is not a literal. #5290
932
+
933
+ ## 2.7.2 / 2019-03-02
934
+
935
+ * [BUGFIX] `prometheus_rule_group_last_evaluation_timestamp_seconds` is now a unix timestamp. #5186
936
+
937
+ ## 2.7.1 / 2019-01-31
938
+
939
+ This release has a fix for a Stored DOM XSS vulnerability that can be triggered when using the query history functionality. Thanks to Dor Tumarkin from Checkmarx for reporting it.
940
+
941
+ * [BUGFIX/SECURITY] Fix a Stored DOM XSS vulnerability with query history. #5163
942
+ * [BUGFIX] `prometheus_rule_group_last_duration_seconds` now reports seconds instead of nanoseconds. #5153
943
+ * [BUGFIX] Make sure the targets are consistently sorted in the targets page. #5161
944
+
945
+ ## 2.7.0 / 2019-01-28
946
+
947
+ We're rolling back the Dockerfile changes introduced in 2.6.0. If you made changes to your docker deployment in 2.6.0, you will need to roll them back. This release also adds experimental support for disk size based retention. To accommodate that we are deprecating the flag `storage.tsdb.retention` in favour of `storage.tsdb.retention.time`. We print a warning if the flag is in use, but it will function without breaking until Prometheus 3.0.
948
+
949
+ * [CHANGE] Rollback Dockerfile to version at 2.5.0. Rollback of the breaking change introduced in 2.6.0. #5122
950
+ * [FEATURE] Add subqueries to PromQL. #4831
951
+ * [FEATURE] [EXPERIMENTAL] Add support for disk size based retention. Note that we don't consider the WAL size which could be significant and the time based retention policy also applies. #5109 prometheus/tsdb#343
952
+ * [FEATURE] Add CORS origin flag. #5011
953
+ * [ENHANCEMENT] Consul SD: Add tagged address to the discovery metadata. #5001
954
+ * [ENHANCEMENT] Kubernetes SD: Add service external IP and external name to the discovery metadata. #4940
955
+ * [ENHANCEMENT] Azure SD: Add support for Managed Identity authentication. #4590
956
+ * [ENHANCEMENT] Azure SD: Add tenant and subscription IDs to the discovery metadata. #4969
957
+ * [ENHANCEMENT] OpenStack SD: Add support for application credentials based authentication. #4968
958
+ * [ENHANCEMENT] Add metric for number of rule groups loaded. #5090
959
+ * [BUGFIX] Avoid duplicate tests for alert unit tests. #4964
960
+ * [BUGFIX] Don't depend on given order when comparing samples in alert unit testing. #5049
961
+ * [BUGFIX] Make sure the retention period doesn't overflow. #5112
962
+ * [BUGFIX] Make sure the blocks don't get very large. #5112
963
+ * [BUGFIX] Don't generate blocks with no samples. prometheus/tsdb#374
964
+ * [BUGFIX] Reintroduce metric for WAL corruptions. prometheus/tsdb#473
965
+
966
+ ## 2.6.1 / 2019-01-15
967
+
968
+ * [BUGFIX] Azure SD: Fix discovery getting stuck sometimes. #5088
969
+ * [BUGFIX] Marathon SD: Use `Tasks.Ports` when `RequirePorts` is `false`. #5026
970
+ * [BUGFIX] Promtool: Fix "out-of-order sample" errors when testing rules. #5069
971
+
972
+ ## 2.6.0 / 2018-12-17
973
+
974
+ * [CHANGE] Remove default flags from the container's entrypoint, run Prometheus from `/etc/prometheus` and symlink the storage directory to `/etc/prometheus/data`. #4976
975
+ * [CHANGE] Promtool: Remove the `update` command. #3839
976
+ * [FEATURE] Add JSON log format via the `--log.format` flag. #4876
977
+ * [FEATURE] API: Add /api/v1/labels endpoint to get all label names. #4835
978
+ * [FEATURE] Web: Allow setting the page's title via the `--web.ui-title` flag. #4841
979
+ * [ENHANCEMENT] Add `prometheus_tsdb_lowest_timestamp_seconds`, `prometheus_tsdb_head_min_time_seconds` and `prometheus_tsdb_head_max_time_seconds` metrics. #4888
980
+ * [ENHANCEMENT] Add `rule_group_last_evaluation_timestamp_seconds` metric. #4852
981
+ * [ENHANCEMENT] Add `prometheus_template_text_expansion_failures_total` and `prometheus_template_text_expansions_total` metrics. #4747
982
+ * [ENHANCEMENT] Set consistent User-Agent header in outgoing requests. #4891
983
+ * [ENHANCEMENT] Azure SD: Error out at load time when authentication parameters are missing. #4907
984
+ * [ENHANCEMENT] EC2 SD: Add the machine's private DNS name to the discovery metadata. #4693
985
+ * [ENHANCEMENT] EC2 SD: Add the operating system's platform to the discovery metadata. #4663
986
+ * [ENHANCEMENT] Kubernetes SD: Add the pod's phase to the discovery metadata. #4824
987
+ * [ENHANCEMENT] Kubernetes SD: Log Kubernetes messages. #4931
988
+ * [ENHANCEMENT] Promtool: Collect CPU and trace profiles. #4897
989
+ * [ENHANCEMENT] Promtool: Support writing output as JSON. #4848
990
+ * [ENHANCEMENT] Remote Read: Return available data if remote read fails partially. #4832
991
+ * [ENHANCEMENT] Remote Write: Improve queue performance. #4772
992
+ * [ENHANCEMENT] Remote Write: Add min_shards parameter to set the minimum number of shards. #4924
993
+ * [ENHANCEMENT] TSDB: Improve WAL reading. #4953
994
+ * [ENHANCEMENT] TSDB: Memory improvements. #4953
995
+ * [ENHANCEMENT] Web: Log stack traces on panic. #4221
996
+ * [ENHANCEMENT] Web UI: Add copy to clipboard button for configuration. #4410
997
+ * [ENHANCEMENT] Web UI: Support console queries at specific times. #4764
998
+ * [ENHANCEMENT] Web UI: group targets by job then instance. #4898 #4806
999
+ * [BUGFIX] Deduplicate handler labels for HTTP metrics. #4732
1000
+ * [BUGFIX] Fix leaked queriers causing shutdowns to hang. #4922
1001
+ * [BUGFIX] Fix configuration loading panics on nil pointer slice elements. #4942
1002
+ * [BUGFIX] API: Correctly skip mismatching targets on /api/v1/targets/metadata. #4905
1003
+ * [BUGFIX] API: Better rounding for incoming query timestamps. #4941
1004
+ * [BUGFIX] Azure SD: Fix panic. #4867
1005
+ * [BUGFIX] Console templates: Fix hover when the metric has a null value. #4906
1006
+ * [BUGFIX] Discovery: Remove all targets when the scrape configuration gets empty. #4819
1007
+ * [BUGFIX] Marathon SD: Fix leaked connections. #4915
1008
+ * [BUGFIX] Marathon SD: Use 'hostPort' member of portMapping to construct target endpoints. #4887
1009
+ * [BUGFIX] PromQL: Fix a goroutine leak in the lexer/parser. #4858
1010
+ * [BUGFIX] Scrape: Pass through content-type for non-compressed output. #4912
1011
+ * [BUGFIX] Scrape: Fix deadlock in the scrape's manager. #4894
1012
+ * [BUGFIX] Scrape: Scrape targets at fixed intervals even after Prometheus restarts. #4926
1013
+ * [BUGFIX] TSDB: Support restored snapshots including the head properly. #4953
1014
+ * [BUGFIX] TSDB: Repair WAL when the last record in a segment is torn. #4953
1015
+ * [BUGFIX] TSDB: Fix unclosed file readers on Windows systems. #4997
1016
+ * [BUGFIX] Web: Avoid proxy to connect to the local gRPC server. #4572
1017
+
1018
+ ## 2.5.0 / 2018-11-06
1019
+
1020
+ * [CHANGE] Group targets by scrape config instead of job name. #4806 #4526
1021
+ * [CHANGE] Marathon SD: Various changes to adapt to Marathon 1.5+. #4499
1022
+ * [CHANGE] Discovery: Split `prometheus_sd_discovered_targets` metric by scrape and notify (Alertmanager SD) as well as by section in the respective configuration. #4753
1023
+ * [FEATURE] Add OpenMetrics support for scraping (EXPERIMENTAL). #4700
1024
+ * [FEATURE] Add unit testing for rules. #4350
1025
+ * [FEATURE] Make maximum number of samples per query configurable via `--query.max-samples` flag. #4513
1026
+ * [FEATURE] Make maximum number of concurrent remote reads configurable via `--storage.remote.read-concurrent-limit` flag. #4656
1027
+ * [ENHANCEMENT] Support s390x platform for Linux. #4605
1028
+ * [ENHANCEMENT] API: Add `prometheus_api_remote_read_queries` metric tracking currently executed or waiting remote read API requests. #4699
1029
+ * [ENHANCEMENT] Remote Read: Add `prometheus_remote_storage_remote_read_queries` metric tracking currently in-flight remote read queries. #4677
1030
+ * [ENHANCEMENT] Remote Read: Reduced memory usage. #4655
1031
+ * [ENHANCEMENT] Discovery: Add `prometheus_sd_discovered_targets`, `prometheus_sd_received_updates_total`, `prometheus_sd_updates_delayed_total`, and `prometheus_sd_updates_total` metrics for discovery subsystem. #4667
1032
+ * [ENHANCEMENT] Discovery: Improve performance of previously slow updates of changes of targets. #4526
1033
+ * [ENHANCEMENT] Kubernetes SD: Add extended metrics. #4458
1034
+ * [ENHANCEMENT] OpenStack SD: Support discovering instances from all projects. #4682
1035
+ * [ENHANCEMENT] OpenStack SD: Discover all interfaces. #4649
1036
+ * [ENHANCEMENT] OpenStack SD: Support `tls_config` for the used HTTP client. #4654
1037
+ * [ENHANCEMENT] Triton SD: Add ability to filter triton_sd targets by pre-defined groups. #4701
1038
+ * [ENHANCEMENT] Web UI: Avoid browser spell-checking in expression field. #4728
1039
+ * [ENHANCEMENT] Web UI: Add scrape duration and last evaluation time in targets and rules pages. #4722
1040
+ * [ENHANCEMENT] Web UI: Improve rule view by wrapping lines. #4702
1041
+ * [ENHANCEMENT] Rules: Error out at load time for invalid templates, rather than at evaluation time. #4537
1042
+ * [ENHANCEMENT] TSDB: Add metrics for WAL operations. #4692
1043
+ * [BUGFIX] Change max/min over_time to handle NaNs properly. #4386
1044
+ * [BUGFIX] Check label name for `count_values` PromQL function. #4585
1045
+ * [BUGFIX] Ensure that vectors and matrices do not contain identical label-sets. #4589
1046
+
1047
+ ## 2.4.3 / 2018-10-04
1048
+
1049
+ * [BUGFIX] Fix panic when using custom EC2 API for SD #4672
1050
+ * [BUGFIX] Fix panic when Zookeeper SD cannot connect to servers #4669
1051
+ * [BUGFIX] Make the skip_head an optional parameter for snapshot API #4674
1052
+
1053
+ ## 2.4.2 / 2018-09-21
1054
+
1055
+ The last release didn't have bugfix included due to a vendoring error.
