ominfra 0.0.0.dev90__py3-none-any.whl → 0.0.0.dev91__py3-none-any.whl

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.
@@ -7,7 +7,7 @@
7
7
  """
8
8
  TODO:
9
9
  - create log group
10
- - log stats - chunk sizes etc
10
+ - log stats - chunk sizes, byte count, num calls, etc
11
11
 
12
12
  ==
13
13
 
@@ -311,6 +311,9 @@ def deep_subclasses(cls: ta.Type[T]) -> ta.Iterator[ta.Type[T]]:
311
311
  # ../../../../../omlish/lite/strings.py
312
312
 
313
313
 
314
+ ##
315
+
316
+
314
317
  def camel_case(name: str, lower: bool = False) -> str:
315
318
  if not name:
316
319
  return ''
@@ -325,6 +328,9 @@ def snake_case(name: str) -> str:
325
328
  return '_'.join([name[l:r].lower() for l, r in zip([None, *uppers], [*uppers, None])]).strip('_')
326
329
 
327
330
 
331
+ ##
332
+
333
+
328
334
  def is_dunder(name: str) -> bool:
329
335
  return (
330
336
  name[:2] == name[-2:] == '__' and
@@ -343,10 +349,31 @@ def is_sunder(name: str) -> bool:
343
349
  )
344
350
 
345
351
 
352
+ ##
353
+
354
+
346
355
  def attr_repr(obj: ta.Any, *attrs: str) -> str:
347
356
  return f'{type(obj).__name__}({", ".join(f"{attr}={getattr(obj, attr)!r}" for attr in attrs)})'
348
357
 
349
358
 
359
+ ##
360
+
361
+
362
+ FORMAT_NUM_BYTES_SUFFIXES: ta.Sequence[str] = ['B', 'kB', 'MB', 'GB', 'TB', 'PB', 'EB']
363
+
364
+
365
+ def format_num_bytes(num_bytes: int) -> str:
366
+ for i, suffix in enumerate(FORMAT_NUM_BYTES_SUFFIXES):
367
+ value = num_bytes / 1024 ** i
368
+ if num_bytes < 1024 ** (i + 1):
369
+ if value.is_integer():
370
+ return f'{int(value)}{suffix}'
371
+ else:
372
+ return f'{value:.2f}{suffix}'
373
+
374
+ return f'{num_bytes / 1024 ** (len(FORMAT_NUM_BYTES_SUFFIXES) - 1):.2f}{FORMAT_NUM_BYTES_SUFFIXES[-1]}'
375
+
376
+
350
377
  ########################################
351
378
  # ../../auth.py
352
379
  """
@@ -1421,129 +1448,6 @@ def check_runtime_version() -> None:
1421
1448
  f'Requires python {REQUIRED_PYTHON_VERSION}, got {sys.version_info} from {sys.executable}') # noqa
1422
1449
 
1423
1450
 
1424
- ########################################
1425
- # ../journald/messages.py
1426
-
1427
-
1428
- @dc.dataclass(frozen=True)
1429
- class JournalctlMessage:
1430
- raw: bytes
1431
- dct: ta.Optional[ta.Mapping[str, ta.Any]] = None
1432
- cursor: ta.Optional[str] = None
1433
- ts_us: ta.Optional[int] = None # microseconds UTC
1434
-
1435
-
1436
- class JournalctlMessageBuilder:
1437
- def __init__(self) -> None:
1438
- super().__init__()
1439
-
1440
- self._buf = DelimitingBuffer(b'\n')
1441
-
1442
- _cursor_field = '__CURSOR'
1443
- _timestamp_field = '_SOURCE_REALTIME_TIMESTAMP'
1444
-
1445
- def _make_message(self, raw: bytes) -> JournalctlMessage:
1446
- dct = None
1447
- cursor = None
1448
- ts = None
1449
-
1450
- try:
1451
- dct = json.loads(raw.decode('utf-8', 'replace'))
1452
- except Exception: # noqa
1453
- log.exception('Failed to parse raw message: %r', raw)
1454
-
1455
- else:
1456
- cursor = dct.get(self._cursor_field)
1457
-
1458
- if tsv := dct.get(self._timestamp_field):
1459
- if isinstance(tsv, str):
1460
- try:
1461
- ts = int(tsv)
1462
- except ValueError:
1463
- try:
1464
- ts = int(float(tsv))
1465
- except ValueError:
1466
- log.exception('Failed to parse timestamp: %r', tsv)
1467
- elif isinstance(tsv, (int, float)):
1468
- ts = int(tsv)
1469
- else:
1470
- log.exception('Invalid timestamp: %r', tsv)
1471
-
1472
- return JournalctlMessage(
1473
- raw=raw,
1474
- dct=dct,
1475
- cursor=cursor,
1476
- ts_us=ts,
1477
- )
1478
-
1479
- def feed(self, data: bytes) -> ta.Sequence[JournalctlMessage]:
1480
- ret: ta.List[JournalctlMessage] = []
1481
- for line in self._buf.feed(data):
1482
- ret.append(self._make_message(check_isinstance(line, bytes))) # type: ignore
1483
- return ret
1484
-
1485
-
1486
- ########################################
1487
- # ../threadworker.py
1488
-
1489
-
1490
- class ThreadWorker(abc.ABC):
1491
- def __init__(
1492
- self,
1493
- *,
1494
- stop_event: ta.Optional[threading.Event] = None,
1495
- ) -> None:
1496
- super().__init__()
1497
-
1498
- if stop_event is None:
1499
- stop_event = threading.Event()
1500
- self._stop_event = stop_event
1501
-
1502
- self._thread: ta.Optional[threading.Thread] = None
1503
-
1504
- self._last_heartbeat: ta.Optional[float] = None
1505
-
1506
- #
1507
-
1508
- def should_stop(self) -> bool:
1509
- return self._stop_event.is_set()
1510
-
1511
- #
1512
-
1513
- @property
1514
- def last_heartbeat(self) -> ta.Optional[float]:
1515
- return self._last_heartbeat
1516
-
1517
- def _heartbeat(self) -> bool:
1518
- self._last_heartbeat = time.time()
1519
-
1520
- if self.should_stop():
1521
- log.info('Stopping: %s', self)
1522
- return False
1523
-
1524
- return True
1525
-
1526
- #
1527
-
1528
- def is_alive(self) -> bool:
1529
- return (thr := self._thread) is not None and thr.is_alive()
1530
-
1531
- def start(self) -> None:
1532
- thr = threading.Thread(target=self._run)
1533
- self._thread = thr
1534
- thr.start()
1535
-
1536
- @abc.abstractmethod
1537
- def _run(self) -> None:
1538
- raise NotImplementedError
1539
-
1540
- def stop(self) -> None:
1541
- raise NotImplementedError
1542
-
1543
- def cleanup(self) -> None: # noqa
1544
- pass
1545
-
1546
-
1547
1451
  ########################################
1548
1452
  # ../../logs.py
1549
1453
  """
@@ -1604,7 +1508,7 @@ class AwsLogMessagePoster:
1604
1508
  - max_items
1605
1509
  - max_bytes - manually build body
1606
1510
  - flush_interval
1607
- - !! sort by timestamp
1511
+ - split sorted chunks if span over 24h
1608
1512
  """
1609
1513
 
