Configure HealthCheck with podman update

New flags in a `podman update` can change the configuration of HealthCheck when the container is started, without having to restart or recreate the container. This can help determine why a given container suddenly started failing HealthCheck without interfering with the services it provides. For example, reconfigure HealthCheck to keep logs longer than the usual last X results, store logs to other destinations, etc. Fixes: https://issues.redhat.com/browse/RHEL-60561 Signed-off-by: Jan Rodák <[email protected]>
containers · Nov 12, 2024 · dfb98ce · dfb98ce
1 parent c8af2f2
commit dfb98ce
Show file tree

Hide file tree

Showing 31 changed files with 1,015 additions and 180 deletions.
diff --git a/cmd/podman/common/create.go b/cmd/podman/common/create.go
@@ -168,78 +168,6 @@ func DefineCreateFlags(cmd *cobra.Command, cf *entities.ContainerCreateOptions,
 		)
 		_ = cmd.RegisterFlagCompletionFunc(groupAddFlagName, completion.AutocompleteNone)
 
-		healthCmdFlagName := "health-cmd"
-		createFlags.StringVar(
-			&cf.HealthCmd,
-			healthCmdFlagName, "",
-			"set a healthcheck command for the container ('none' disables the existing healthcheck)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthCmdFlagName, completion.AutocompleteNone)
-
-		healthIntervalFlagName := "health-interval"
-		createFlags.StringVar(
-			&cf.HealthInterval,
-			healthIntervalFlagName, define.DefaultHealthCheckInterval,
-			"set an interval for the healthcheck (a value of disable results in no automatic timer setup)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthIntervalFlagName, completion.AutocompleteNone)
-
-		healthLogDestinationFlagName := "health-log-destination"
-		createFlags.StringVar(
-			&cf.HealthLogDestination,
-			healthLogDestinationFlagName, define.DefaultHealthCheckLocalDestination,
-			"set the destination of the HealthCheck log. Directory path, local or events_logger (local use container state file)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthLogDestinationFlagName, completion.AutocompleteNone)
-
-		healthMaxLogCountFlagName := "health-max-log-count"
-		createFlags.UintVar(
-			&cf.HealthMaxLogCount,
-			healthMaxLogCountFlagName, define.DefaultHealthMaxLogCount,
-			"set maximum number of attempts in the HealthCheck log file. ('0' value means an infinite number of attempts in the log file)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthMaxLogCountFlagName, completion.AutocompleteNone)
-
-		healthMaxLogSizeFlagName := "health-max-log-size"
-		createFlags.UintVar(
-			&cf.HealthMaxLogSize,
-			healthMaxLogSizeFlagName, define.DefaultHealthMaxLogSize,
-			"set maximum length in characters of stored HealthCheck log. ('0' value means an infinite log length)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthMaxLogSizeFlagName, completion.AutocompleteNone)
-
-		healthRetriesFlagName := "health-retries"
-		createFlags.UintVar(
-			&cf.HealthRetries,
-			healthRetriesFlagName, define.DefaultHealthCheckRetries,
-			"the number of retries allowed before a healthcheck is considered to be unhealthy",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthRetriesFlagName, completion.AutocompleteNone)
-
-		healthStartPeriodFlagName := "health-start-period"
-		createFlags.StringVar(
-			&cf.HealthStartPeriod,
-			healthStartPeriodFlagName, define.DefaultHealthCheckStartPeriod,
-			"the initialization time needed for a container to bootstrap",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthStartPeriodFlagName, completion.AutocompleteNone)
-
-		healthTimeoutFlagName := "health-timeout"
-		createFlags.StringVar(
-			&cf.HealthTimeout,
-			healthTimeoutFlagName, define.DefaultHealthCheckTimeout,
-			"the maximum time allowed to complete the healthcheck before an interval is considered failed",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthTimeoutFlagName, completion.AutocompleteNone)
-
-		healthOnFailureFlagName := "health-on-failure"
-		createFlags.StringVar(
-			&cf.HealthOnFailure,
-			healthOnFailureFlagName, "none",
-			"action to take once the container turns unhealthy",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(healthOnFailureFlagName, AutocompleteHealthOnFailure)
-
 		createFlags.BoolVar(
 			&cf.HTTPProxy,
 			"http-proxy", podmanConfig.ContainersConfDefaultsRO.Containers.HTTPProxy,
@@ -311,11 +239,6 @@ func DefineCreateFlags(cmd *cobra.Command, cf *entities.ContainerCreateOptions,
 		)
 		_ = cmd.RegisterFlagCompletionFunc(logOptFlagName, AutocompleteLogOpt)
 