1056
+
1057
+ * [BUGFIX] Handle WAL corruptions properly prometheus/tsdb#389
1058
+ * [BUGFIX] Handle WAL migrations correctly on Windows prometheus/tsdb#392
1059
+
1060
+ ## 2.4.1 / 2018-09-19
1061
+
1062
+ * [ENHANCEMENT] New TSDB metrics prometheus/tsdb#375 prometheus/tsdb#363
1063
+ * [BUGFIX] Render UI correctly for Windows #4616
1064
+
1065
+ ## 2.4.0 / 2018-09-11
1066
+
1067
+ This release includes multiple bugfixes and features. Further, the WAL implementation has been re-written so the storage is not forward compatible. Prometheus 2.3 storage will work on 2.4 but not vice-versa.
1068
+
1069
+ * [CHANGE] Reduce remote write default retries #4279
1070
+ * [CHANGE] Remove /heap endpoint #4460
1071
+ * [FEATURE] Persist alert 'for' state across restarts #4061
1072
+ * [FEATURE] Add API providing per target metric metadata #4183
1073
+ * [FEATURE] Add API providing recording and alerting rules #4318 #4501
1074
+ * [ENHANCEMENT] Brand new WAL implementation for TSDB. Forwards incompatible with previous WAL.
1075
+ * [ENHANCEMENT] Show rule evaluation errors in UI #4457
1076
+ * [ENHANCEMENT] Throttle resends of alerts to Alertmanager #4538
1077
+ * [ENHANCEMENT] Send EndsAt along with the alert to Alertmanager #4550
1078
+ * [ENHANCEMENT] Limit the samples returned by remote read endpoint #4532
1079
+ * [ENHANCEMENT] Limit the data read in through remote read #4239
1080
+ * [ENHANCEMENT] Coalesce identical SD configurations #3912
1081
+ * [ENHANCEMENT] `promtool`: Add new commands for debugging and querying #4247 #4308 #4346 #4454
1082
+ * [ENHANCEMENT] Update console examples for node_exporter v0.16.0 #4208
1083
+ * [ENHANCEMENT] Optimize PromQL aggregations #4248
1084
+ * [ENHANCEMENT] Remote read: Add Offset to hints #4226
1085
+ * [ENHANCEMENT] `consul_sd`: Add support for ServiceMeta field #4280
1086
+ * [ENHANCEMENT] `ec2_sd`: Maintain order of subnet_id label #4405
1087
+ * [ENHANCEMENT] `ec2_sd`: Add support for custom endpoint to support EC2 compliant APIs #4333
1088
+ * [ENHANCEMENT] `ec2_sd`: Add instance_owner label #4514
1089
+ * [ENHANCEMENT] `azure_sd`: Add support for VMSS discovery and multiple environments #4202 #4569
1090
+ * [ENHANCEMENT] `gce_sd`: Add instance_id label #4488
1091
+ * [ENHANCEMENT] Forbid rule-abiding robots from indexing #4266
1092
+ * [ENHANCEMENT] Log virtual memory limits on startup #4418
1093
+ * [BUGFIX] Wait for service discovery to stop before exiting #4508
1094
+ * [BUGFIX] Render SD configs properly #4338
1095
+ * [BUGFIX] Only add LookbackDelta to vector selectors #4399
1096
+ * [BUGFIX] `ec2_sd`: Handle panic-ing nil pointer #4469
1097
+ * [BUGFIX] `consul_sd`: Stop leaking connections #4443
1098
+ * [BUGFIX] Use templated labels also to identify alerts #4500
1099
+ * [BUGFIX] Reduce floating point errors in stddev and related functions #4533
1100
+ * [BUGFIX] Log errors while encoding responses #4359
1101
+
1102
+ ## 2.3.2 / 2018-07-12
1103
+
1104
+ * [BUGFIX] Fix various tsdb bugs #4369
1105
+ * [BUGFIX] Reorder startup and shutdown to prevent panics. #4321
1106
+ * [BUGFIX] Exit with non-zero code on error #4296
1107
+ * [BUGFIX] discovery/kubernetes/ingress: fix scheme discovery #4329
1108
+ * [BUGFIX] Fix race in zookeeper sd #4355
1109
+ * [BUGFIX] Better timeout handling in promql #4291 #4300
1110
+ * [BUGFIX] Propagate errors when selecting series from the tsdb #4136
1111
+
1112
+ ## 2.3.1 / 2018-06-19
1113
+
1114
+ * [BUGFIX] Avoid infinite loop on duplicate NaN values. #4275
1115
+ * [BUGFIX] Fix nil pointer deference when using various API endpoints #4282
1116
+ * [BUGFIX] config: set target group source index during unmarshaling #4245
1117
+ * [BUGFIX] discovery/file: fix logging #4178
1118
+ * [BUGFIX] kubernetes_sd: fix namespace filtering #4285
1119
+ * [BUGFIX] web: restore old path prefix behavior #4273
1120
+ * [BUGFIX] web: remove security headers added in 2.3.0 #4259
1121
+
1122
+ ## 2.3.0 / 2018-06-05
1123
+
1124
+ * [CHANGE] `marathon_sd`: use `auth_token` and `auth_token_file` for token-based authentication instead of `bearer_token` and `bearer_token_file` respectively.
1125
+ * [CHANGE] Metric names for HTTP server metrics changed
1126
+ * [FEATURE] Add query commands to promtool
1127
+ * [FEATURE] Add security headers to HTTP server responses
1128
+ * [FEATURE] Pass query hints via remote read API
1129
+ * [FEATURE] Basic auth passwords can now be configured via file across all configuration
1130
+ * [ENHANCEMENT] Optimize PromQL and API serialization for memory usage and allocations
1131
+ * [ENHANCEMENT] Limit number of dropped targets in web UI
1132
+ * [ENHANCEMENT] Consul and EC2 service discovery allow using server-side filtering for performance improvement
1133
+ * [ENHANCEMENT] Add advanced filtering configuration to EC2 service discovery
1134
+ * [ENHANCEMENT] `marathon_sd`: adds support for basic and bearer authentication, plus all other common HTTP client options (TLS config, proxy URL, etc.)
1135
+ * [ENHANCEMENT] Provide machine type metadata and labels in GCE service discovery
1136
+ * [ENHANCEMENT] Add pod controller kind and name to Kubernetes service discovery data
1137
+ * [ENHANCEMENT] Move TSDB to flock-based log file that works with Docker containers
1138
+ * [BUGFIX] Properly propagate storage errors in PromQL
1139
+ * [BUGFIX] Fix path prefix for web pages
1140
+ * [BUGFIX] Fix goroutine leak in Consul service discovery
1141
+ * [BUGFIX] Fix races in scrape manager
1142
+ * [BUGFIX] Fix OOM for very large k in PromQL topk() queries
1143
+ * [BUGFIX] Make remote write more resilient to unavailable receivers
1144
+ * [BUGFIX] Make remote write shutdown cleanly
1145
+ * [BUGFIX] Don't leak files on errors in TSDB's tombstone cleanup
1146
+ * [BUGFIX] Unary minus expressions now removes the metric name from results
1147
+ * [BUGFIX] Fix bug that lead to wrong amount of samples considered for time range expressions
1148
+
1149
+ ## 2.2.1 / 2018-03-13
1150
+
1151
+ * [BUGFIX] Fix data loss in TSDB on compaction
1152
+ * [BUGFIX] Correctly stop timer in remote-write path
1153
+ * [BUGFIX] Fix deadlock triggered by loading targets page
1154
+ * [BUGFIX] Fix incorrect buffering of samples on range selection queries
1155
+ * [BUGFIX] Handle large index files on windows properly
1156
+
1157
+ ## 2.2.0 / 2018-03-08
1158
+
1159
+ * [CHANGE] Rename file SD mtime metric.
1160
+ * [CHANGE] Send target update on empty pod IP in Kubernetes SD.
1161
+ * [FEATURE] Add API endpoint for flags.
1162
+ * [FEATURE] Add API endpoint for dropped targets.
1163
+ * [FEATURE] Display annotations on alerts page.
1164
+ * [FEATURE] Add option to skip head data when taking snapshots.
1165
+ * [ENHANCEMENT] Federation performance improvement.
1166
+ * [ENHANCEMENT] Read bearer token file on every scrape.
1167
+ * [ENHANCEMENT] Improve typeahead on `/graph` page.
1168
+ * [ENHANCEMENT] Change rule file formatting.
1169
+ * [ENHANCEMENT] Set consul server default to `localhost:8500`.
1170
+ * [ENHANCEMENT] Add dropped Alertmanagers to API info endpoint.
1171
+ * [ENHANCEMENT] Add OS type meta label to Azure SD.
1172
+ * [ENHANCEMENT] Validate required fields in SD configuration.
1173
+ * [BUGFIX] Prevent stack overflow on deep recursion in TSDB.
1174
+ * [BUGFIX] Correctly read offsets in index files that are greater than 4GB.
1175
+ * [BUGFIX] Fix scraping behavior for empty labels.
1176
+ * [BUGFIX] Drop metric name for bool modifier.
1177
+ * [BUGFIX] Fix races in discovery.
1178
+ * [BUGFIX] Fix Kubernetes endpoints SD for empty subsets.
1179
+ * [BUGFIX] Throttle updates from SD providers, which caused increased CPU usage and allocations.
1180
+ * [BUGFIX] Fix TSDB block reload issue.
1181
+ * [BUGFIX] Fix PromQL printing of empty `without()`.
1182
+ * [BUGFIX] Don't reset FiredAt for inactive alerts.
1183
+ * [BUGFIX] Fix erroneous file version changes and repair existing data.
1184
+
1185
+ ## 2.1.0 / 2018-01-19
1186
+
1187
+ * [FEATURE] New Service Discovery UI showing labels before and after relabelling.
1188
+ * [FEATURE] New Admin APIs added to v1 to delete, snapshot and remove tombstones.
1189
+ * [ENHANCEMENT] The graph UI autocomplete now includes your previous queries.
1190
+ * [ENHANCEMENT] Federation is now much faster for large numbers of series.
1191
+ * [ENHANCEMENT] Added new metrics to measure rule timings.
1192
+ * [ENHANCEMENT] Rule evaluation times added to the rules UI.
1193
+ * [ENHANCEMENT] Added metrics to measure modified time of file SD files.
1194
+ * [ENHANCEMENT] Kubernetes SD now includes POD UID in discovery metadata.
1195
+ * [ENHANCEMENT] The Query APIs now return optional stats on query execution times.
1196
+ * [ENHANCEMENT] The index now no longer has the 4GiB size limit and is also smaller.
1197
+ * [BUGFIX] Remote read `read_recent` option is now false by default.
1198
+ * [BUGFIX] Pass the right configuration to each Alertmanager (AM) when using multiple AM configs.
1199
+ * [BUGFIX] Fix not-matchers not selecting series with labels unset.
1200
+ * [BUGFIX] tsdb: Fix occasional panic in head block.
1201
+ * [BUGFIX] tsdb: Close files before deletion to fix retention issues on Windows and NFS.
1202
+ * [BUGFIX] tsdb: Cleanup and do not retry failing compactions.
1203
+ * [BUGFIX] tsdb: Close WAL while shutting down.
