cpl 2.1.0 → 2.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 468f47cb0e1cf3cdb710b885949188c68f40fd74ad4236379a5a6c6fd55739c1
-  data.tar.gz: fc83d4ef5ee3ded08d52ca5f2df92e5ee261df6fcefd27f673452d0ba265a650
+  metadata.gz: 27661f3a2e2aff2f7018e2cb0e77835ee4de88a8ee0268ec549cafa9d02dfbaa
+  data.tar.gz: 001a114b051c3501b5449eb6d3e4c9f7dea3fa608c6e4595e2129c3bcad8f855
 SHA512:
-  metadata.gz: dac1051680d2b3c69473636a9babd19b8799ea6e2eaadbb26582836641f7a73e843e4a26c32f41206edb0a6b21ab89b9b62cc08087360d75a562c3b6bb40188d
-  data.tar.gz: 52ca1f46ff0d459f450e1a1bd82b6efce780e9b6480ebbda955722eef4bea3a98d71cdaff076691b4cfe93e77c42285f358bfb4ba9adf857c961108302a4f5d6
+  metadata.gz: 38dfd9ca9aa2d28dc7df577677e5732f37740448beb857269ff156da7a8ff2f8627872c958b63e9d1d308ce091df1376e44c09aff96bc82f128edb887e1aba94
+  data.tar.gz: 14c73bf485166dd5be51ff276b6cef8aa3152bb0482fd92b6752e47575cd43ab0bd620ba1c49de619863cff955b628477eb32188a8f76393423c50b1a81e5b07

data/CHANGELOG.md

@@ -16,6 +16,22 @@ _Please add entries here for your pull requests that have not yet been released.
 
 ### Fixed
 
+- Fixed issue where `ps:wait` command hangs forever if workloads are suspended. [PR 198](https://github.com/shakacode/control-plane-flow/pull/198) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
+
+### Added
+
+- Added a timeout for `run` jobs (6 hours by default, but configurable through `runner_job_timeout` in `controlplane.yml`). [PR 194](https://github.com/shakacode/control-plane-flow/pull/194) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
+
+### Changed
+
+- `run` command now overrides the `--image`, `--cpu`, and `--memory` for each job separately, which completely removes any race conditions when running simultaneous jobs with different overrides. [PR 182](https://github.com/shakacode/control-plane-flow/pull/182) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
+- `run` jobs now use a CPU size of 1 (1 core) and a memory size of 2Gi (2 gibibytes) by default (configurable through `runner_job_default_cpu` and `runner_job_default_memory` in `controlplane.yml`). [PR 182](https://github.com/shakacode/control-plane-flow/pull/182) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
+- `run` command now keeps ENV values synced between original and runner workloads. [PR 196](https://github.com/shakacode/control-plane-flow/pull/196) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
+
+## [2.1.0] - 2024-05-27
+
+### Fixed
+
 - Fixed issue where release script was not running from the app image. [PR 183](https://github.com/shakacode/control-plane-flow/pull/183) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
 - Fixed issue where deprecated options were not being warned. [PR 183](https://github.com/shakacode/control-plane-flow/pull/183) by [Rafael Gomes](https://github.com/rafaelgomesxyz).
 
@@ -214,7 +230,8 @@ _Please add entries here for your pull requests that have not yet been released.
 
 - Initial release
 
-[Unreleased]: https://github.com/shakacode/control-plane-flow/compare/v2.0.2...HEAD
+[Unreleased]: https://github.com/shakacode/control-plane-flow/compare/v2.1.0...HEAD
+[2.1.0]: https://github.com/shakacode/control-plane-flow/compare/v2.0.2...v2.1.0
 [2.0.2]: https://github.com/shakacode/control-plane-flow/compare/v2.0.1...v2.0.2
 [2.0.1]: https://github.com/shakacode/control-plane-flow/compare/v2.0.0...v2.0.1
 [2.0.0]: https://github.com/shakacode/control-plane-flow/compare/v1.4.0...v2.0.0

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    cpl (2.1.0)
+    cpl (2.2.0)
       debug (~> 1.7.1)
       dotenv (~> 2.8.1)
       jwt (~> 2.8.1)
@@ -52,7 +52,7 @@ GEM
     rdoc (6.7.0)
       psych (>= 4.0.0)
     regexp_parser (2.9.0)
-    reline (0.5.7)
+    reline (0.5.8)
       io-console (~> 0.5)
     rexml (3.2.6)
     rspec (3.12.0)

data/README.md

@@ -249,6 +249,18 @@ aliases:
     # when running `cpl run`.
     fix_terminal_size: true
 