-		createFlags.BoolVar(
-			&cf.NoHealthCheck,
-			"no-healthcheck", false,
-			"Disable healthchecks on container",
-		)
 		createFlags.BoolVar(
 			&cf.OOMKillDisable,
 			"oom-kill-disable", false,
@@ -452,46 +375,6 @@ func DefineCreateFlags(cmd *cobra.Command, cf *entities.ContainerCreateOptions,
 		)
 		_ = cmd.RegisterFlagCompletionFunc(secretFlagName, AutocompleteSecrets)
 
-		startupHCCmdFlagName := "health-startup-cmd"
-		createFlags.StringVar(
-			&cf.StartupHCCmd,
-			startupHCCmdFlagName, "",
-			"Set a startup healthcheck command for the container",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(startupHCCmdFlagName, completion.AutocompleteNone)
-
-		startupHCIntervalFlagName := "health-startup-interval"
-		createFlags.StringVar(
-			&cf.StartupHCInterval,
-			startupHCIntervalFlagName, define.DefaultHealthCheckInterval,
-			"Set an interval for the startup healthcheck",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(startupHCIntervalFlagName, completion.AutocompleteNone)
-
-		startupHCRetriesFlagName := "health-startup-retries"
-		createFlags.UintVar(
-			&cf.StartupHCRetries,
-			startupHCRetriesFlagName, 0,
-			"Set the maximum number of retries before the startup healthcheck will restart the container",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(startupHCRetriesFlagName, completion.AutocompleteNone)
-
-		startupHCSuccessesFlagName := "health-startup-success"
-		createFlags.UintVar(
-			&cf.StartupHCSuccesses,
-			startupHCSuccessesFlagName, 0,
-			"Set the number of consecutive successes before the startup healthcheck is marked as successful and the normal healthcheck begins (0 indicates any success will start the regular healthcheck)",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(startupHCSuccessesFlagName, completion.AutocompleteNone)
-
-		startupHCTimeoutFlagName := "health-startup-timeout"
-		createFlags.StringVar(
-			&cf.StartupHCTimeout,
-			startupHCTimeoutFlagName, define.DefaultHealthCheckTimeout,
-			"Set the maximum amount of time that the startup healthcheck may take before it is considered failed",
-		)
-		_ = cmd.RegisterFlagCompletionFunc(startupHCTimeoutFlagName, completion.AutocompleteNone)
-
 		stopSignalFlagName := "stop-signal"
 		createFlags.StringVar(
 			&cf.StopSignal,
@@ -665,6 +548,140 @@ func DefineCreateFlags(cmd *cobra.Command, cf *entities.ContainerCreateOptions,
 			`If a container with the same name exists, replace it`,
 		)
 	}
+	if mode == entities.CreateMode || mode == entities.UpdateMode {
+		createFlags.BoolVar(
+			&cf.NoHealthCheck,
+			"no-healthcheck", false,
+			"Disable healthchecks on container",
+		)
+
+		healthCmdFlagName := "health-cmd"
+		createFlags.StringVar(
+			&cf.HealthCmd,
+			healthCmdFlagName, "",
+			"set a healthcheck command for the container ('none' disables the existing healthcheck)",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthCmdFlagName, completion.AutocompleteNone)
+
+		info := ""
+		if mode == entities.UpdateMode {
+			info = "Changing this setting resets timer."
+		}
+		healthIntervalFlagName := "health-interval"
+		createFlags.StringVar(
+			&cf.HealthInterval,
+			healthIntervalFlagName, define.DefaultHealthCheckInterval,
+			"set an interval for the healthcheck. (a value of disable results in no automatic timer setup) "+info,
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthIntervalFlagName, completion.AutocompleteNone)
+
+		warning := ""
+		if mode == entities.UpdateMode {
+			warning = "Warning: Changing this setting may cause the loss of previous logs!"
+		}
+		healthLogDestinationFlagName := "health-log-destination"
+		createFlags.StringVar(
+			&cf.HealthLogDestination,
+			healthLogDestinationFlagName, define.DefaultHealthCheckLocalDestination,
+			"set the destination of the HealthCheck log. Directory path, local or events_logger (local use container state file) "+warning,
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthLogDestinationFlagName, completion.AutocompleteNone)
+
+		healthMaxLogCountFlagName := "health-max-log-count"
+		createFlags.UintVar(
+			&cf.HealthMaxLogCount,
+			healthMaxLogCountFlagName, define.DefaultHealthMaxLogCount,