1204
+
1205
+ ## 2.0.0 / 2017-11-08
1206
+
1207
+ This release includes a completely rewritten storage, huge performance
1208
+ improvements, but also many backwards incompatible changes. For more
1209
+ information, read the announcement blog post and migration guide.
1210
+
1211
+ <https://prometheus.io/blog/2017/11/08/announcing-prometheus-2-0/>
1212
+ <https://prometheus.io/docs/prometheus/2.0/migration/>
1213
+
1214
+ * [CHANGE] Completely rewritten storage layer, with WAL. This is not backwards compatible with 1.x storage, and many flags have changed/disappeared.
1215
+ * [CHANGE] New staleness behavior. Series now marked stale after target scrapes no longer return them, and soon after targets disappear from service discovery.
1216
+ * [CHANGE] Rules files use YAML syntax now. Conversion tool added to promtool.
1217
+ * [CHANGE] Removed `count_scalar`, `drop_common_labels` functions and `keep_common` modifier from PromQL.
1218
+ * [CHANGE] Rewritten exposition format parser with much higher performance. The Protobuf exposition format is no longer supported.
1219
+ * [CHANGE] Example console templates updated for new storage and metrics names. Examples other than node exporter and Prometheus removed.
1220
+ * [CHANGE] Admin and lifecycle APIs now disabled by default, can be re-enabled via flags
1221
+ * [CHANGE] Flags switched to using Kingpin, all flags are now --flagname rather than -flagname.
1222
+ * [FEATURE/CHANGE] Remote read can be configured to not read data which is available locally. This is enabled by default.
1223
+ * [FEATURE] Rules can be grouped now. Rules within a rule group are executed sequentially.
1224
+ * [FEATURE] Added experimental GRPC apis
1225
+ * [FEATURE] Add timestamp() function to PromQL.
1226
+ * [ENHANCEMENT] Remove remote read from the query path if no remote storage is configured.
1227
+ * [ENHANCEMENT] Bump Consul HTTP client timeout to not match the Consul SD watch timeout.
1228
+ * [ENHANCEMENT] Go-conntrack added to provide HTTP connection metrics.
1229
+ * [BUGFIX] Fix connection leak in Consul SD.
1230
+
1231
+ ## 1.8.2 / 2017-11-04
1232
+
1233
+ * [BUGFIX] EC2 service discovery: Do not crash if tags are empty.
1234
+
1235
+ ## 1.8.1 / 2017-10-19
1236
+
1237
+ * [BUGFIX] Correctly handle external labels on remote read endpoint
1238
+
1239
+ ## 1.8.0 / 2017-10-06
1240
+
1241
+ * [CHANGE] Rule links link to the _Console_ tab rather than the _Graph_ tab to
1242
+ not trigger expensive range queries by default.
1243
+ * [FEATURE] Ability to act as a remote read endpoint for other Prometheus
1244
+ servers.
1245
+ * [FEATURE] K8s SD: Support discovery of ingresses.
1246
+ * [FEATURE] Consul SD: Support for node metadata.
1247
+ * [FEATURE] Openstack SD: Support discovery of hypervisors.
1248
+ * [FEATURE] Expose current Prometheus config via `/status/config`.
1249
+ * [FEATURE] Allow to collapse jobs on `/targets` page.
1250
+ * [FEATURE] Add `/-/healthy` and `/-/ready` endpoints.
1251
+ * [FEATURE] Add color scheme support to console templates.
1252
+ * [ENHANCEMENT] Remote storage connections use HTTP keep-alive.
1253
+ * [ENHANCEMENT] Improved logging about remote storage.
1254
+ * [ENHANCEMENT] Relaxed URL validation.
1255
+ * [ENHANCEMENT] Openstack SD: Handle instances without IP.
1256
+ * [ENHANCEMENT] Make remote storage queue manager configurable.
1257
+ * [ENHANCEMENT] Validate metrics returned from remote read.
1258
+ * [ENHANCEMENT] EC2 SD: Set a default region.
1259
+ * [ENHANCEMENT] Changed help link to `https://prometheus.io/docs`.
1260
+ * [BUGFIX] Fix floating-point precision issue in `deriv` function.
1261
+ * [BUGFIX] Fix pprof endpoints when -web.route-prefix or -web.external-url is
1262
+ used.
1263
+ * [BUGFIX] Fix handling of `null` target groups in file-based SD.
1264
+ * [BUGFIX] Set the sample timestamp in date-related PromQL functions.
1265
+ * [BUGFIX] Apply path prefix to redirect from deprecated graph URL.
1266
+ * [BUGFIX] Fixed tests on MS Windows.
1267
+ * [BUGFIX] Check for invalid UTF-8 in label values after relabeling.
1268
+
1269
+ ## 1.7.2 / 2017-09-26
1270
+
1271
+ * [BUGFIX] Correctly remove all targets from DNS service discovery if the
1272
+ corresponding DNS query succeeds and returns an empty result.
1273
+ * [BUGFIX] Correctly parse resolution input in expression browser.
1274
+ * [BUGFIX] Consistently use UTC in the date picker of the expression browser.
1275
+ * [BUGFIX] Correctly handle multiple ports in Marathon service discovery.
1276
+ * [BUGFIX] Fix HTML escaping so that HTML templates compile with Go1.9.
1277
+ * [BUGFIX] Prevent number of remote write shards from going negative.
1278
+ * [BUGFIX] In the graphs created by the expression browser, render very large
1279
+ and small numbers in a readable way.
1280
+ * [BUGFIX] Fix a rarely occurring iterator issue in varbit encoded chunks.
1281
+
1282
+ ## 1.7.1 / 2017-06-12
1283
+
1284
+ * [BUGFIX] Fix double prefix redirect.
1285
+
1286
+ ## 1.7.0 / 2017-06-06
1287
+
1288
+ * [CHANGE] Compress remote storage requests and responses with unframed/raw snappy.
1289
+ * [CHANGE] Properly ellide secrets in config.
1290
+ * [FEATURE] Add OpenStack service discovery.
1291
+ * [FEATURE] Add ability to limit Kubernetes service discovery to certain namespaces.
1292
+ * [FEATURE] Add metric for discovered number of Alertmanagers.
1293
+ * [ENHANCEMENT] Print system information (uname) on start up.
1294
+ * [ENHANCEMENT] Show gaps in graphs on expression browser.
1295
+ * [ENHANCEMENT] Promtool linter checks counter naming and more reserved labels.
1296
+ * [BUGFIX] Fix broken Mesos discovery.
1297
+ * [BUGFIX] Fix redirect when external URL is set.
1298
+ * [BUGFIX] Fix mutation of active alert elements by notifier.
1299
+ * [BUGFIX] Fix HTTP error handling for remote write.
1300
+ * [BUGFIX] Fix builds for Solaris/Illumos.
1301
+ * [BUGFIX] Fix overflow checking in global config.
1302
+ * [BUGFIX] Fix log level reporting issue.
1303
+ * [BUGFIX] Fix ZooKeeper serverset discovery can become out-of-sync.
1304
+
1305
+ ## 1.6.3 / 2017-05-18
1306
+
1307
+ * [BUGFIX] Fix disappearing Alertmanager targets in Alertmanager discovery.
1308
+ * [BUGFIX] Fix panic with remote_write on ARMv7.
1309
+ * [BUGFIX] Fix stacked graphs to adapt min/max values.
1310
+
1311
+ ## 1.6.2 / 2017-05-11
1312
+
1313
+ * [BUGFIX] Fix potential memory leak in Kubernetes service discovery
1314
+
1315
+ ## 1.6.1 / 2017-04-19
1316
+
1317
+ * [BUGFIX] Don't panic if storage has no FPs even after initial wait
1318
+
1319
+ ## 1.6.0 / 2017-04-14
1320
+
1321
+ * [CHANGE] Replaced the remote write implementations for various backends by a
1322
+ generic write interface with example adapter implementation for various
1323
+ backends. Note that both the previous and the current remote write
1324
+ implementations are **experimental**.
1325
+ * [FEATURE] New flag `-storage.local.target-heap-size` to tell Prometheus about
1326
+ the desired heap size. This deprecates the flags
1327
+ `-storage.local.memory-chunks` and `-storage.local.max-chunks-to-persist`,
1328
+ which are kept for backward compatibility.
1329
+ * [FEATURE] Add `check-metrics` to `promtool` to lint metric names.
1330
+ * [FEATURE] Add Joyent Triton discovery.
1331
+ * [FEATURE] `X-Prometheus-Scrape-Timeout-Seconds` header in HTTP scrape
1332
+ requests.
1333
+ * [FEATURE] Remote read interface, including example for InfluxDB. **Experimental.**
1334
+ * [FEATURE] Enable Consul SD to connect via TLS.
1335
+ * [FEATURE] Marathon SD supports multiple ports.
1336
+ * [FEATURE] Marathon SD supports bearer token for authentication.
1337
+ * [FEATURE] Custom timeout for queries.
1338
+ * [FEATURE] Expose `buildQueryUrl` in `graph.js`.
1339
+ * [FEATURE] Add `rickshawGraph` property to the graph object in console
1340
+ templates.
1341
+ * [FEATURE] New metrics exported by Prometheus itself:
1342
+ * Summary `prometheus_engine_query_duration_seconds`
1343
+ * Counter `prometheus_evaluator_iterations_missed_total`
1344
+ * Counter `prometheus_evaluator_iterations_total`
1345
+ * Gauge `prometheus_local_storage_open_head_chunks`
1346
+ * Gauge `prometheus_local_storage_target_heap_size`
1347
+ * [ENHANCEMENT] Reduce shut-down time by interrupting an ongoing checkpoint
1348
+ before starting the final checkpoint.
1349
+ * [ENHANCEMENT] Auto-tweak times between checkpoints to limit time spent in
1350
+ checkpointing to 50%.
1351
+ * [ENHANCEMENT] Improved crash recovery deals better with certain index
1352
+ corruptions.
1353
+ * [ENHANCEMENT] Graphing deals better with constant time series.
1354
+ * [ENHANCEMENT] Retry remote writes on recoverable errors.
1355
+ * [ENHANCEMENT] Evict unused chunk descriptors during crash recovery to limit
1356
+ memory usage.
1357
+ * [ENHANCEMENT] Smoother disk usage during series maintenance.
1358
+ * [ENHANCEMENT] Targets on targets page sorted by instance within a job.
1359
+ * [ENHANCEMENT] Sort labels in federation.
1360
+ * [ENHANCEMENT] Set `GOGC=40` by default, which results in much better memory
1361
+ utilization at the price of slightly higher CPU usage. If `GOGC` is set by
1362
+ the user, it is still honored as usual.
1363
+ * [ENHANCEMENT] Close head chunks after being idle for the duration of the
1364
+ configured staleness delta. This helps to persist and evict head chunk of
1365
+ stale series more quickly.
1366
+ * [ENHANCEMENT] Stricter checking of relabel config.
1367
+ * [ENHANCEMENT] Cache busters for static web content.
1368
+ * [ENHANCEMENT] Send Prometheus-specific user-agent header during scrapes.
1369
+ * [ENHANCEMENT] Improved performance of series retention cut-off.
1370
+ * [ENHANCEMENT] Mitigate impact of non-atomic sample ingestion on
1371
+ `histogram_quantile` by enforcing buckets to be monotonic.
1372
+ * [ENHANCEMENT] Released binaries built with Go 1.8.1.