1610
1514
  DEFAULT_URL = 'https://logs.{region_name}.amazonaws.com/' # noqa
@@ -1677,7 +1581,7 @@ class AwsLogMessagePoster:
1677
1581
  message=m.message,
1678
1582
  timestamp=m.ts_ms,
1679
1583
  )
1680
- for m in messages
1584
+ for m in sorted(messages, key=lambda m: m.ts_ms)
1681
1585
  ],
1682
1586
  )
1683
1587
 
@@ -1709,6 +1613,142 @@ class AwsLogMessagePoster:
1709
1613
  return [post]
1710
1614
 
1711
1615
 
1616
+ ########################################
1617
+ # ../../../../journald/messages.py
1618
+
1619
+
1620
+ @dc.dataclass(frozen=True)
1621
+ class JournalctlMessage:
1622
+ raw: bytes
1623
+ dct: ta.Optional[ta.Mapping[str, ta.Any]] = None
1624
+ cursor: ta.Optional[str] = None
1625
+ ts_us: ta.Optional[int] = None # microseconds UTC
1626
+
1627
+
1628
+ class JournalctlMessageBuilder:
1629
+ def __init__(self) -> None:
1630
+ super().__init__()
1631
+
1632
+ self._buf = DelimitingBuffer(b'\n')
1633
+
1634
+ _cursor_field = '__CURSOR'
1635
+
1636
+ _timestamp_fields: ta.Sequence[str] = [
1637
+ '_SOURCE_REALTIME_TIMESTAMP',
1638
+ '__REALTIME_TIMESTAMP',
1639
+ ]
1640
+
1641
+ def _get_message_timestamp(self, dct: ta.Mapping[str, ta.Any]) -> ta.Optional[int]:
1642
+ for fld in self._timestamp_fields:
1643
+ if (tsv := dct.get(fld)) is None:
1644
+ continue
1645
+
1646
+ if isinstance(tsv, str):
1647
+ try:
1648
+ return int(tsv)
1649
+ except ValueError:
1650
+ try:
1651
+ return int(float(tsv))
1652
+ except ValueError:
1653
+ log.exception('Failed to parse timestamp: %r', tsv)
1654
+
1655
+ elif isinstance(tsv, (int, float)):
1656
+ return int(tsv)
1657
+
1658
+ log.error('Invalid timestamp: %r', dct)
1659
+ return None
1660
+
1661
+ def _make_message(self, raw: bytes) -> JournalctlMessage:
1662
+ dct = None
1663
+ cursor = None
1664
+ ts = None
1665
+
1666
+ try:
1667
+ dct = json.loads(raw.decode('utf-8', 'replace'))
1668
+ except Exception: # noqa
1669
+ log.exception('Failed to parse raw message: %r', raw)
1670
+
1671
+ else:
1672
+ cursor = dct.get(self._cursor_field)
1673
+ ts = self._get_message_timestamp(dct)
1674
+
1675
+ return JournalctlMessage(
1676
+ raw=raw,
1677
+ dct=dct,
1678
+ cursor=cursor,
1679
+ ts_us=ts,
1680
+ )
1681
+
1682
+ def feed(self, data: bytes) -> ta.Sequence[JournalctlMessage]:
1683
+ ret: ta.List[JournalctlMessage] = []
1684
+ for line in self._buf.feed(data):
1685
+ ret.append(self._make_message(check_isinstance(line, bytes))) # type: ignore
1686
+ return ret
1687
+
1688
+
1689
+ ########################################
1690
+ # ../../../../threadworker.py
1691
+ """
1692
+ TODO:
1693
+ - implement stop lol
1694
+ - collective heartbeat monitoring - ThreadWorkerGroups
1695
+ """
1696
+
1697
+
1698
+ class ThreadWorker(abc.ABC):
1699
+ def __init__(
1700
+ self,
1701
+ *,
1702
+ stop_event: ta.Optional[threading.Event] = None,
1703
+ ) -> None:
1704
+ super().__init__()
1705
+
1706
+ if stop_event is None:
1707
+ stop_event = threading.Event()
1708
+ self._stop_event = stop_event
1709
+
1710
+ self._thread: ta.Optional[threading.Thread] = None
1711
+
1712
+ self._last_heartbeat: ta.Optional[float] = None
1713
+
1714
+ #
1715
+
1716
+ def should_stop(self) -> bool:
1717
+ return self._stop_event.is_set()
1718
+
1719
+ #
1720
+
1721
+ @property
1722
+ def last_heartbeat(self) -> ta.Optional[float]:
1723
+ return self._last_heartbeat
1724
+
1725
+ def _heartbeat(self) -> bool:
1726
+ self._last_heartbeat = time.time()
1727
+
1728
+ if self.should_stop():
1729
+ log.info('Stopping: %s', self)
1730
+ return False
1731
+
1732
+ return True
1733
+
1734
+ #
1735
+
1736
+ def is_alive(self) -> bool:
1737
+ return (thr := self._thread) is not None and thr.is_alive()
1738
+
1739
+ def start(self) -> None:
1740
+ thr = threading.Thread(target=self._run)
1741
+ self._thread = thr
1742
+ thr.start()
1743
+
1744
+ @abc.abstractmethod
1745
+ def _run(self) -> None:
1746
+ raise NotImplementedError
1747
+
1748
+ def stop(self) -> None:
1749
+ raise NotImplementedError
1750
+
1751
+
1712
1752
  ########################################
1713
1753
  # ../../../../../omlish/lite/subprocesses.py
1714
1754
 