+			"set maximum number of attempts in the HealthCheck log file. ('0' value means an infinite number of attempts in the log file)",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthMaxLogCountFlagName, completion.AutocompleteNone)
+
+		healthMaxLogSizeFlagName := "health-max-log-size"
+		createFlags.UintVar(
+			&cf.HealthMaxLogSize,
+			healthMaxLogSizeFlagName, define.DefaultHealthMaxLogSize,
+			"set maximum length in characters of stored HealthCheck log. ('0' value means an infinite log length)",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthMaxLogSizeFlagName, completion.AutocompleteNone)
+
+		healthRetriesFlagName := "health-retries"
+		createFlags.UintVar(
+			&cf.HealthRetries,
+			healthRetriesFlagName, define.DefaultHealthCheckRetries,
+			"the number of retries allowed before a healthcheck is considered to be unhealthy",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthRetriesFlagName, completion.AutocompleteNone)
+
+		healthStartPeriodFlagName := "health-start-period"
+		createFlags.StringVar(
+			&cf.HealthStartPeriod,
+			healthStartPeriodFlagName, define.DefaultHealthCheckStartPeriod,
+			"the initialization time needed for a container to bootstrap",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthStartPeriodFlagName, completion.AutocompleteNone)
+
+		healthTimeoutFlagName := "health-timeout"
+		createFlags.StringVar(
+			&cf.HealthTimeout,
+			healthTimeoutFlagName, define.DefaultHealthCheckTimeout,
+			"the maximum time allowed to complete the healthcheck before an interval is considered failed",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthTimeoutFlagName, completion.AutocompleteNone)
+
+		healthOnFailureFlagName := "health-on-failure"
+		createFlags.StringVar(
+			&cf.HealthOnFailure,
+			healthOnFailureFlagName, "none",
+			"action to take once the container turns unhealthy",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(healthOnFailureFlagName, AutocompleteHealthOnFailure)
+
+		// Startup HealthCheck
+
+		startupHCCmdFlagName := "health-startup-cmd"
+		createFlags.StringVar(
+			&cf.StartupHCCmd,
+			startupHCCmdFlagName, "",
+			"Set a startup healthcheck command for the container",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(startupHCCmdFlagName, completion.AutocompleteNone)
+
+		info = ""
+		if mode == entities.UpdateMode {
+			info = "Changing this setting resets the timer, depending on the state of the container."
+		}
+		startupHCIntervalFlagName := "health-startup-interval"
+		createFlags.StringVar(
+			&cf.StartupHCInterval,
+			startupHCIntervalFlagName, define.DefaultHealthCheckInterval,
+			"Set an interval for the startup healthcheck. "+info,
+		)
+		_ = cmd.RegisterFlagCompletionFunc(startupHCIntervalFlagName, completion.AutocompleteNone)
+
+		startupHCRetriesFlagName := "health-startup-retries"
+		createFlags.UintVar(
+			&cf.StartupHCRetries,
+			startupHCRetriesFlagName, 0,
+			"Set the maximum number of retries before the startup healthcheck will restart the container",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(startupHCRetriesFlagName, completion.AutocompleteNone)
+
+		startupHCSuccessesFlagName := "health-startup-success"
+		createFlags.UintVar(
+			&cf.StartupHCSuccesses,
+			startupHCSuccessesFlagName, 0,
+			"Set the number of consecutive successes before the startup healthcheck is marked as successful and the normal healthcheck begins (0 indicates any success will start the regular healthcheck)",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(startupHCSuccessesFlagName, completion.AutocompleteNone)
+
+		startupHCTimeoutFlagName := "health-startup-timeout"
+		createFlags.StringVar(
+			&cf.StartupHCTimeout,
+			startupHCTimeoutFlagName, define.DefaultHealthCheckTimeout,
+			"Set the maximum amount of time that the startup healthcheck may take before it is considered failed",
+		)
+		_ = cmd.RegisterFlagCompletionFunc(startupHCTimeoutFlagName, completion.AutocompleteNone)
+	}
+
 	// Restart is allowed for created, updated, and infra ctr
 	if mode == entities.InfraMode || mode == entities.CreateMode || mode == entities.UpdateMode {
 		restartFlagName := "restart"