1373
+ * [BUGFIX] Send `instance=""` with federation if `instance` not set.
1374
+ * [BUGFIX] Update to new `client_golang` to get rid of unwanted quantile
1375
+ metrics in summaries.
1376
+ * [BUGFIX] Introduce several additional guards against data corruption.
1377
+ * [BUGFIX] Mark storage dirty and increment
1378
+ `prometheus_local_storage_persist_errors_total` on all relevant errors.
1379
+ * [BUGFIX] Propagate storage errors as 500 in the HTTP API.
1380
+ * [BUGFIX] Fix int64 overflow in timestamps in the HTTP API.
1381
+ * [BUGFIX] Fix deadlock in Zookeeper SD.
1382
+ * [BUGFIX] Fix fuzzy search problems in the web-UI auto-completion.
1383
+
1384
+ ## 1.5.3 / 2017-05-11
1385
+
1386
+ * [BUGFIX] Fix potential memory leak in Kubernetes service discovery
1387
+
1388
+ ## 1.5.2 / 2017-02-10
1389
+
1390
+ * [BUGFIX] Fix series corruption in a special case of series maintenance where
1391
+ the minimum series-file-shrink-ratio kicks in.
1392
+ * [BUGFIX] Fix two panic conditions both related to processing a series
1393
+ scheduled to be quarantined.
1394
+ * [ENHANCEMENT] Binaries built with Go1.7.5.
1395
+
1396
+ ## 1.5.1 / 2017-02-07
1397
+
1398
+ * [BUGFIX] Don't lose fully persisted memory series during checkpointing.
1399
+ * [BUGFIX] Fix intermittently failing relabeling.
1400
+ * [BUGFIX] Make `-storage.local.series-file-shrink-ratio` work.
1401
+ * [BUGFIX] Remove race condition from TestLoop.
1402
+
1403
+ ## 1.5.0 / 2017-01-23
1404
+
1405
+ * [CHANGE] Use lexicographic order to sort alerts by name.
1406
+ * [FEATURE] Add Joyent Triton discovery.
1407
+ * [FEATURE] Add scrape targets and alertmanager targets API.
1408
+ * [FEATURE] Add various persistence related metrics.
1409
+ * [FEATURE] Add various query engine related metrics.
1410
+ * [FEATURE] Add ability to limit scrape samples, and related metrics.
1411
+ * [FEATURE] Add labeldrop and labelkeep relabelling actions.
1412
+ * [FEATURE] Display current working directory on status-page.
1413
+ * [ENHANCEMENT] Strictly use ServiceAccount for in cluster configuration on Kubernetes.
1414
+ * [ENHANCEMENT] Various performance and memory-management improvements.
1415
+ * [BUGFIX] Fix basic auth for alertmanagers configured via flag.
1416
+ * [BUGFIX] Don't panic on decoding corrupt data.
1417
+ * [BUGFIX] Ignore dotfiles in data directory.
1418
+ * [BUGFIX] Abort on intermediate federation errors.
1419
+
1420
+ ## 1.4.1 / 2016-11-28
1421
+
1422
+ * [BUGFIX] Fix Consul service discovery
1423
+
1424
+ ## 1.4.0 / 2016-11-25
1425
+
1426
+ * [FEATURE] Allow configuring Alertmanagers via service discovery
1427
+ * [FEATURE] Display used Alertmanagers on runtime page in the UI
1428
+ * [FEATURE] Support profiles in AWS EC2 service discovery configuration
1429
+ * [ENHANCEMENT] Remove duplicated logging of Kubernetes client errors
1430
+ * [ENHANCEMENT] Add metrics about Kubernetes service discovery
1431
+ * [BUGFIX] Update alert annotations on re-evaluation
1432
+ * [BUGFIX] Fix export of group modifier in PromQL queries
1433
+ * [BUGFIX] Remove potential deadlocks in several service discovery implementations
1434
+ * [BUGFIX] Use proper float64 modulo in PromQL `%` binary operations
1435
+ * [BUGFIX] Fix crash bug in Kubernetes service discovery
1436
+
1437
+ ## 1.3.1 / 2016-11-04
1438
+
1439
+ This bug-fix release pulls in the fixes from the 1.2.3 release.
1440
+
1441
+ * [BUGFIX] Correctly handle empty Regex entry in relabel config.
1442
+ * [BUGFIX] MOD (`%`) operator doesn't panic with small floating point numbers.
1443
+ * [BUGFIX] Updated miekg/dns vendoring to pick up upstream bug fixes.
1444
+ * [ENHANCEMENT] Improved DNS error reporting.
1445
+
1446
+ ## 1.2.3 / 2016-11-04
1447
+
1448
+ Note that this release is chronologically after 1.3.0.
1449
+
1450
+ * [BUGFIX] Correctly handle end time before start time in range queries.
1451
+ * [BUGFIX] Error on negative `-storage.staleness-delta`
1452
+ * [BUGFIX] Correctly handle empty Regex entry in relabel config.
1453
+ * [BUGFIX] MOD (`%`) operator doesn't panic with small floating point numbers.
1454
+ * [BUGFIX] Updated miekg/dns vendoring to pick up upstream bug fixes.
1455
+ * [ENHANCEMENT] Improved DNS error reporting.
1456
+
1457
+ ## 1.3.0 / 2016-11-01
1458
+
1459
+ This is a breaking change to the Kubernetes service discovery.
1460
+
1461
+ * [CHANGE] Rework Kubernetes SD.
1462
+ * [FEATURE] Add support for interpolating `target_label`.
1463
+ * [FEATURE] Add GCE metadata as Prometheus meta labels.
1464
+ * [ENHANCEMENT] Add EC2 SD metrics.
1465
+ * [ENHANCEMENT] Add Azure SD metrics.
1466
+ * [ENHANCEMENT] Add fuzzy search to `/graph` textarea.
1467
+ * [ENHANCEMENT] Always show instance labels on target page.
1468
+ * [BUGFIX] Validate query end time is not before start time.
1469
+ * [BUGFIX] Error on negative `-storage.staleness-delta`
1470
+
1471
+ ## 1.2.2 / 2016-10-30
1472
+
1473
+ * [BUGFIX] Correctly handle on() in alerts.
1474
+ * [BUGFIX] UI: Deal properly with aborted requests.
1475
+ * [BUGFIX] UI: Decode URL query parameters properly.
1476
+ * [BUGFIX] Storage: Deal better with data corruption (non-monotonic timestamps).
1477
+ * [BUGFIX] Remote storage: Re-add accidentally removed timeout flag.
1478
+ * [BUGFIX] Updated a number of vendored packages to pick up upstream bug fixes.
1479
+
1480
+ ## 1.2.1 / 2016-10-10
1481
+
1482
+ * [BUGFIX] Count chunk evictions properly so that the server doesn't
1483
+ assume it runs out of memory and subsequently throttles ingestion.
1484
+ * [BUGFIX] Use Go1.7.1 for prebuilt binaries to fix issues on MacOS Sierra.
1485
+
1486
+ ## 1.2.0 / 2016-10-07
1487
+
1488
+ * [FEATURE] Cleaner encoding of query parameters in `/graph` URLs.
1489
+ * [FEATURE] PromQL: Add `minute()` function.
1490
+ * [FEATURE] Add GCE service discovery.
1491
+ * [FEATURE] Allow any valid UTF-8 string as job name.
1492
+ * [FEATURE] Allow disabling local storage.
1493
+ * [FEATURE] EC2 service discovery: Expose `ec2_instance_state`.
1494
+ * [ENHANCEMENT] Various performance improvements in local storage.
1495
+ * [BUGFIX] Zookeeper service discovery: Remove deleted nodes.
1496
+ * [BUGFIX] Zookeeper service discovery: Resync state after Zookeeper failure.
1497
+ * [BUGFIX] Remove JSON from HTTP Accept header.
1498
+ * [BUGFIX] Fix flag validation of Alertmanager URL.
1499
+ * [BUGFIX] Fix race condition on shutdown.
1500
+ * [BUGFIX] Do not fail Consul discovery on Prometheus startup when Consul
1501
+ is down.
1502
+ * [BUGFIX] Handle NaN in `changes()` correctly.
1503
+ * [CHANGE] **Experimental** remote write path: Remove use of gRPC.
1504
+ * [CHANGE] **Experimental** remote write path: Configuration via config file
1505
+ rather than command line flags.
1506
+ * [FEATURE] **Experimental** remote write path: Add HTTP basic auth and TLS.
1507
+ * [FEATURE] **Experimental** remote write path: Support for relabelling.
1508
+
1509
+ ## 1.1.3 / 2016-09-16
1510
+
1511
+ * [ENHANCEMENT] Use golang-builder base image for tests in CircleCI.
1512
+ * [ENHANCEMENT] Added unit tests for federation.
1513
+ * [BUGFIX] Correctly de-dup metric families in federation output.
1514
+
1515
+ ## 1.1.2 / 2016-09-08
1516
+
1517
+ * [BUGFIX] Allow label names that coincide with keywords.
1518
+
1519
+ ## 1.1.1 / 2016-09-07
1520
+
1521
+ * [BUGFIX] Fix IPv6 escaping in service discovery integrations
1522
+ * [BUGFIX] Fix default scrape port assignment for IPv6
1523
+
1524
+ ## 1.1.0 / 2016-09-03
1525
+
1526
+ * [FEATURE] Add `quantile` and `quantile_over_time`.
1527
+ * [FEATURE] Add `stddev_over_time` and `stdvar_over_time`.
1528
+ * [FEATURE] Add various time and date functions.
1529
+ * [FEATURE] Added `toUpper` and `toLower` formatting to templates.
1530
+ * [FEATURE] Allow relabeling of alerts.
1531
+ * [FEATURE] Allow URLs in targets defined via a JSON file.
1532
+ * [FEATURE] Add idelta function.
1533
+ * [FEATURE] 'Remove graph' button on the /graph page.
1534
+ * [FEATURE] Kubernetes SD: Add node name and host IP to pod discovery.
1535
+ * [FEATURE] New remote storage write path. EXPERIMENTAL!
1536
+ * [ENHANCEMENT] Improve time-series index lookups.
1537
+ * [ENHANCEMENT] Forbid invalid relabel configurations.
1538
+ * [ENHANCEMENT] Improved various tests.
1539
+ * [ENHANCEMENT] Add crash recovery metric 'started_dirty'.
1540
+ * [ENHANCEMENT] Fix (and simplify) populating series iterators.
1541
+ * [ENHANCEMENT] Add job link on target page.
1542
+ * [ENHANCEMENT] Message on empty Alerts page.
1543
+ * [ENHANCEMENT] Various internal code refactorings and clean-ups.
1544
+ * [ENHANCEMENT] Various improvements in the build system.
1545
+ * [BUGFIX] Catch errors when unmarshaling delta/doubleDelta encoded chunks.
1546
+ * [BUGFIX] Fix data race in lexer and lexer test.
1547
+ * [BUGFIX] Trim stray whitespace from bearer token file.
1548
+ * [BUGFIX] Avoid divide-by-zero panic on query_range?step=0.
1549
+ * [BUGFIX] Detect invalid rule files at startup.
1550
+ * [BUGFIX] Fix counter reset treatment in PromQL.
1551
+ * [BUGFIX] Fix rule HTML escaping issues.
1552
+ * [BUGFIX] Remove internal labels from alerts sent to AM.