+    # Sets a default CPU size for `cpl run` jobs (can be overridden per job through `--cpu`).
+    # If not specified, defaults to "1" (1 core).
+    runner_job_default_cpu: "2"
+
+    # Sets a default memory size for `cpl run` jobs (can be overridden per job through `--memory`).
+    # If not specified, defaults to "2Gi" (2 gibibytes).
+    runner_job_default_memory: "4Gi"
+
+    # Sets the maximum number of seconds that `cpl run` jobs can execute before being stopped.
+    # If not specified, defaults to 21600 (6 hours).
+    runner_job_timeout: 1000
+
     # Apps with a deployed image created before this amount of days will be listed for deletion
     # when running the command `cpl cleanup-stale-apps`.
     stale_app_image_deployed_days: 5

data/docs/commands.md

@@ -371,12 +371,17 @@ cpl ps:swait -a $APP_NAME -w $WORKLOAD_NAME
 - - log async fetching for non-interactive mode
 - The Dockerfile entrypoint is used as the command by default, which assumes `exec "${@}"` to be present,
   and the args ["bash", "-c", cmd_to_run] are passed
-- The entrypoint can be overriden through `--entrypoint`, which must be a single command or a script path that exists in the container,
+- The entrypoint can be overridden through `--entrypoint`, which must be a single command or a script path that exists in the container,
   and the args ["bash", "-c", cmd_to_run] are passed,
   unless the entrypoint is `bash`, in which case the args ["-c", cmd_to_run] are passed
 - Providing `--entrypoint none` sets the entrypoint to `bash` by default
 - If `fix_terminal_size` is `true` in the `.controlplane/controlplane.yml` file,
-  the remote terminal size will be fixed to match the local terminal size (may also be overriden through `--terminal-size`)
+  the remote terminal size will be fixed to match the local terminal size (may also be overridden through `--terminal-size`)
+- By default, all jobs use a CPU size of 1 (1 core) and a memory size of 2Gi (2 gibibytes)
+  (can be configured through `runner_job_default_cpu` and `runner_job_default_memory` in `controlplane.yml`,
+  and also overridden per job through `--cpu` and `--memory`)
+- By default, the job is stopped if it takes longer than 6 hours to finish
+  (can be configured though `runner_job_timeout` in `controlplane.yml`)
 
 ```sh
 # Opens shell (bash by default).

data/examples/controlplane.yml

@@ -84,6 +84,18 @@ aliases:
     # when running `cpl run`.
     fix_terminal_size: true
 
+    # Sets a default CPU size for `cpl run` jobs (can be overridden per job through `--cpu`).
+    # If not specified, defaults to "1" (1 core).
+    runner_job_default_cpu: "2"
+
+    # Sets a default memory size for `cpl run` jobs (can be overridden per job through `--memory`).
+    # If not specified, defaults to "2Gi" (2 gibibytes).
+    runner_job_default_memory: "4Gi"
+
+    # Sets the maximum number of seconds that `cpl run` jobs can execute before being stopped.
+    # If not specified, defaults to 21600 (6 hours).
+    runner_job_timeout: 1000
+
     # Apps with a deployed image created before this amount of days will be listed for deletion
     # when running the command `cpl cleanup-stale-apps`.
     stale_app_image_deployed_days: 5

data/lib/command/ps_wait.rb

@@ -22,13 +22,17 @@ module Command
       ```
     EX
 
-    def call
+    def call # rubocop:disable Metrics/MethodLength
       @workloads = [config.options[:workload]] if config.options[:workload]
       @workloads ||= config[:app_workloads] + config[:additional_workloads]
 
       @workloads.reverse_each do |workload|
-        step("Waiting for workload '#{workload}' to be ready", retry_on_failure: true) do
-          cp.workload_deployments_ready?(workload, location: config.location, expected_status: true)
+        if cp.workload_suspended?(workload)
+          progress.puts("Workload '#{workload}' is suspended. Skipping...")
+        else
+          step("Waiting for workload '#{workload}' to be ready", retry_on_failure: true) do
+            cp.workload_deployments_ready?(workload, location: config.location, expected_status: true)
+          end
         end
       end
     end

data/lib/command/run.rb

@@ -36,12 +36,17 @@ module Command
       - - log async fetching for non-interactive mode
       - The Dockerfile entrypoint is used as the command by default, which assumes `exec "${@}"` to be present,
         and the args ["bash", "-c", cmd_to_run] are passed
-      - The entrypoint can be overriden through `--entrypoint`, which must be a single command or a script path that exists in the container,
+      - The entrypoint can be overridden through `--entrypoint`, which must be a single command or a script path that exists in the container,
         and the args ["bash", "-c", cmd_to_run] are passed,
         unless the entrypoint is `bash`, in which case the args ["-c", cmd_to_run] are passed
       - Providing `--entrypoint none` sets the entrypoint to `bash` by default
       - If `fix_terminal_size` is `true` in the `.controlplane/controlplane.yml` file,