@@ -1816,7 +1856,344 @@ def subprocess_try_output_str(*args: str, **kwargs: ta.Any) -> ta.Optional[str]:
1816
1856
 
1817
1857
 
1818
1858
  ########################################
1819
- # ../journald/tailer.py
1859
+ # ../../../../journald/tailer.py
1860
+ """
1861
+ TODO:
1862
+ - https://www.rootusers.com/how-to-change-log-rate-limiting-in-linux/
1863
+
1864
+ ==
1865
+
1866
+ https://www.freedesktop.org/software/systemd/man/latest/journalctl.html
1867
+
1868
+ Source Options
1869
+ --system, --user :: Show messages from system services and the kernel (with --system). Show messages from service of
1870
+ current user (with --user). If neither is specified, show all messages that the user can see. The
1871
+ --user option affects how --unit= arguments are treated. See --unit=. Note that --user only works if
1872
+ persistent logging is enabled, via the Storage= setting in journald.conf(5).
1873
+ -M, --machine= :: Show messages from a running, local container. Specify a container name to connect to.
1874
+ -m, --merge :: Show entries interleaved from all available journals, including remote ones.
1875
+ -D DIR, --directory=DIR :: Takes a directory path as argument. If specified, journalctl will operate on the specified
1876
+ journal directory DIR instead of the default runtime and system journal paths.
1877
+ -i GLOB, --file=GLOB :: Takes a file glob as an argument. If specified, journalctl will operate on the specified journal
1878
+ files matching GLOB instead of the default runtime and system journal paths. May be specified
1879
+ multiple times, in which case files will be suitably interleaved.
1880
+ --root=ROOT :: Takes a directory path as an argument. If specified, journalctl will operate on journal directories and
1881
+ catalog file hierarchy underneath the specified directory instead of the root directory (e.g.
1882
+ --update-catalog will create ROOT/var/lib/systemd/catalog/database, and journal files under
1883
+ ROOT/run/journal/ or ROOT/var/log/journal/ will be displayed).
1884
+ --image=IMAGE :: Takes a path to a disk image file or block device node. If specified, journalctl will operate on the
1885
+ file system in the indicated disk image. This option is similar to --root=, but operates on file
1886
+ systems stored in disk images or block devices, thus providing an easy way to extract log data from
1887
+ disk images. The disk image should either contain just a file system or a set of file systems within a
1888
+ GPT partition table, following the Discoverable Partitions Specification. For further information on
1889
+ supported disk images, see systemd-nspawn(1)'s switch of the same name.
1890
+ --image-policy=policy :: Takes an image policy string as argument, as per systemd.image-policy(7). The policy is
1891
+ enforced when operating on the disk image specified via --image=, see above. If not specified
1892
+ defaults to the "*" policy, i.e. all recognized file systems in the image are used.
1893
+ --namespace=NAMESPACE :: Takes a journal namespace identifier string as argument. If not specified the data collected by
1894
+ the default namespace is shown. If specified shows the log data of the specified namespace
1895
+ instead. If the namespace is specified as "*" data from all namespaces is shown, interleaved.
1896
+ If the namespace identifier is prefixed with "+" data from the specified namespace and the
1897
+ default namespace is shown, interleaved, but no other. For details about journal namespaces see
1898
+ systemd-journald.service(8).
1899
+
1900
+ Filtering Options
1901
+ -S, --since=, -U, --until= :: Start showing entries on or newer than the specified date, or on or older than the
1902
+ specified date, respectively. Date specifications should be of the format
1903
+ "2012-10-30 18:17:16". If the time part is omitted, "00:00:00" is assumed. If only the
1904
+ seconds component is omitted, ":00" is assumed. If the date component is omitted, the
1905
+ current day is assumed. Alternatively the strings "yesterday", "today", "tomorrow" are
1906
+ understood, which refer to 00:00:00 of the day before the current day, the current day, or
1907
+ the day after the current day, respectively. "now" refers to the current time. Finally,
1908
+ relative times may be specified, prefixed with "-" or "+", referring to times before or
1909
+ after the current time, respectively. For complete time and date specification, see
1910
+ systemd.time(7). Note that --output=short-full prints timestamps that follow precisely
1911
+ this format.
1912
+ -c, --cursor= :: Start showing entries from the location in the journal specified by the passed cursor.
1913
+ --after-cursor= :: Start showing entries from the location in the journal after the location specified by the passed
1914
+ cursor. The cursor is shown when the --show-cursor option is used.
1915
+ --cursor-file=FILE :: If FILE exists and contains a cursor, start showing entries after this location. Otherwise show
1916
+ entries according to the other given options. At the end, write the cursor of the last entry to
1917
+ FILE. Use this option to continually read the journal by sequentially calling journalctl.
1918
+ -b [[ID][±offset]|all], --boot[=[ID][±offset]|all] :: Show messages from a specific boot. This will add a match for
1919
+ "_BOOT_ID=". The argument may be empty, in which case logs for the
1920
+ current boot will be shown. If the boot ID is omitted, a positive
1921
+ offset will look up the boots starting from the beginning of the
1922
+ journal, and an equal-or-less-than zero offset will look up boots
1923
+ starting from the end of the journal. Thus, 1 means the first boot
1924
+ found in the journal in chronological order, 2 the second and so
1925
+ on; while -0 is the last boot, -1 the boot before last, and so on.
1926
+ An empty offset is equivalent to specifying -0, except when the
1927
+ current boot is not the last boot (e.g. because --directory= was
1928
+ specified to look at logs from a different machine). If the
1929
+ 32-character ID is specified, it may optionally be followed by
1930
+ offset which identifies the boot relative to the one given by boot
1931
+ ID. Negative values mean earlier boots and positive values mean
1932