1553
+
1554
+ ## 1.0.2 / 2016-08-24
1555
+
1556
+ * [BUGFIX] Clean up old targets after config reload.
1557
+
1558
+ ## 1.0.1 / 2016-07-21
1559
+
1560
+ * [BUGFIX] Exit with error on non-flag command-line arguments.
1561
+ * [BUGFIX] Update example console templates to new HTTP API.
1562
+ * [BUGFIX] Re-add logging flags.
1563
+
1564
+ ## 1.0.0 / 2016-07-18
1565
+
1566
+ * [CHANGE] Remove deprecated query language keywords
1567
+ * [CHANGE] Change Kubernetes SD to require specifying Kubernetes role
1568
+ * [CHANGE] Use service address in Consul SD if available
1569
+ * [CHANGE] Standardize all Prometheus internal metrics to second units
1570
+ * [CHANGE] Remove unversioned legacy HTTP API
1571
+ * [CHANGE] Remove legacy ingestion of JSON metric format
1572
+ * [CHANGE] Remove deprecated `target_groups` configuration
1573
+ * [FEATURE] Add binary power operation to PromQL
1574
+ * [FEATURE] Add `count_values` aggregator
1575
+ * [FEATURE] Add `-web.route-prefix` flag
1576
+ * [FEATURE] Allow `on()`, `by()`, `without()` in PromQL with empty label sets
1577
+ * [ENHANCEMENT] Make `topk/bottomk` query functions aggregators
1578
+ * [BUGFIX] Fix annotations in alert rule printing
1579
+ * [BUGFIX] Expand alert templating at evaluation time
1580
+ * [BUGFIX] Fix edge case handling in crash recovery
1581
+ * [BUGFIX] Hide testing package flags from help output
1582
+
1583
+ ## 0.20.0 / 2016-06-15
1584
+
1585
+ This release contains multiple breaking changes to the configuration schema.
1586
+
1587
+ * [FEATURE] Allow configuring multiple Alertmanagers
1588
+ * [FEATURE] Add server name to TLS configuration
1589
+ * [FEATURE] Add labels for all node addresses and discover node port if available in Kubernetes SD
1590
+ * [ENHANCEMENT] More meaningful configuration errors
1591
+ * [ENHANCEMENT] Round scraping timestamps to milliseconds in web UI
1592
+ * [ENHANCEMENT] Make number of storage fingerprint locks configurable
1593
+ * [BUGFIX] Fix date parsing in console template graphs
1594
+ * [BUGFIX] Fix static console files in Docker images
1595
+ * [BUGFIX] Fix console JS XHR requests for IE11
1596
+ * [BUGFIX] Add missing path prefix in new status page
1597
+ * [CHANGE] Rename `target_groups` to `static_configs` in config files
1598
+ * [CHANGE] Rename `names` to `files` in file SD configuration
1599
+ * [CHANGE] Remove kubelet port config option in Kubernetes SD configuration
1600
+
1601
+ ## 0.19.3 / 2016-06-14
1602
+
1603
+ * [BUGFIX] Handle Marathon apps with zero ports
1604
+ * [BUGFIX] Fix startup panic in retrieval layer
1605
+
1606
+ ## 0.19.2 / 2016-05-29
1607
+
1608
+ * [BUGFIX] Correctly handle `GROUP_LEFT` and `GROUP_RIGHT` without labels in
1609
+ string representation of expressions and in rules.
1610
+ * [BUGFIX] Use `-web.external-url` for new status endpoints.
1611
+
1612
+ ## 0.19.1 / 2016-05-25
1613
+
1614
+ * [BUGFIX] Handle service discovery panic affecting Kubernetes SD
1615
+ * [BUGFIX] Fix web UI display issue in some browsers
1616
+
1617
+ ## 0.19.0 / 2016-05-24
1618
+
1619
+ This version contains a breaking change to the query language. Please read
1620
+ the documentation on the grouping behavior of vector matching:
1621
+
1622
+ <https://prometheus.io/docs/querying/operators/#vector-matching>
1623
+
1624
+ * [FEATURE] Add experimental Microsoft Azure service discovery
1625
+ * [FEATURE] Add `ignoring` modifier for binary operations
1626
+ * [FEATURE] Add pod discovery to Kubernetes service discovery
1627
+ * [CHANGE] Vector matching takes grouping labels from one-side
1628
+ * [ENHANCEMENT] Support time range on /api/v1/series endpoint
1629
+ * [ENHANCEMENT] Partition status page into individual pages
1630
+ * [BUGFIX] Fix issue of hanging target scrapes
1631
+
1632
+ ## 0.18.0 / 2016-04-18
1633
+
1634
+ * [BUGFIX] Fix operator precedence in PromQL
1635
+ * [BUGFIX] Never drop still open head chunk
1636
+ * [BUGFIX] Fix missing 'keep_common' when printing AST node
1637
+ * [CHANGE/BUGFIX] Target identity considers path and parameters additionally to host and port
1638
+ * [CHANGE] Rename metric `prometheus_local_storage_invalid_preload_requests_total` to `prometheus_local_storage_non_existent_series_matches_total`
1639
+ * [CHANGE] Support for old alerting rule syntax dropped
1640
+ * [FEATURE] Deduplicate targets within the same scrape job
1641
+ * [FEATURE] Add varbit chunk encoding (higher compression, more CPU usage – disabled by default)
1642
+ * [FEATURE] Add `holt_winters` query function
1643
+ * [FEATURE] Add relative complement `unless` operator to PromQL
1644
+ * [ENHANCEMENT] Quarantine series file if data corruption is encountered (instead of crashing)
1645
+ * [ENHANCEMENT] Validate Alertmanager URL
1646
+ * [ENHANCEMENT] Use UTC for build timestamp
1647
+ * [ENHANCEMENT] Improve index query performance (especially for active time series)
1648
+ * [ENHANCEMENT] Instrument configuration reload duration
1649
+ * [ENHANCEMENT] Instrument retrieval layer
1650
+ * [ENHANCEMENT] Add Go version to `prometheus_build_info` metric
1651
+
1652
+ ## 0.17.0 / 2016-03-02
1653
+
1654
+ This version no longer works with Alertmanager 0.0.4 and earlier!
1655
+ The alerting rule syntax has changed as well but the old syntax is supported
1656
+ up until version 0.18.
1657
+
1658
+ All regular expressions in PromQL are anchored now, matching the behavior of
1659
+ regular expressions in config files.
1660
+
1661
+ * [CHANGE] Integrate with Alertmanager 0.1.0 and higher
1662
+ * [CHANGE] Degraded storage mode renamed to rushed mode
1663
+ * [CHANGE] New alerting rule syntax
1664
+ * [CHANGE] Add label validation on ingestion
1665
+ * [CHANGE] Regular expression matchers in PromQL are anchored
1666
+ * [FEATURE] Add `without` aggregation modifier
1667
+ * [FEATURE] Send alert resolved notifications to Alertmanager
1668
+ * [FEATURE] Allow millisecond precision in configuration file
1669
+ * [FEATURE] Support AirBnB's Smartstack Nerve for service discovery
1670
+ * [ENHANCEMENT] Storage switches less often between regular and rushed mode.
1671
+ * [ENHANCEMENT] Storage switches into rushed mode if there are too many memory chunks.
1672
+ * [ENHANCEMENT] Added more storage instrumentation
1673
+ * [ENHANCEMENT] Improved instrumentation of notification handler
1674
+ * [BUGFIX] Do not count head chunks as chunks waiting for persistence
1675
+ * [BUGFIX] Handle OPTIONS HTTP requests to the API correctly
1676
+ * [BUGFIX] Parsing of ranges in PromQL fixed
1677
+ * [BUGFIX] Correctly validate URL flag parameters
1678
+ * [BUGFIX] Log argument parse errors
1679
+ * [BUGFIX] Properly handle creation of target with bad TLS config
1680
+ * [BUGFIX] Fix of checkpoint timing issue
1681
+
1682
+ ## 0.16.2 / 2016-01-18
1683
+
1684
+ * [FEATURE] Multiple authentication options for EC2 discovery added
1685
+ * [FEATURE] Several meta labels for EC2 discovery added
1686
+ * [FEATURE] Allow full URLs in static target groups (used e.g. by the `blackbox_exporter`)
1687
+ * [FEATURE] Add Graphite remote-storage integration
1688
+ * [FEATURE] Create separate Kubernetes targets for services and their endpoints
1689
+ * [FEATURE] Add `clamp_{min,max}` functions to PromQL
1690
+ * [FEATURE] Omitted time parameter in API query defaults to now
1691
+ * [ENHANCEMENT] Less frequent time series file truncation
1692
+ * [ENHANCEMENT] Instrument number of manually deleted time series
1693
+ * [ENHANCEMENT] Ignore lost+found directory during storage version detection
1694
+ * [CHANGE] Kubernetes `masters` renamed to `api_servers`
1695
+ * [CHANGE] "Healthy" and "unhealthy" targets are now called "up" and "down" in the web UI
1696
+ * [CHANGE] Remove undocumented 2nd argument of the `delta` function.
1697
+ (This is a BREAKING CHANGE for users of the undocumented 2nd argument.)
1698
+ * [BUGFIX] Return proper HTTP status codes on API errors
1699
+ * [BUGFIX] Fix Kubernetes authentication configuration
1700
+ * [BUGFIX] Fix stripped OFFSET from in rule evaluation and display
1701
+ * [BUGFIX] Do not crash on failing Consul SD initialization
1702
+ * [BUGFIX] Revert changes to metric auto-completion
1703
+ * [BUGFIX] Add config overflow validation for TLS configuration
1704
+ * [BUGFIX] Skip already watched Zookeeper nodes in serverset SD
1705
+ * [BUGFIX] Don't federate stale samples
1706
+ * [BUGFIX] Move NaN to end of result for `topk/bottomk/sort/sort_desc/min/max`
1707
+ * [BUGFIX] Limit extrapolation of `delta/rate/increase`
1708
+ * [BUGFIX] Fix unhandled error in rule evaluation
1709
+
1710
+ Some changes to the Kubernetes service discovery were integration since
1711
+ it was released as a beta feature.
1712
+
1713
+ ## 0.16.1 / 2015-10-16
1714
+
1715
+ * [FEATURE] Add `irate()` function.
1716
+ * [ENHANCEMENT] Improved auto-completion in expression browser.
1717
+ * [CHANGE] Kubernetes SD moves node label to instance label.
1718
+ * [BUGFIX] Escape regexes in console templates.
1719
+
1720
+ ## 0.16.0 / 2015-10-09
1721
+
1722
+ BREAKING CHANGES:
1723
+
1724
+ * Release tarballs now contain the built binaries in a nested directory.
1725
+ * The `hash_mod` relabeling action now uses MD5 hashes instead of FNV hashes to
1726
+ achieve a better distribution.
1727
+ * The DNS-SD meta label `__meta_dns_srv_name` was renamed to `__meta_dns_name`
1728
+ to reflect support for DNS record types other than `SRV`.
1729
+ * The default full refresh interval for the file-based service discovery has been
1730
+ increased from 30 seconds to 5 minutes.
1731
+ * In relabeling, parts of a source label that weren't matched by
1732
+ the specified regular expression are no longer included in the replacement
1733
+ output.
1734
+ * Queries no longer interpolate between two data points. Instead, the resulting
1735
+ value will always be the latest value before the evaluation query timestamp.