diff --git a/cmd/podman/containers/update.go b/cmd/podman/containers/update.go
@@ -17,7 +17,7 @@ import (
 )
 
 var (
-	updateDescription = `Updates the cgroup configuration of a given container`
+	updateDescription = `Updates the configuration of an existing container, allowing changes to resource limits and healthchecks`
 
 	updateCommand = &cobra.Command{
 		Use:               "update [options] CONTAINER",
@@ -89,9 +89,15 @@ func update(cmd *cobra.Command, args []string) error {
 		return err
 	}
 
+	healthCheckConfig := specgenutil.GetChangedHealthCheckConfiguration(cmd)
+	if err != nil {
+		return err
+	}
+
 	opts := &entities.ContainerUpdateOptions{
-		NameOrID: strings.TrimPrefix(args[0], "/"),
-		Specgen:  s,
+		NameOrID:                        strings.TrimPrefix(args[0], "/"),
+		Specgen:                         s,
+		ChangedHealthCheckConfiguration: &healthCheckConfig,
 	}
 	rep, err := registry.ContainerEngine().ContainerUpdate(context.Background(), opts)
 	if err != nil {

diff --git a/docs/source/markdown/options/health-cmd.md b/docs/source/markdown/options/health-cmd.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-cmd**=*"command"* | *'["command", "arg1", ...]'*

diff --git a/docs/source/markdown/options/health-interval.md b/docs/source/markdown/options/health-interval.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-interval**=*interval*

diff --git a/docs/source/markdown/options/health-log-destination.md b/docs/source/markdown/options/health-log-destination.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-log-destination**=*directory_path*

diff --git a/docs/source/markdown/options/health-max-log-count.md b/docs/source/markdown/options/health-max-log-count.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-max-log-count**=*number of stored logs*

diff --git a/docs/source/markdown/options/health-max-log-size.md b/docs/source/markdown/options/health-max-log-size.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-max-log-size**=*size of stored logs*

diff --git a/docs/source/markdown/options/health-on-failure.md b/docs/source/markdown/options/health-on-failure.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-on-failure**=*action*

diff --git a/docs/source/markdown/options/health-retries.md b/docs/source/markdown/options/health-retries.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-retries**=*retries*

diff --git a/docs/source/markdown/options/health-start-period.md b/docs/source/markdown/options/health-start-period.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-start-period**=*period*

diff --git a/docs/source/markdown/options/health-startup-cmd.md b/docs/source/markdown/options/health-startup-cmd.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-startup-cmd**=*"command"* | *'["command", "arg1", ...]'*

diff --git a/docs/source/markdown/options/health-startup-interval.md b/docs/source/markdown/options/health-startup-interval.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-startup-interval**=*interval*

diff --git a/docs/source/markdown/options/health-startup-retries.md b/docs/source/markdown/options/health-startup-retries.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-startup-retries**=*retries*

diff --git a/docs/source/markdown/options/health-startup-success.md b/docs/source/markdown/options/health-startup-success.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-startup-success**=*retries*

diff --git a/docs/source/markdown/options/health-startup-timeout.md b/docs/source/markdown/options/health-startup-timeout.md
@@ -1,5 +1,5 @@
 ####> This option file is used in:
-####>   podman create, run
+####>   podman create, run, update
 ####> If file is edited, make sure the changes
 ####> are applicable to all of those.
 #### **--health-startup-timeout**=*timeout*