+ later boots. If offset is not specified, a value of zero is
1933
+ assumed, and the logs for the boot given by ID are shown. The
1934
+ special argument all can be used to negate the effect of an
1935
+ earlier use of -b.
1936
+ -u, --unit=UNIT|PATTERN :: Show messages for the specified systemd unit UNIT (such as a service unit), or for any of the
1937
+ units matched by PATTERN. If a pattern is specified, a list of unit names found in the
1938
+ journal is compared with the specified pattern and all that match are used. For each unit
1939
+ name, a match is added for messages from the unit ("_SYSTEMD_UNIT=UNIT"), along with
1940
+ additional matches for messages from systemd and messages about coredumps for the specified
1941
+ unit. A match is also added for "_SYSTEMD_SLICE=UNIT", such that if the provided UNIT is a
1942
+ systemd.slice(5) unit, all logs of children of the slice will be shown. With --user, all
1943
+ --unit= arguments will be converted to match user messages as if specified with --user-unit=.
1944
+ This parameter can be specified multiple times.
1945
+ --user-unit= :: Show messages for the specified user session unit. This will add a match for messages from the unit
1946
+ ("_SYSTEMD_USER_UNIT=" and "_UID=") and additional matches for messages from session systemd and
1947
+ messages about coredumps for the specified unit. A match is also added for "_SYSTEMD_USER_SLICE=UNIT",
1948
+ such that if the provided UNIT is a systemd.slice(5) unit, all logs of children of the unit will be
1949
+ shown. This parameter can be specified multiple times.
1950
+ -t, --identifier=SYSLOG_IDENTIFIER :: Show messages for the specified syslog identifier SYSLOG_IDENTIFIER. This
1951
+ parameter can be specified multiple times.
1952
+ -T, --exclude-identifier=SYSLOG_IDENTIFIER :: Exclude messages for the specified syslog identifier SYSLOG_IDENTIFIER.
1953
+ This parameter can be specified multiple times.
1954
+ -p, --priority= :: Filter output by message priorities or priority ranges. Takes either a single numeric or textual log
1955
+ level (i.e. between 0/"emerg" and 7/"debug"), or a range of numeric/text log levels in the form
1956
+ FROM..TO. The log levels are the usual syslog log levels as documented in syslog(3), i.e. "emerg"
1957
+ (0), "alert" (1), "crit" (2), "err" (3), "warning" (4), "notice" (5), "info" (6), "debug" (7). If a
1958
+ single log level is specified, all messages with this log level or a lower (hence more important) log
1959
+ level are shown. If a range is specified, all messages within the range are shown, including both the
1960
+ start and the end value of the range. This will add "PRIORITY=" matches for the specified priorities.
1961
+ --facility= :: Filter output by syslog facility. Takes a comma-separated list of numbers or facility names. The names
1962
+ are the usual syslog facilities as documented in syslog(3). --facility=help may be used to display a list
1963
+ of known facility names and exit.
1964
+ -g, --grep= :: Filter output to entries where the MESSAGE= field matches the specified regular expression.
1965
+ PERL-compatible regular expressions are used, see pcre2pattern(3) for a detailed description of the
1966
+ syntax. If the pattern is all lowercase, matching is case insensitive. Otherwise, matching is case
1967
+ sensitive. This can be overridden with the --case-sensitive option, see below. When used with --lines=
1968
+ (not prefixed with "+"), --reverse is implied.
1969
+ --case-sensitive[=BOOLEAN] :: Make pattern matching case sensitive or case insensitive.
1970
+ -k, --dmesg :: Show only kernel messages. This implies -b and adds the match "_TRANSPORT=kernel".
1971
+
1972
+ Output Options
1973
+ -o, --output= :: Controls the formatting of the journal entries that are shown. Takes one of the following options:
1974
+ short :: is the default and generates an output that is mostly identical to the formatting of classic syslog files,
1975
+ showing one line per journal entry.
1976
+ short-full :: is very similar, but shows timestamps in the format the --since= and --until= options accept. Unlike the
1977
+ timestamp information shown in short output mode this mode includes weekday, year and timezone
1978
+ information in the output, and is locale-independent.
1979
+ short-iso :: is very similar, but shows timestamps in the RFC 3339 profile of ISO 8601.
1980
+ short-iso-precise :: as for short-iso but includes full microsecond precision.
1981
+ short-precise :: is very similar, but shows classic syslog timestamps with full microsecond precision.
1982
+ short-monotonic :: is very similar, but shows monotonic timestamps instead of wallclock timestamps.
1983
+ short-delta :: as for short-monotonic but includes the time difference to the previous entry. Maybe unreliable time
1984
+ differences are marked by a "*".
1985
+ short-unix :: is very similar, but shows seconds passed since January 1st 1970 UTC instead of wallclock timestamps
1986
+ ("UNIX time"). The time is shown with microsecond accuracy.
1987
+ verbose :: shows the full-structured entry items with all fields.
1988
+ export :: serializes the journal into a binary (but mostly text-based) stream suitable for backups and network
1989
+ transfer (see Journal Export Format for more information). To import the binary stream back into native
1990
+ journald format use systemd-journal-remote(8).
1991
+ json :: formats entries as JSON objects, separated by newline characters (see Journal JSON Format for more
1992
+ information). Field values are generally encoded as JSON strings, with three exceptions: Fields larger than
1993
+ 4096 bytes are encoded as null values. (This may be turned off by passing --all, but be aware that this may
1994
+ allocate overly long JSON objects.) Journal entries permit non-unique fields within the same log entry. JSON
1995
+ does not allow non-unique fields within objects. Due to this, if a non-unique field is encountered a JSON
1996
+ array is used as field value, listing all field values as elements. Fields containing non-printable or
1997
+ non-UTF8 bytes are encoded as arrays containing the raw bytes individually formatted as unsigned numbers. Note
1998
+ that this encoding is reversible (with the exception of the size limit).
1999