1736
+ * Regular expressions supplied via the configuration are now anchored to match
1737
+ full strings instead of substrings.
1738
+ * Global labels are not appended upon storing time series anymore. Instead,
1739
+ they are only appended when communicating with external systems
1740
+ (Alertmanager, remote storages, federation). They have thus also been renamed
1741
+ from `global.labels` to `global.external_labels`.
1742
+ * The names and units of metrics related to remote storage sample appends have
1743
+ been changed.
1744
+ * The experimental support for writing to InfluxDB has been updated to work
1745
+ with InfluxDB 0.9.x. 0.8.x versions of InfluxDB are not supported anymore.
1746
+ * Escape sequences in double- and single-quoted string literals in rules or query
1747
+ expressions are now interpreted like escape sequences in Go string literals
1748
+ (<https://golang.org/ref/spec#String_literals>).
1749
+
1750
+ Future breaking changes / deprecated features:
1751
+
1752
+ * The `delta()` function had an undocumented optional second boolean argument
1753
+ to make it behave like `increase()`. This second argument will be removed in
1754
+ the future. Migrate any occurrences of `delta(x, 1)` to use `increase(x)`
1755
+ instead.
1756
+ * Support for filter operators between two scalar values (like `2 > 1`) will be
1757
+ removed in the future. These will require a `bool` modifier on the operator,
1758
+ e.g. `2 > bool 1`.
1759
+
1760
+ All changes:
1761
+
1762
+ * [CHANGE] Renamed `global.labels` to `global.external_labels`.
1763
+ * [CHANGE] Vendoring is now done via govendor instead of godep.
1764
+ * [CHANGE] Change web UI root page to show the graphing interface instead of
1765
+ the server status page.
1766
+ * [CHANGE] Append global labels only when communicating with external systems
1767
+ instead of storing them locally.
1768
+ * [CHANGE] Change all regexes in the configuration to do full-string matches
1769
+ instead of substring matches.
1770
+ * [CHANGE] Remove interpolation of vector values in queries.
1771
+ * [CHANGE] For alert `SUMMARY`/`DESCRIPTION` template fields, cast the alert
1772
+ value to `float64` to work with common templating functions.
1773
+ * [CHANGE] In relabeling, don't include unmatched source label parts in the
1774
+ replacement.
1775
+ * [CHANGE] Change default full refresh interval for the file-based service
1776
+ discovery from 30 seconds to 5 minutes.
1777
+ * [CHANGE] Rename the DNS-SD meta label `__meta_dns_srv_name` to
1778
+ `__meta_dns_name` to reflect support for other record types than `SRV`.
1779
+ * [CHANGE] Release tarballs now contain the binaries in a nested directory.
1780
+ * [CHANGE] Update InfluxDB write support to work with InfluxDB 0.9.x.
1781
+ * [FEATURE] Support full "Go-style" escape sequences in strings and add raw
1782
+ string literals.
1783
+ * [FEATURE] Add EC2 service discovery support.
1784
+ * [FEATURE] Allow configuring TLS options in scrape configurations.
1785
+ * [FEATURE] Add instrumentation around configuration reloads.
1786
+ * [FEATURE] Add `bool` modifier to comparison operators to enable boolean
1787
+ (`0`/`1`) output instead of filtering.
1788
+ * [FEATURE] In Zookeeper serverset discovery, provide `__meta_serverset_shard`
1789
+ label with the serverset shard number.
1790
+ * [FEATURE] Provide `__meta_consul_service_id` meta label in Consul service
1791
+ discovery.
1792
+ * [FEATURE] Allow scalar expressions in recording rules to enable use cases
1793
+ such as building constant metrics.
1794
+ * [FEATURE] Add `label_replace()` and `vector()` query language functions.
1795
+ * [FEATURE] In Consul service discovery, fill in the `__meta_consul_dc`
1796
+ datacenter label from the Consul agent when it's not set in the Consul SD
1797
+ config.
1798
+ * [FEATURE] Scrape all services upon empty services list in Consul service
1799
+ discovery.
1800
+ * [FEATURE] Add `labelmap` relabeling action to map a set of input labels to a
1801
+ set of output labels using regular expressions.
1802
+ * [FEATURE] Introduce `__tmp` as a relabeling label prefix that is guaranteed
1803
+ to not be used by Prometheus internally.
1804
+ * [FEATURE] Kubernetes-based service discovery.
1805
+ * [FEATURE] Marathon-based service discovery.
1806
+ * [FEATURE] Support multiple series names in console graphs JavaScript library.
1807
+ * [FEATURE] Allow reloading configuration via web handler at `/-/reload`.
1808
+ * [FEATURE] Updates to promtool to reflect new Prometheus configuration
1809
+ features.
1810
+ * [FEATURE] Add `proxy_url` parameter to scrape configurations to enable use of
1811
+ proxy servers.
1812
+ * [FEATURE] Add console templates for Prometheus itself.
1813
+ * [FEATURE] Allow relabeling the protocol scheme of targets.
1814
+ * [FEATURE] Add `predict_linear()` query language function.
1815
+ * [FEATURE] Support for authentication using bearer tokens, client certs, and
1816
+ CA certs.
1817
+ * [FEATURE] Implement unary expressions for vector types (`-foo`, `+foo`).
1818
+ * [FEATURE] Add console templates for the SNMP exporter.
1819
+ * [FEATURE] Make it possible to relabel target scrape query parameters.
1820
+ * [FEATURE] Add support for `A` and `AAAA` records in DNS service discovery.
1821
+ * [ENHANCEMENT] Fix several flaky tests.
1822
+ * [ENHANCEMENT] Switch to common routing package.
1823
+ * [ENHANCEMENT] Use more resilient metric decoder.
1824
+ * [ENHANCEMENT] Update vendored dependencies.
1825
+ * [ENHANCEMENT] Add compression to more HTTP handlers.
1826
+ * [ENHANCEMENT] Make -web.external-url flag help string more verbose.
1827
+ * [ENHANCEMENT] Improve metrics around remote storage queues.
1828
+ * [ENHANCEMENT] Use Go 1.5.1 instead of Go 1.4.2 in builds.
1829
+ * [ENHANCEMENT] Update the architecture diagram in the `README.md`.
1830
+ * [ENHANCEMENT] Time out sample appends in retrieval layer if the storage is
1831
+ backlogging.
1832
+ * [ENHANCEMENT] Make `hash_mod` relabeling action use MD5 instead of FNV to
1833
+ enable better hash distribution.
1834
+ * [ENHANCEMENT] Better tracking of targets between same service discovery
1835
+ mechanisms in one scrape configuration.
1836
+ * [ENHANCEMENT] Handle parser and query evaluation runtime panics more
1837
+ gracefully.
1838
+ * [ENHANCEMENT] Add IDs to H2 tags on status page to allow anchored linking.
1839
+ * [BUGFIX] Fix watching multiple paths with Zookeeper serverset discovery.
1840
+ * [BUGFIX] Fix high CPU usage on configuration reload.
1841
+ * [BUGFIX] Fix disappearing `__params` on configuration reload.
1842
+ * [BUGFIX] Make `labelmap` action available through configuration.
1843
+ * [BUGFIX] Fix direct access of protobuf fields.
1844
+ * [BUGFIX] Fix panic on Consul request error.
1845
+ * [BUGFIX] Redirect of graph endpoint for prefixed setups.
1846
+ * [BUGFIX] Fix series file deletion behavior when purging archived series.
1847
+ * [BUGFIX] Fix error checking and logging around checkpointing.
1848
+ * [BUGFIX] Fix map initialization in target manager.
1849
+ * [BUGFIX] Fix draining of file watcher events in file-based service discovery.
1850
+ * [BUGFIX] Add `POST` handler for `/debug` endpoints to fix CPU profiling.
1851
+ * [BUGFIX] Fix several flaky tests.
1852
+ * [BUGFIX] Fix busylooping in case a scrape configuration has no target
1853
+ providers defined.
1854
+ * [BUGFIX] Fix exit behavior of static target provider.
1855
+ * [BUGFIX] Fix configuration reloading loop upon shutdown.
1856
+ * [BUGFIX] Add missing check for nil expression in expression parser.
1857
+ * [BUGFIX] Fix error handling bug in test code.
1858
+ * [BUGFIX] Fix Consul port meta label.
1859
+ * [BUGFIX] Fix lexer bug that treated non-Latin Unicode digits as digits.
1860
+ * [CLEANUP] Remove obsolete federation example from console templates.
1861
+ * [CLEANUP] Remove duplicated Bootstrap JS inclusion on graph page.
1862
+ * [CLEANUP] Switch to common log package.
1863
+ * [CLEANUP] Update build environment scripts and Makefiles to work better with
1864
+ native Go build mechanisms and new Go 1.5 experimental vendoring support.
1865
+ * [CLEANUP] Remove logged notice about 0.14.x configuration file format change.
1866
+ * [CLEANUP] Move scrape-time metric label modification into SampleAppenders.
1867
+ * [CLEANUP] Switch from `github.com/client_golang/model` to
1868
+ `github.com/common/model` and related type cleanups.
1869
+ * [CLEANUP] Switch from `github.com/client_golang/extraction` to
1870
+ `github.com/common/expfmt` and related type cleanups.
1871
+ * [CLEANUP] Exit Prometheus when the web server encounters a startup error.
1872
+ * [CLEANUP] Remove non-functional alert-silencing links on alerting page.
1873
+ * [CLEANUP] General cleanups to comments and code, derived from `golint`,
1874
+ `go vet`, or otherwise.
1875
+ * [CLEANUP] When entering crash recovery, tell users how to cleanly shut down
1876
+ Prometheus.
1877
+ * [CLEANUP] Remove internal support for multi-statement queries in query engine.
1878
+ * [CLEANUP] Update AUTHORS.md.
1879
+ * [CLEANUP] Don't warn/increment metric upon encountering equal timestamps for
1880
+ the same series upon append.
1881
+ * [CLEANUP] Resolve relative paths during configuration loading.
1882
+
1883
+ ## 0.15.1 / 2015-07-27
1884
+
1885
+ * [BUGFIX] Fix vector matching behavior when there is a mix of equality and
1886
+ non-equality matchers in a vector selector and one matcher matches no series.
1887
+ * [ENHANCEMENT] Allow overriding `GOARCH` and `GOOS` in Makefile.INCLUDE.
1888
+ * [ENHANCEMENT] Update vendored dependencies.
1889
+
1890
+ ## 0.15.0 / 2015-07-21
1891
+
1892
+ BREAKING CHANGES:
1893
+
1894
+ * Relative paths for rule files are now evaluated relative to the config file.
1895
+ * External reachability flags (`-web.*`) consolidated.
1896
+ * The default storage directory has been changed from `/tmp/metrics`
1897
+ to `data` in the local directory.
1898
+ * The `rule_checker` tool has been replaced by `promtool` with
1899
+ different flags and more functionality.
1900
+ * Empty labels are now removed upon ingestion into the
1901
+ storage. Matching empty labels is now equivalent to matching unset
1902
+ labels (`mymetric{label=""}` now matches series that don't have
1903
+ `label` set at all).
1904
+ * The special `__meta_consul_tags` label in Consul service discovery
1905
+ now starts and ends with tag separators to enable easier regex
1906
+ matching.
1907
+ * The default scrape interval has been changed back from 1 minute to
1908
+ 10 seconds.