-        the remote terminal size will be fixed to match the local terminal size (may also be overriden through `--terminal-size`)
+        the remote terminal size will be fixed to match the local terminal size (may also be overridden through `--terminal-size`)
+      - By default, all jobs use a CPU size of 1 (1 core) and a memory size of 2Gi (2 gibibytes)
+        (can be configured through `runner_job_default_cpu` and `runner_job_default_memory` in `controlplane.yml`,
+        and also overridden per job through `--cpu` and `--memory`)
+      - By default, the job is stopped if it takes longer than 6 hours to finish
+        (can be configured though `runner_job_timeout` in `controlplane.yml`)
     DESC
     EXAMPLES = <<~EX
       ```sh
@@ -84,12 +89,17 @@ module Command
       ```
     EX
 
+    DEFAULT_JOB_CPU = "1"
+    DEFAULT_JOB_MEMORY = "2Gi"
+    DEFAULT_JOB_TIMEOUT = 21_600 # 6 hours
+    DEFAULT_JOB_HISTORY_LIMIT = 10
     MAGIC_END = "---cpl run command finished---"
 
     attr_reader :interactive, :detached, :location, :original_workload, :runner_workload,
+                :default_image, :default_cpu, :default_memory, :job_timeout, :job_history_limit,
                 :container, :expected_deployed_version, :job, :replica, :command
 
-    def call # rubocop:disable Metrics/MethodLength
+    def call # rubocop:disable Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity
       @interactive = config.options[:interactive] || interactive_command?
       @detached = config.options[:detached]
       @log_method = config.options[:log_method]
@@ -97,6 +107,11 @@ module Command
       @location = config.location
       @original_workload = config.options[:workload] || config[:one_off_workload]
       @runner_workload = "#{original_workload}-runner"
+      @default_image = "#{config.app}:#{Controlplane::NO_IMAGE_AVAILABLE}"
+      @default_cpu = config.current[:runner_job_default_cpu] || DEFAULT_JOB_CPU
+      @default_memory = config.current[:runner_job_default_memory] || DEFAULT_JOB_MEMORY
+      @job_timeout = config.current[:runner_job_timeout] || DEFAULT_JOB_TIMEOUT
+      @job_history_limit = DEFAULT_JOB_HISTORY_LIMIT
 
       unless interactive
         @internal_sigint = false
@@ -110,12 +125,10 @@ module Command
         end
       end
 
-      if cp.fetch_workload(runner_workload).nil?
-        create_runner_workload
-        wait_for_runner_workload_create
-      end
+      create_runner_workload if cp.fetch_workload(runner_workload).nil?
+      wait_for_runner_workload_deploy
       update_runner_workload
-      wait_for_runner_workload_update
+      wait_for_runner_workload_update if expected_deployed_version
 
       start_job
       wait_for_replica_for_job
@@ -154,6 +167,11 @@ module Command
         container_spec.delete("livenessProbe")
         container_spec.delete("readinessProbe")
 
+        # Set image, CPU, and memory to default values
+        container_spec["image"] = default_image
+        container_spec["cpu"] = default_cpu
+        container_spec["memory"] = default_memory
+
         # Ensure cron workload won't run per schedule
         spec["defaultOptions"]["suspend"] = true
 
@@ -163,9 +181,14 @@ module Command
 
         # Set cron job props
         spec["type"] = "cron"
+        spec["job"] = {
+          # Next job set to run on January 1st, 2029
+          "schedule" => "0 0 1 1 1",
 
-        # Next job set to run on January 1st, 2029
-        spec["job"] = { "schedule" => "0 0 1 1 1", "restartPolicy" => "Never" }
+          "restartPolicy" => "Never",
+          "activeDeadlineSeconds" => job_timeout,
+          "historyLimit" => job_history_limit
+        }
 
         # Create runner workload
         cp.apply_hash("kind" => "workload", "name" => runner_workload, "spec" => spec)
@@ -173,46 +196,60 @@ module Command
     end
 
     def update_runner_workload # rubocop:disable Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity
-      step("Updating runner workload '#{runner_workload}' based on '#{original_workload}'") do # rubocop:disable Metrics/BlockLength
-        @expected_deployed_version = cp.cron_workload_deployed_version(runner_workload)
-        should_update = false
+      should_update = false
+      spec = nil
 
+      step("Checking if runner workload '#{runner_workload}' needs to be updated") do # rubocop:disable Metrics/BlockLength
         _, original_container_spec = base_workload_specs(original_workload)
         spec, container_spec = base_workload_specs(runner_workload)
 
-        # Override image if specified
-        image = config.options[:image]
-        image_link = if image
-                       image = cp.latest_image if image == "latest"
-                       "/org/#{config.org}/image/#{image}"
-                     else
-                       original_container_spec["image"]
-                     end
-        if container_spec["image"] != image_link
-          container_spec["image"] = image_link
+        # Keep ENV synced between original and runner workloads