+ json-pretty :: formats entries as JSON data structures, but formats them in multiple lines in order to make them more
2000
+ readable by humans.
2001
+ json-sse :: formats entries as JSON data structures, but wraps them in a format suitable for Server-Sent Events.
2002
+ json-seq :: formats entries as JSON data structures, but prefixes them with an ASCII Record Separator character (0x1E)
2003
+ and suffixes them with an ASCII Line Feed character (0x0A), in accordance with JavaScript Object Notation
2004
+ (JSON) Text Sequences ("application/json-seq").
2005
+ cat :: generates a very terse output, only showing the actual message of each journal entry with no metadata, not even
2006
+ a timestamp. If combined with the --output-fields= option will output the listed fields for each log record,
2007
+ instead of the message.
2008
+ with-unit :: similar to short-full, but prefixes the unit and user unit names instead of the traditional syslog
2009
+ identifier. Useful when using templated instances, as it will include the arguments in the unit names.
2010
+ --truncate-newline :: Truncate each log message at the first newline character on output, so that only the first line of
2011
+ each message is displayed.
2012
+ --output-fields= :: A comma separated list of the fields which should be included in the output. This has an effect only
2013
+ for the output modes which would normally show all fields (verbose, export, json, json-pretty,
2014
+ json-sse and json-seq), as well as on cat. For the former, the "__CURSOR", "__REALTIME_TIMESTAMP",
2015
+ "__MONOTONIC_TIMESTAMP", and "_BOOT_ID" fields are always printed.
2016
+ -n, --lines= :: Show the most recent journal events and limit the number of events shown. The argument is a positive
2017
+ integer or "all" to disable the limit. Additionally, if the number is prefixed with "+", the oldest
2018
+ journal events are used instead. The default value is 10 if no argument is given. If --follow is used,
2019
+ this option is implied. When not prefixed with "+" and used with --grep=, --reverse is implied.
2020
+ -r, --reverse :: Reverse output so that the newest entries are displayed first.
2021
+ --show-cursor :: The cursor is shown after the last entry after two dashes:
2022
+ -- cursor: s=0639… :: The format of the cursor is private and subject to change.
2023
+ --utc :: Express time in Coordinated Universal Time (UTC).
2024
+ -x, --catalog :: Augment log lines with explanation texts from the message catalog. This will add explanatory help texts
2025
+ to log messages in the output where this is available. These short help texts will explain the context
2026
+ of an error or log event, possible solutions, as well as pointers to support forums, developer
2027
+ documentation, and any other relevant manuals. Note that help texts are not available for all messages,
2028
+ but only for selected ones. For more information on the message catalog, see Journal Message Catalogs.
2029
+ Note: when attaching journalctl output to bug reports, please do not use -x.
2030
+ --no-hostname :: Don't show the hostname field of log messages originating from the local host. This switch has an
2031
+ effect only on the short family of output modes (see above). Note: this option does not remove
2032
+ occurrences of the hostname from log entries themselves, so it does not prevent the hostname from being
2033
+ visible in the logs.
2034
+ --no-full, --full, -l :: Ellipsize fields when they do not fit in available columns. The default is to show full fields,
2035
+ allowing them to wrap or be truncated by the pager, if one is used. The old options -l/--full
2036
+ are not useful anymore, except to undo --no-full.
2037
+ -a, --all :: Show all fields in full, even if they include unprintable characters or are very long. By default, fields
2038
+ with unprintable characters are abbreviated as "blob data". (Note that the pager may escape unprintable
2039
+ characters again.)
2040
+ -f, --follow :: Show only the most recent journal entries, and continuously print new entries as they are appended to
2041
+ the journal.
2042
+ --no-tail :: Show all stored output lines, even in follow mode. Undoes the effect of --lines=.
2043
+ -q, --quiet :: Suppresses all informational messages (i.e. "-- Journal begins at …", "-- Reboot --"), any warning
2044
+ messages regarding inaccessible system journals when run as a normal user.
2045
+
2046
+ Pager Control Options
2047
+ --no-pager :: Do not pipe output into a pager.
2048
+ -e, --pager-end :: Immediately jump to the end of the journal inside the implied pager tool. This implies -n1000 to
2049
+ guarantee that the pager will not buffer logs of unbounded size. This may be overridden with an
2050
+ explicit -n with some other numeric value, while -nall will disable this cap. Note that this option
2051
+ is only supported for the less(1) pager.
2052
+
2053
+ Forward Secure Sealing (FSS) Options
2054
+ --interval= :: Specifies the change interval for the sealing key when generating an FSS key pair with --setup-keys.
2055
+ Shorter intervals increase CPU consumption but shorten the time range of undetectable journal
2056
+ alterations. Defaults to 15min.
2057
+ --verify-key= :: Specifies the FSS verification key to use for the --verify operation.
2058
+ --force :: When --setup-keys is passed and Forward Secure Sealing (FSS) has already been configured, recreate FSS keys.
2059
+
2060
+ Commands
2061
+ -N, --fields :: Print all field names currently used in all entries of the journal.
2062
+ -F, --field= :: Print all possible data values the specified field can take in all entries of the journal.
2063
+ --list-boots :: Show a tabular list of boot numbers (relative to the current boot), their IDs, and the timestamps of the
2064
+ first and last message pertaining to the boot. When specified with -n/--lines=[+]N option, only the
2065
+ first (when the number prefixed with "+") or the last (without prefix) N entries will be shown. When
2066
+ specified with -r/--reverse, the list will be shown in the reverse order.
2067
+ --disk-usage :: Shows the current disk usage of all journal files. This shows the sum of the disk usage of all archived
2068
+ and active journal files.
2069
+ --vacuum-size=, --vacuum-time=, --vacuum-files=