1909
+
1910
+ All changes:
1911
+
1912
+ * [CHANGE] Change default storage directory to `data` in the current
1913
+ working directory.
1914
+ * [CHANGE] Consolidate external reachability flags (`-web.*`)into one.
1915
+ * [CHANGE] Deprecate `keeping_extra` modifier keyword, rename it to
1916
+ `keep_common`.
1917
+ * [CHANGE] Improve label matching performance and treat unset labels
1918
+ like empty labels in label matchers.
1919
+ * [CHANGE] Remove `rule_checker` tool and add generic `promtool` CLI
1920
+ tool which allows checking rules and configuration files.
1921
+ * [CHANGE] Resolve rule files relative to config file.
1922
+ * [CHANGE] Restore default ScrapeInterval of 1 minute instead of 10 seconds.
1923
+ * [CHANGE] Surround `__meta_consul_tags` value with tag separators.
1924
+ * [CHANGE] Update node disk console for new filesystem labels.
1925
+ * [FEATURE] Add Consul's `ServiceAddress`, `Address`, and `ServicePort` as
1926
+ meta labels to enable setting a custom scrape address if needed.
1927
+ * [FEATURE] Add `hashmod` relabel action to allow for horizontal
1928
+ sharding of Prometheus servers.
1929
+ * [FEATURE] Add `honor_labels` scrape configuration option to not
1930
+ overwrite any labels exposed by the target.
1931
+ * [FEATURE] Add basic federation support on `/federate`.
1932
+ * [FEATURE] Add optional `RUNBOOK` field to alert statements.
1933
+ * [FEATURE] Add pre-relabel target labels to status page.
1934
+ * [FEATURE] Add version information endpoint under `/version`.
1935
+ * [FEATURE] Added initial stable API version 1 under `/api/v1`,
1936
+ including ability to delete series and query more metadata.
1937
+ * [FEATURE] Allow configuring query parameters when scraping metrics endpoints.
1938
+ * [FEATURE] Allow deleting time series via the new v1 API.
1939
+ * [FEATURE] Allow individual ingested metrics to be relabeled.
1940
+ * [FEATURE] Allow loading rule files from an entire directory.
1941
+ * [FEATURE] Allow scalar expressions in range queries, improve error messages.
1942
+ * [FEATURE] Support Zookeeper Serversets as a service discovery mechanism.
1943
+ * [ENHANCEMENT] Add circleci yaml for Dockerfile test build.
1944
+ * [ENHANCEMENT] Always show selected graph range, regardless of available data.
1945
+ * [ENHANCEMENT] Change expression input field to multi-line textarea.
1946
+ * [ENHANCEMENT] Enforce strict monotonicity of time stamps within a series.
1947
+ * [ENHANCEMENT] Export build information as metric.
1948
+ * [ENHANCEMENT] Improve UI of `/alerts` page.
1949
+ * [ENHANCEMENT] Improve display of target labels on status page.
1950
+ * [ENHANCEMENT] Improve initialization and routing functionality of web service.
1951
+ * [ENHANCEMENT] Improve target URL handling and display.
1952
+ * [ENHANCEMENT] New dockerfile using alpine-glibc base image and make.
1953
+ * [ENHANCEMENT] Other minor fixes.
1954
+ * [ENHANCEMENT] Preserve alert state across reloads.
1955
+ * [ENHANCEMENT] Prettify flag help output even more.
1956
+ * [ENHANCEMENT] README.md updates.
1957
+ * [ENHANCEMENT] Raise error on unknown config parameters.
1958
+ * [ENHANCEMENT] Refine v1 HTTP API output.
1959
+ * [ENHANCEMENT] Show original configuration file contents on status
1960
+ page instead of serialized YAML.
1961
+ * [ENHANCEMENT] Start HUP signal handler earlier to not exit upon HUP
1962
+ during startup.
1963
+ * [ENHANCEMENT] Updated vendored dependencies.
1964
+ * [BUGFIX] Do not panic in `StringToDuration()` on wrong duration unit.
1965
+ * [BUGFIX] Exit on invalid rule files on startup.
1966
+ * [BUGFIX] Fix a regression in the `.Path` console template variable.
1967
+ * [BUGFIX] Fix chunk descriptor loading.
1968
+ * [BUGFIX] Fix consoles "Prometheus" link to point to /
1969
+ * [BUGFIX] Fix empty configuration file cases
1970
+ * [BUGFIX] Fix float to int conversions in chunk encoding, which were
1971
+ broken for some architectures.
1972
+ * [BUGFIX] Fix overflow detection for serverset config.
1973
+ * [BUGFIX] Fix race conditions in retrieval layer.
1974
+ * [BUGFIX] Fix shutdown deadlock in Consul SD code.
1975
+ * [BUGFIX] Fix the race condition targets in the Makefile.
1976
+ * [BUGFIX] Fix value display error in web console.
1977
+ * [BUGFIX] Hide authentication credentials in config `String()` output.
1978
+ * [BUGFIX] Increment dirty counter metric in storage only if
1979
+ `setDirty(true)` is called.
1980
+ * [BUGFIX] Periodically refresh services in Consul to recover from
1981
+ missing events.
1982
+ * [BUGFIX] Prevent overwrite of default global config when loading a
1983
+ configuration.
1984
+ * [BUGFIX] Properly lex `\r` as whitespace in expression language.
1985
+ * [BUGFIX] Validate label names in JSON target groups.
1986
+ * [BUGFIX] Validate presence of regex field in relabeling configurations.
1987
+ * [CLEANUP] Clean up initialization of remote storage queues.
1988
+ * [CLEANUP] Fix `go vet` and `golint` violations.
1989
+ * [CLEANUP] General cleanup of rules and query language code.
1990
+ * [CLEANUP] Improve and simplify Dockerfile build steps.
1991
+ * [CLEANUP] Improve and simplify build infrastructure, use go-bindata
1992
+ for web assets. Allow building without git.
1993
+ * [CLEANUP] Move all utility packages into common `util` subdirectory.
1994
+ * [CLEANUP] Refactor main, flag handling, and web package.
1995
+ * [CLEANUP] Remove unused methods from `Rule` interface.
1996
+ * [CLEANUP] Simplify default config handling.
1997
+ * [CLEANUP] Switch human-readable times on web UI to UTC.
1998
+ * [CLEANUP] Use `templates.TemplateExpander` for all page templates.
1999
+ * [CLEANUP] Use new v1 HTTP API for querying and graphing.
2000
+
2001
+ ## 0.14.0 / 2015-06-01
2002
+
2003
+ * [CHANGE] Configuration format changed and switched to YAML.
2004
+ (See the provided [migration tool](https://github.com/prometheus/migrate/releases).)
2005
+ * [ENHANCEMENT] Redesign of state-preserving target discovery.
2006
+ * [ENHANCEMENT] Allow specifying scrape URL scheme and basic HTTP auth for non-static targets.
2007
+ * [FEATURE] Allow attaching meaningful labels to targets via relabeling.
2008
+ * [FEATURE] Configuration/rule reloading at runtime.
2009
+ * [FEATURE] Target discovery via file watches.
2010
+ * [FEATURE] Target discovery via Consul.
2011
+ * [ENHANCEMENT] Simplified binary operation evaluation.
2012
+ * [ENHANCEMENT] More stable component initialization.
2013
+ * [ENHANCEMENT] Added internal expression testing language.
2014
+ * [BUGFIX] Fix graph links with path prefix.
2015
+ * [ENHANCEMENT] Allow building from source without git.
2016
+ * [ENHANCEMENT] Improve storage iterator performance.
2017
+ * [ENHANCEMENT] Change logging output format and flags.
2018
+ * [BUGFIX] Fix memory alignment bug for 32bit systems.
2019
+ * [ENHANCEMENT] Improve web redirection behavior.
2020
+ * [ENHANCEMENT] Allow overriding default hostname for Prometheus URLs.
2021
+ * [BUGFIX] Fix double slash in URL sent to alertmanager.
2022
+ * [FEATURE] Add resets() query function to count counter resets.
2023
+ * [FEATURE] Add changes() query function to count the number of times a gauge changed.
2024
+ * [FEATURE] Add increase() query function to calculate a counter's increase.
2025
+ * [ENHANCEMENT] Limit retrievable samples to the storage's retention window.
2026
+
2027
+ ## 0.13.4 / 2015-05-23
2028
+
2029
+ * [BUGFIX] Fix a race while checkpointing fingerprint mappings.
2030
+
2031
+ ## 0.13.3 / 2015-05-11
2032
+
2033
+ * [BUGFIX] Handle fingerprint collisions properly.
2034
+ * [CHANGE] Comments in rules file must start with `#`. (The undocumented `//`
2035
+ and `/*...*/` comment styles are no longer supported.)
2036
+ * [ENHANCEMENT] Switch to custom expression language parser and evaluation
2037
+ engine, which generates better error messages, fixes some parsing edge-cases,
2038
+ and enables other future enhancements (like the ones below).
2039
+ * [ENHANCEMENT] Limit maximum number of concurrent queries.
2040
+ * [ENHANCEMENT] Terminate running queries during shutdown.
2041
+
2042
+ ## 0.13.2 / 2015-05-05
2043
+
2044
+ * [MAINTENANCE] Updated vendored dependencies to their newest versions.
2045
+ * [MAINTENANCE] Include rule_checker and console templates in release tarball.
2046
+ * [BUGFIX] Sort NaN as the lowest value.
2047
+ * [ENHANCEMENT] Add square root, stddev and stdvar functions.
2048
+ * [BUGFIX] Use scrape_timeout for scrape timeout, not scrape_interval.
2049
+ * [ENHANCEMENT] Improve chunk and chunkDesc loading, increase performance when
2050
+ reading from disk.
2051
+ * [BUGFIX] Show correct error on wrong DNS response.
2052
+
2053
+ ## 0.13.1 / 2015-04-09
2054
+
2055
+ * [BUGFIX] Treat memory series with zero chunks correctly in series maintenance.
2056
+ * [ENHANCEMENT] Improve readability of usage text even more.
2057
+
2058
+ ## 0.13.0 / 2015-04-08
2059
+
2060
+ * [ENHANCEMENT] Double-delta encoding for chunks, saving typically 40% of
2061
+ space, both in RAM and on disk.
2062
+ * [ENHANCEMENT] Redesign of chunk persistence queuing, increasing performance
2063
+ on spinning disks significantly.
2064
+ * [ENHANCEMENT] Redesign of sample ingestion, increasing ingestion performance.
2065
+ * [FEATURE] Added ln, log2, log10 and exp functions to the query language.
2066
+ * [FEATURE] Experimental write support to InfluxDB.
2067
+ * [FEATURE] Allow custom timestamps in instant query API.
2068
+ * [FEATURE] Configurable path prefix for URLs to support proxies.
2069
+ * [ENHANCEMENT] Increase of rule_checker CLI usability.
2070
+ * [CHANGE] Show special float values as gaps.
2071
+ * [ENHANCEMENT] Made usage output more readable.
2072
+ * [ENHANCEMENT] Increased resilience of the storage against data corruption.
2073
+ * [ENHANCEMENT] Various improvements around chunk encoding.
2074
+ * [ENHANCEMENT] Nicer formatting of target health table on /status.
2075
+ * [CHANGE] Rename UNREACHABLE to UNHEALTHY, ALIVE to HEALTHY.