+        original_env_str = original_container_spec["env"]&.sort_by { |env| env["name"] }.to_s
+        env_str = container_spec["env"]&.sort_by { |env| env["name"] }.to_s
+        if original_env_str != env_str
+          container_spec["env"] = original_container_spec["env"]
+          should_update = true
+        end
+
+        if container_spec["image"] != default_image
+          container_spec["image"] = default_image
           should_update = true
         end
 
-        # Container overrides
-        if config.options[:cpu] && container_spec["cpu"] != config.options[:cpu]
-          container_spec["cpu"] = config.options[:cpu]
+        if container_spec["cpu"] != default_cpu
+          container_spec["cpu"] = default_cpu
           should_update = true
         end
-        if config.options[:memory] && container_spec["memory"] != config.options[:memory]
-          container_spec["memory"] = config.options[:memory]
+
+        if container_spec["memory"] != default_memory
+          container_spec["memory"] = default_memory
           should_update = true
         end
 
-        next true unless should_update
+        if spec["job"]["activeDeadlineSeconds"] != job_timeout
+          spec["job"]["activeDeadlineSeconds"] = job_timeout
+          should_update = true
+        end
 
+        if spec["job"]["historyLimit"] != job_history_limit
+          spec["job"]["historyLimit"] = job_history_limit
+          should_update = true
+        end
+
+        true
+      end
+
+      return unless should_update
+
+      step("Updating runner workload '#{runner_workload}'") do
         # Update runner workload
-        @expected_deployed_version += 1
+        @expected_deployed_version = cp.cron_workload_deployed_version(runner_workload) + 1
         cp.apply_hash("kind" => "workload", "name" => runner_workload, "spec" => spec)
       end
     end
 
-    def wait_for_runner_workload_create
-      step("Waiting for runner workload '#{runner_workload}' to be created", retry_on_failure: true) do
+    def wait_for_runner_workload_deploy
+      step("Waiting for runner workload '#{runner_workload}' to be deployed", retry_on_failure: true) do
         !cp.cron_workload_deployed_version(runner_workload).nil?
       end
     end
@@ -302,12 +339,14 @@ module Command
     def base_workload_specs(workload)
       spec = cp.fetch_workload!(workload).fetch("spec")
       container_spec = spec["containers"].detect { _1["name"] == original_workload } || spec["containers"].first
-      @container = container_spec["name"]
 
       [spec, container_spec]
     end
 
     def build_job_start_yaml # rubocop:disable Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity
+      _, original_container_spec = base_workload_specs(original_workload)
+      @container = original_container_spec["name"]
+
       job_start_hash = { "name" => container }
 
       if config.options[:use_local_token]
@@ -335,6 +374,18 @@ module Command
         job_start_hash["args"].push('eval "$CPL_RUNNER_SCRIPT"')
       end
 
+      image = config.options[:image]
+      image_link = if image
+                     image = cp.latest_image if image == "latest"
+                     "/org/#{config.org}/image/#{image}"
+                   else
+                     original_container_spec["image"]
+                   end
+
+      job_start_hash["image"] = image_link
+      job_start_hash["cpu"] = config.options[:cpu] if config.options[:cpu]
+      job_start_hash["memory"] = config.options[:memory] if config.options[:memory]
+
       job_start_hash.to_yaml
     end
 
@@ -434,6 +485,8 @@ module Command
     end
 
     def print_detached_commands
+      return unless replica
+
       app_workload_replica_config = app_workload_replica_args.join(" ")
       progress.puts(
         "\n\n" \
@@ -451,7 +504,7 @@ module Command
         Shell.debug("JOB STATUS", status)
 
         case status
-        when "active"
+        when "active", "pending"
           sleep 1
         when "successful"
           break ExitCode::SUCCESS

data/lib/core/controlplane.rb

@@ -256,6 +256,11 @@ class Controlplane # rubocop:disable Metrics/ClassLength
     api.update_workload(org: org, gvc: gvc, workload: workload, data: data)
   end
 
+  def workload_suspended?(workload)
+    details = fetch_workload!(workload)
+    details["spec"]["defaultOptions"]["suspend"]
+  end
+
   def workload_force_redeployment(workload)
     cmd = "cpln workload force-redeployment #{workload} #{gvc_org}"
     perform!(cmd)

data/lib/cpl/version.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
 module Cpl
-  VERSION = "2.1.0"
+  VERSION = "2.2.0"
   MIN_CPLN_VERSION = "2.0.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: cpl
 version: !ruby/object:Gem::Version
-  version: 2.1.0
+  version: 2.2.0
 platform: ruby
 authors:
 - Justin Gordon
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-05-27 00:00:00.000000000 Z
+date: 2024-06-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: debug