2070
+ --vacuum-size= :: removes the oldest archived journal files until the disk space they use falls below the specified
2071
+ size. Accepts the usual "K", "M", "G" and "T" suffixes (to the base of 1024).
2072
+ --vacuum-time= :: removes archived journal files older than the specified timespan. Accepts the usual "s" (default),
2073
+ "m", "h", "days", "weeks", "months", and "years" suffixes, see systemd.time(7) for details.
2074
+ --vacuum-files= :: leaves only the specified number of separate journal files.
2075
+ Note that running --vacuum-size= has only an indirect effect on the output shown by --disk-usage, as the latter
2076
+ includes active journal files, while the vacuuming operation only operates on archived journal files. Similarly,
2077
+ --vacuum-files= might not actually reduce the number of journal files to below the specified number, as it will not
2078
+ remove active journal files.
2079
+ --vacuum-size=, --vacuum-time= and --vacuum-files= may be combined in a single invocation to enforce any combination
2080
+ of a size, a time and a number of files limit on the archived journal files. Specifying any of these three parameters
2081
+ as zero is equivalent to not enforcing the specific limit, and is thus redundant.
2082
+ These three switches may also be combined with --rotate into one command. If so, all active files are rotated first,
2083
+ and the requested vacuuming operation is executed right after. The rotation has the effect that all currently active
2084
+ files are archived (and potentially new, empty journal files opened as replacement), and hence the vacuuming operation
2085
+ has the greatest effect as it can take all log data written so far into account.
2086
+ --verify :: Check the journal file for internal consistency. If the file has been generated with FSS enabled and the FSS
2087
+ verification key has been specified with --verify-key=, authenticity of the journal file is verified.
2088
+ --sync :: Asks the journal daemon to write all yet unwritten journal data to the backing file system and synchronize all
2089
+ journals. This call does not return until the synchronization operation is complete. This command guarantees
2090
+ that any log messages written before its invocation are safely stored on disk at the time it returns.
2091
+ --relinquish-var :: Asks the journal daemon for the reverse operation to --flush: if requested the daemon will write
2092
+ further log data to /run/log/journal/ and stops writing to /var/log/journal/. A subsequent call to
2093
+ --flush causes the log output to switch back to /var/log/journal/, see above.
2094
+ --smart-relinquish-var :: Similar to --relinquish-var, but executes no operation if the root file system and
2095
+ /var/log/journal/ reside on the same mount point. This operation is used during system
2096
+ shutdown in order to make the journal daemon stop writing data to /var/log/journal/ in case
2097
+ that directory is located on a mount point that needs to be unmounted.
2098
+ --flush :: Asks the journal daemon to flush any log data stored in /run/log/journal/ into /var/log/journal/, if
2099
+ persistent storage is enabled. This call does not return until the operation is complete. Note that this call
2100
+ is idempotent: the data is only flushed from /run/log/journal/ into /var/log/journal/ once during system
2101
+ runtime (but see --relinquish-var below), and this command exits cleanly without executing any operation if
2102
+ this has already happened. This command effectively guarantees that all data is flushed to /var/log/journal/
2103
+ at the time it returns.
2104
+ --rotate :: Asks the journal daemon to rotate journal files. This call does not return until the rotation operation is
2105
+ complete. Journal file rotation has the effect that all currently active journal files are marked as
2106
+ archived and renamed, so that they are never written to in future. New (empty) journal files are then
2107
+ created in their place. This operation may be combined with --vacuum-size=, --vacuum-time= and
2108
+ --vacuum-file= into a single command, see above.
2109
+ --header :: Instead of showing journal contents, show internal header information of the journal fields accessed. This
2110
+ option is particularly useful when trying to identify out-of-order journal entries, as happens for example
2111
+ when the machine is booted with the wrong system time.
2112
+ --list-catalog [128-bit-ID…] :: List the contents of the message catalog as a table of message IDs, plus their short
2113
+ description strings. If any 128-bit-IDs are specified, only those entries are shown.
2114
+ --dump-catalog [128-bit-ID…] :: Show the contents of the message catalog, with entries separated by a line consisting of
2115
+ two dashes and the ID (the format is the same as .catalog files). If any 128-bit-IDs are
2116
+ specified, only those entries are shown.
2117
+ --update-catalog :: Update the message catalog index. This command needs to be executed each time new catalog files are
2118
+ installed, removed, or updated to rebuild the binary catalog index.
2119
+ --setup-keys :: Instead of showing journal contents, generate a new key pair for Forward Secure Sealing (FSS). This will
2120
+ generate a sealing key and a verification key. The sealing key is stored in the journal data directory
2121
+ and shall remain on the host. The verification key should be stored externally. Refer to the Seal=
2122
+ option in journald.conf(5) for information on Forward Secure Sealing and for a link to a refereed
2123
+ scholarly paper detailing the cryptographic theory it is based on.
2124
+ -h, --help :: Print a short help text and exit.
2125
+ --version :: Print a short version string and exit.
2126
+
2127
+ Environment
2128
+ $SYSTEMD_LOG_LEVEL :: The maximum log level of emitted messages (messages with a higher log level, i.e. less important
2129
+ ones, will be suppressed). Takes a comma-separated list of values. A value may be either one of
2130
+ (in order of decreasing importance) emerg, alert, crit, err, warning, notice, info, debug, or an
2131
+ integer in the range 0…7. See syslog(3) for more information. Each value may optionally be
2132