2076
+ * [BUGFIX] Strip trailing slash in alertmanager URL.
2077
+ * [BUGFIX] Avoid +InfYs and similar, just display +Inf.
2078
+ * [BUGFIX] Fixed HTML-escaping at various places.
2079
+ * [BUGFIX] Fixed special value handling in division and modulo of the query
2080
+ language.
2081
+ * [BUGFIX] Fix embed-static.sh.
2082
+ * [CLEANUP] Added initial HTTP API tests.
2083
+ * [CLEANUP] Misc. other code cleanups.
2084
+ * [MAINTENANCE] Updated vendored dependencies to their newest versions.
2085
+
2086
+ ## 0.12.0 / 2015-03-04
2087
+
2088
+ * [CHANGE] Use client_golang v0.3.1. THIS CHANGES FINGERPRINTING AND INVALIDATES
2089
+ ALL PERSISTED FINGERPRINTS. You have to wipe your storage to use this or
2090
+ later versions. There is a version guard in place that will prevent you to
2091
+ run Prometheus with the stored data of an older Prometheus.
2092
+ * [BUGFIX] The change above fixes a weakness in the fingerprinting algorithm.
2093
+ * [ENHANCEMENT] The change above makes fingerprinting faster and less allocation
2094
+ intensive.
2095
+ * [FEATURE] OR operator and vector matching options. See docs for details.
2096
+ * [ENHANCEMENT] Scientific notation and special float values (Inf, NaN) now
2097
+ supported by the expression language.
2098
+ * [CHANGE] Dockerfile makes Prometheus use the Docker volume to store data
2099
+ (rather than /tmp/metrics).
2100
+ * [CHANGE] Makefile uses Go 1.4.2.
2101
+
2102
+ ## 0.11.1 / 2015-02-27
2103
+
2104
+ * [BUGFIX] Make series maintenance complete again. (Ever since 0.9.0rc4,
2105
+ or commit 0851945, series would not be archived, chunk descriptors would
2106
+ not be evicted, and stale head chunks would never be closed. This happened
2107
+ due to accidental deletion of a line calling a (well tested :) function.
2108
+ * [BUGFIX] Do not double count head chunks read from checkpoint on startup.
2109
+ Also fix a related but less severe bug in counting chunk descriptors.
2110
+ * [BUGFIX] Check last time in head chunk for head chunk timeout, not first.
2111
+ * [CHANGE] Update vendoring due to vendoring changes in client_golang.
2112
+ * [CLEANUP] Code cleanups.
2113
+ * [ENHANCEMENT] Limit the number of 'dirty' series counted during checkpointing.
2114
+
2115
+ ## 0.11.0 / 2015-02-23
2116
+
2117
+ * [FEATURE] Introduce new metric type Histogram with server-side aggregation.
2118
+ * [FEATURE] Add offset operator.
2119
+ * [FEATURE] Add floor, ceil and round functions.
2120
+ * [CHANGE] Change instance identifiers to be host:port.
2121
+ * [CHANGE] Dependency management and vendoring changed/improved.
2122
+ * [CHANGE] Flag name changes to create consistency between various Prometheus
2123
+ binaries.
2124
+ * [CHANGE] Show unlimited number of metrics in autocomplete.
2125
+ * [CHANGE] Add query timeout.
2126
+ * [CHANGE] Remove labels on persist error counter.
2127
+ * [ENHANCEMENT] Various performance improvements for sample ingestion.
2128
+ * [ENHANCEMENT] Various Makefile improvements.
2129
+ * [ENHANCEMENT] Various console template improvements, including
2130
+ proof-of-concept for federation via console templates.
2131
+ * [ENHANCEMENT] Fix graph JS glitches and simplify graphing code.
2132
+ * [ENHANCEMENT] Dramatically decrease resources for file embedding.
2133
+ * [ENHANCEMENT] Crash recovery saves lost series data in 'orphaned' directory.
2134
+ * [BUGFIX] Fix aggregation grouping key calculation.
2135
+ * [BUGFIX] Fix Go download path for various architectures.
2136
+ * [BUGFIX] Fixed the link of the Travis build status image.
2137
+ * [BUGFIX] Fix Rickshaw/D3 version mismatch.
2138
+ * [CLEANUP] Various code cleanups.
2139
+
2140
+ ## 0.10.0 / 2015-01-26
2141
+
2142
+ * [CHANGE] More efficient JSON result format in query API. This requires
2143
+ up-to-date versions of PromDash and prometheus_cli, too.
2144
+ * [ENHANCEMENT] Excluded non-minified Bootstrap assets and the Bootstrap maps
2145
+ from embedding into the binary. Those files are only used for debugging,
2146
+ and then you can use -web.use-local-assets. By including fewer files, the
2147
+ RAM usage during compilation is much more manageable.
2148
+ * [ENHANCEMENT] Help link points to <https://prometheus.github.io> now.
2149
+ * [FEATURE] Consoles for haproxy and cloudwatch.
2150
+ * [BUGFIX] Several fixes to graphs in consoles.
2151
+ * [CLEANUP] Removed a file size check that did not check anything.
2152
+
2153
+ ## 0.9.0 / 2015-01-23
2154
+
2155
+ * [CHANGE] Reworked command line flags, now more consistent and taking into
2156
+ account needs of the new storage backend (see below).
2157
+ * [CHANGE] Metric names are dropped after certain transformations.
2158
+ * [CHANGE] Changed partitioning of summary metrics exported by Prometheus.
2159
+ * [CHANGE] Got rid of Gerrit as a review tool.
2160
+ * [CHANGE] 'Tabular' view now the default (rather than 'Graph') to avoid
2161
+ running very expensive queries accidentally.
2162
+ * [CHANGE] On-disk format for stored samples changed. For upgrading, you have
2163
+ to nuke your old files completely. See "Complete rewrite of the storage
2164
+ * [CHANGE] Removed 2nd argument from `delta`.
2165
+ * [FEATURE] Added a `deriv` function.
2166
+ * [FEATURE] Console templates.
2167
+ * [FEATURE] Added `absent` function.
2168
+ * [FEATURE] Allow omitting the metric name in queries.
2169
+ * [BUGFIX] Removed all known race conditions.
2170
+ * [BUGFIX] Metric mutations now handled correctly in all cases.
2171
+ * [ENHANCEMENT] Proper double-start protection.
2172
+ * [ENHANCEMENT] Complete rewrite of the storage layer. Benefits include:
2173
+ * Better query performance.
2174
+ * More samples in less RAM.
2175
+ * Better memory management.
2176
+ * Scales up to millions of time series and thousands of samples ingested
2177
+ per second.
2178
+ * Purging of obsolete samples much cleaner now, up to completely
2179
+ "forgetting" obsolete time series.
2180
+ * Proper instrumentation to diagnose the storage layer with... well...
2181
+ Prometheus.
2182
+ * Pure Go implementation, no need for cgo and shared C libraries anymore.
2183
+ * Better concurrency.
2184
+ * [ENHANCEMENT] Copy-on-write semantics in the AST layer.
2185
+ * [ENHANCEMENT] Switched from Go 1.3 to Go 1.4.
2186
+ * [ENHANCEMENT] Vendored external dependencies with godeps.
2187
+ * [ENHANCEMENT] Numerous Web UI improvements, moved to Bootstrap3 and
2188
+ Rickshaw 1.5.1.
2189
+ * [ENHANCEMENT] Improved Docker integration.
2190
+ * [ENHANCEMENT] Simplified the Makefile contraption.
2191
+ * [CLEANUP] Put meta-data files into proper shape (LICENSE, README.md etc.)
2192
+ * [CLEANUP] Removed all legitimate 'go vet' and 'golint' warnings.
2193
+ * [CLEANUP] Removed dead code.
2194
+
2195
+ ## 0.8.0 / 2014-09-04
2196
+
2197
+ * [ENHANCEMENT] Stagger scrapes to spread out load.
2198
+ * [BUGFIX] Correctly quote HTTP Accept header.
2199
+
2200
+ ## 0.7.0 / 2014-08-06
2201
+
2202
+ * [FEATURE] Added new functions: abs(), topk(), bottomk(), drop_common_labels().
2203
+ * [FEATURE] Let console templates get graph links from expressions.
2204
+ * [FEATURE] Allow console templates to dynamically include other templates.
2205
+ * [FEATURE] Template consoles now have access to their URL.
2206
+ * [BUGFIX] Fixed time() function to return evaluation time, not wallclock time.
2207
+ * [BUGFIX] Fixed HTTP connection leak when targets returned a non-200 status.
2208
+ * [BUGFIX] Fixed link to console templates in UI.
2209
+ * [PERFORMANCE] Removed extra memory copies while scraping targets.
2210
+ * [ENHANCEMENT] Switched from Go 1.2.1 to Go 1.3.
2211
+ * [ENHANCEMENT] Made metrics exported by Prometheus itself more consistent.
2212
+ * [ENHANCEMENT] Removed incremental backoffs for unhealthy targets.
2213
+ * [ENHANCEMENT] Dockerfile also builds Prometheus support tools now.
2214
+
2215
+ ## 0.6.0 / 2014-06-30
2216
+
2217
+ * [FEATURE] Added console and alert templates support, along with various template functions.
2218
+ * [PERFORMANCE] Much faster and more memory-efficient flushing to disk.
2219
+ * [ENHANCEMENT] Query results are now only logged when debugging.
2220
+ * [ENHANCEMENT] Upgraded to new Prometheus client library for exposing metrics.
2221
+ * [BUGFIX] Samples are now kept in memory until fully flushed to disk.
2222
+ * [BUGFIX] Non-200 target scrapes are now treated as an error.
2223
+ * [BUGFIX] Added installation step for missing dependency to Dockerfile.
2224
+ * [BUGFIX] Removed broken and unused "User Dashboard" link.
2225
+
2226
+ ## 0.5.0 / 2014-05-28
2227
+
2228
+ * [BUGFIX] Fixed next retrieval time display on status page.
2229
+ * [BUGFIX] Updated some variable references in tools subdir.
2230
+ * [FEATURE] Added support for scraping metrics via the new text format.
2231
+ * [PERFORMANCE] Improved label matcher performance.
2232
+ * [PERFORMANCE] Removed JSON indentation in query API, leading to smaller response sizes.
2233
+ * [ENHANCEMENT] Added internal check to verify temporal order of streams.
2234
+ * [ENHANCEMENT] Some internal refactorings.
2235
+
2236
+ ## 0.4.0 / 2014-04-17
2237
+
2238
+ * [FEATURE] Vectors and scalars may now be reversed in binary operations (`<scalar> <binop> <vector>`).
2239
+ * [FEATURE] It's possible to shutdown Prometheus via a `/-/quit` web endpoint now.
2240
+ * [BUGFIX] Fix for a deadlock race condition in the memory storage.
2241
+ * [BUGFIX] Mac OS X build fixed.
2242
+ * [BUGFIX] Built from Go 1.2.1, which has internal fixes to race conditions in garbage collection handling.
2243
+ * [ENHANCEMENT] Internal storage interface refactoring that allows building e.g. the `rule_checker` tool without LevelDB dynamic library dependencies.
2244
+ * [ENHANCEMENT] Cleanups around shutdown handling.
2245
+ * [PERFORMANCE] Preparations for better memory reuse during marshaling / unmarshaling.