+ prefixed with one of console, syslog, kmsg or journal followed by a colon to set the maximum log
2133
+ level for that specific log target (e.g. SYSTEMD_LOG_LEVEL=debug,console:info specifies to log at
2134
+ debug level except when logging to the console which should be at info level). Note that the
2135
+ global maximum log level takes priority over any per target maximum log levels.
2136
+ $SYSTEMD_LOG_COLOR :: A boolean. If true, messages written to the tty will be colored according to priority. This
2137
+ setting is only useful when messages are written directly to the terminal, because journalctl(1)
2138
+ and other tools that display logs will color messages based on the log level on their own.
2139
+ $SYSTEMD_LOG_TIME :: A boolean. If true, console log messages will be prefixed with a timestamp. This setting is only
2140
+ useful when messages are written directly to the terminal or a file, because journalctl(1) and
2141
+ other tools that display logs will attach timestamps based on the entry metadata on their own.
2142
+ $SYSTEMD_LOG_LOCATION :: A boolean. If true, messages will be prefixed with a filename and line number in the source
2143
+ code where the message originates. Note that the log location is often attached as metadata to
2144
+ journal entries anyway. Including it directly in the message text can nevertheless be
2145
+ convenient when debugging programs.
2146
+ $SYSTEMD_LOG_TID :: A boolean. If true, messages will be prefixed with the current numerical thread ID (TID). Note that
2147
+ the this information is attached as metadata to journal entries anyway. Including it directly in the
2148
+ message text can nevertheless be convenient when debugging programs.
2149
+ $SYSTEMD_LOG_TARGET :: The destination for log messages. One of console (log to the attached tty), console-prefixed (log
2150
+ to the attached tty but with prefixes encoding the log level and "facility", see syslog(3), kmsg
2151
+ (log to the kernel circular log buffer), journal (log to the journal), journal-or-kmsg (log to
2152
+ the journal if available, and to kmsg otherwise), auto (determine the appropriate log target
2153
+ automatically, the default), null (disable log output).
2154
+ $SYSTEMD_LOG_RATELIMIT_KMSG :: Whether to ratelimit kmsg or not. Takes a boolean. Defaults to "true". If disabled,
2155
+ systemd will not ratelimit messages written to kmsg.
2156
+ $SYSTEMD_PAGER :: Pager to use when --no-pager is not given; overrides $PAGER. If neither $SYSTEMD_PAGER nor $PAGER are
2157
+ set, a set of well-known pager implementations are tried in turn, including less(1) and more(1), until
2158
+ one is found. If no pager implementation is discovered no pager is invoked. Setting this environment
2159
+ variable to an empty string or the value "cat" is equivalent to passing --no-pager. Note: if
2160
+ $SYSTEMD_PAGERSECURE is not set, $SYSTEMD_PAGER (as well as $PAGER) will be silently ignored.
2161
+ $SYSTEMD_LESS :: Override the options passed to less (by default "FRSXMK"). Users might want to change two options in
2162
+ particular:
2163
+ K :: This option instructs the pager to exit immediately when Ctrl+C is pressed. To allow less to handle Ctrl+C itself
2164
+ to switch back to the pager command prompt, unset this option. If the value of $SYSTEMD_LESS does not include
2165
+ "K", and the pager that is invoked is less, Ctrl+C will be ignored by the executable, and needs to be handled by
2166
+ the pager.
2167
+ X :: This option instructs the pager to not send termcap initialization and deinitialization strings to the terminal.
2168
+ It is set by default to allow command output to remain visible in the terminal even after the pager exits.
2169
+ Nevertheless, this prevents some pager functionality from working, in particular paged output cannot be scrolled
2170
+ with the mouse. Note that setting the regular $LESS environment variable has no effect for less invocations by
2171
+ systemd tools.
2172
+ $SYSTEMD_LESSCHARSET :: Override the charset passed to less (by default "utf-8", if the invoking terminal is determined
2173
+ to be UTF-8 compatible). Note that setting the regular $LESSCHARSET environment variable has no
2174
+ effect for less invocations by systemd tools.
2175
+ $SYSTEMD_PAGERSECURE :: Takes a boolean argument. When true, the "secure" mode of the pager is enabled; if false,
2176
+ disabled. If $SYSTEMD_PAGERSECURE is not set at all, secure mode is enabled if the effective UID
2177
+ is not the same as the owner of the login session, see geteuid(2) and sd_pid_get_owner_uid(3).
2178
+ In secure mode, LESSSECURE=1 will be set when invoking the pager, and the pager shall disable
2179
+ commands that open or create new files or start new subprocesses. When $SYSTEMD_PAGERSECURE is
2180
+ not set at all, pagers which are not known to implement secure mode will not be used. (Currently
2181
+ only less(1) implements secure mode.) Note: when commands are invoked with elevated privileges,
2182
+ for example under sudo(8) or pkexec(1), care must be taken to ensure that unintended interactive
2183
+ features are not enabled. "Secure" mode for the pager may be enabled automatically as describe
2184
+ above. Setting SYSTEMD_PAGERSECURE=0 or not removing it from the inherited environment allows
2185
+ the user to invoke arbitrary commands. Note that if the $SYSTEMD_PAGER or $PAGER variables are
2186
+ to be honoured, $SYSTEMD_PAGERSECURE must be set too. It might be reasonable to completely
2187
+ disable the pager using --no-pager instead.
2188
+ $SYSTEMD_COLORS :: Takes a boolean argument. When true, systemd and related utilities will use colors in their output,
2189
+ otherwise the output will be monochrome. Additionally, the variable can take one of the following
2190
+ special values: "16", "256" to restrict the use of colors to the base 16 or 256 ANSI colors,
2191
+ respectively. This can be specified to override the automatic decision based on $TERM and what the
2192
+ console is connected to.
2193
+ $SYSTEMD_URLIFY :: The value must be a boolean. Controls whether clickable links should be generated in the output for
2194
+ terminal emulators supporting this. This can be specified to override the decision that systemd makes
2195
+ based on $TERM and other conditions.
2196
+ """
1820
2197
 
1821
2198
 
1822
2199
  class JournalctlTailerWorker(ThreadWorker):
@@ -1887,11 +2264,11 @@ class JournalctlTailerWorker(ThreadWorker):
1887
2264
 
1888
2265
  while True:
1889
2266
  if not self._heartbeat():
1890
- break
2267
+ return
1891
2268
 
1892
2269
  while stdout.readable():
1893
2270
  if not self._heartbeat():
1894
- break
2271
+ return
1895
2272
 
1896
2273
  buf = stdout.read(self._read_size)
1897
2274
  if not buf:
@@ -1901,11 +2278,18 @@ class JournalctlTailerWorker(ThreadWorker):
1901
2278
  log.debug('Journalctl read buffer: %r', buf)
1902
2279
  msgs = self._mb.feed(buf)
1903
2280
  if msgs:
1904
- self._output.put(msgs)
2281
+ while True:
2282
+ try:
2283
+ self._output.put(msgs, timeout=1.)
2284
+ except queue.Full:
2285
+ if not self._heartbeat():
2286
+ return
2287
+ else:
2288
+ break
1905
2289
 
1906
2290
  if self._proc.poll() is not None:
1907
2291
  log.critical('Journalctl process terminated')
1908
- break
2292
+ return
1909
2293
 
1910
2294
  log.debug('Journalctl readable')
1911
2295
  time.sleep(self._sleep_s)
@@ -2039,15 +2423,18 @@ class JournalctlToAws:
2039
2423
 
2040
2424
  @cached_nullary
2041
2425
  def _journalctl_tailer_worker(self) -> JournalctlTailerWorker:
2042
- ac: ta.Optional[str] = self._config.journalctl_after_cursor
2043
- if ac is None:
2044
- ac = self._read_cursor_file()
2045
- if ac is not None:
2046
- log.info('Starting from cursor %s', ac)
2426
+ ac: ta.Optional[str] = None
2047
2427
 
2048
2428
  if (since := self._config.journalctl_since):
2049
2429
  log.info('Starting since %s', since)
2050
2430
 
2431
+ else:
2432
+ ac = self._config.journalctl_after_cursor
2433
+ if ac is None:
2434
+ ac = self._read_cursor_file()
2435
+ if ac is not None:
2436
+ log.info('Starting from cursor %s', ac)
2437
+
2051
2438
  return JournalctlTailerWorker(
2052
2439
  self._journalctl_message_queue(),
2053
2440
 
@@ -2063,9 +2450,9 @@ class JournalctlToAws:
2063
2450
  def run(self) -> None:
2064
2451
  self._ensure_locked()
2065
2452
 
2066
- q = self._journalctl_message_queue()
2067
- jtw = self._journalctl_tailer_worker()
2068
- mp = self._aws_log_message_poster()
2453
+ q = self._journalctl_message_queue() # type: queue.Queue[ta.Sequence[JournalctlMessage]]
2454
+ jtw = self._journalctl_tailer_worker() # type: JournalctlTailerWorker
2455
+ mp = self._aws_log_message_poster() # type: AwsLogMessagePoster
2069
2456
 
2070
2457
  jtw.start()
2071
2458
 
@@ -2075,7 +2462,13 @@ class JournalctlToAws:
2075
2462
  log.critical('Journalctl tailer worker died')
2076
2463
  break
2077
2464
 
2078
- msgs: ta.Sequence[JournalctlMessage] = q.get()
2465
+ try:
2466
+ msgs: ta.Sequence[JournalctlMessage] = q.get(timeout=1.)
2467
+ except queue.Empty:
2468
+ msgs = []
2469
+ if not msgs:
2470
+ continue
2471
+
2079
2472
  log.debug('%r', msgs)
2080
2473
 
2081
2474
  cur_cursor: ta.Optional[str] = None
@@ -2088,10 +2481,14 @@ class JournalctlToAws:
2088
2481
  log.warning('Empty queue chunk')
2089
2482
  continue
2090
2483
 
2091
- [post] = mp.feed([mp.Message(
2092
- message=json.dumps(m.dct),
2093
- ts_ms=int(time.time() * 1000.),
2094
- ) for m in msgs])
2484
+ feed_msgs = []
2485
+ for m in msgs:
2486
+ feed_msgs.append(mp.Message(
2487
+ message=json.dumps(m.dct, sort_keys=True),
2488
+ ts_ms=int((m.ts_us / 1000.) if m.ts_us is not None else (time.time() * 1000.)),
2489
+ ))
2490
+
2491
+ [post] = mp.feed(feed_msgs)
2095
2492
  log.debug('%r', post)
2096
2493
 
2097
2494
  if not self._config.dry_run:
@@ -2158,9 +2555,13 @@ def _main() -> None:
2158
2555
 
2159
2556
  #
2160
2557
 
2161
- for a in ['after_cursor', 'since', 'dry_run']:
2162
- if (pa := getattr(args, a)):
2163
- config = dc.replace(config, **{a: pa})
2558
+ for ca, pa in [
2559
+ ('journalctl_after_cursor', 'after_cursor'),
2560
+ ('journalctl_since', 'since'),
2561
+ ('dry_run', 'dry_run'),
2562
+ ]:
2563
+ if (av := getattr(args, pa)):
2564
+ config = dc.replace(config, **{ca: av})
2164
2565
 
2165
2566
  #
2166
2567