Starting of constantly restarting container fails #196
Comments
dimitar-dimitrow
added a commit
to bosch-io/container-management
that referenced
this issue
Nov 23, 2023
- Do not release container resources when restart manager is canceled, they are already released upon exit. - Apply restart policy when it is updated. Signed-off-by: Dimitar Dimitrov <dimitar.dimitrov3@bosch.com>
k-gostev
pushed a commit
that referenced
this issue
Nov 23, 2023
k-gostev
pushed a commit
that referenced
this issue
Apr 30, 2024
dimitar-dimitrow
added a commit
that referenced
this issue
May 10, 2024
[#234] Merge `dev-m5` branch into `main` * [#51] Improve containerd client unit tests (#203) * [#201] Optimized, deterministic intermediate desired state feedback messages (#204) * [#208] Add file flag to the CLI create command (#209) * [#191] Container remains Stopped after container-management service restart (#214) * [#210] Remove command should accept more than one container ID (#212) * [#196] Starting of constantly restarting container fails (#216) * [#67] Implement signed images verification (#215) * [#91] Provide unit tests covering signed images verification (#220) * [#213] Add quiet flag, to the list command. (#221) * [#217] CLI Remove command improvements (#224) --------- Signed-off-by: Daniel Milchev fixed-term.daniel.milchev@bosch.io Signed-off-by: Stoyan Zoubev <Stoyan.Zoubev@bosch.com> Signed-off-by: Kristiyan Gostev <kristiyan.gostev@bosch.com> Signed-off-by: Dimitar Dimitrov <dimitar.dimitrov3@bosch.com> Co-authored-by: Daniel Milchev <fixed-term.daniel.milchev@bosch.io> Co-authored-by: Stoyan Zoubev <Stoyan.Zoubev@bosch.io> Co-authored-by: Dimitar Dimitrov <dimitar.dimitrov3@bosch.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
When a constantly restarting container update is received from the Update Manager, the Container's Update Agent, tries to start the container. However, there seems to be a concurrency problem, between the UA and the auto restart mechanism. The update activity finishes with status INCOMPLETE and the following error is logged:
time="2023-08-31T09:28:17.50888366Z" level=error msg="[container-management][ctrd_client.go:255][pkg:github][func:com/eclipse-kanto/container-management/containerm/ctr] error creating task for container ID = cd0be8fb-e533-4cce-bbc4-f0db9d0775d8 \n\t Error: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: time=\"2023-08-31T09:28:17Z\" level=fatal msg=\"no sandbox present for cd0be8fb-e533-4cce-bbc4-f0db9d0775d8\": unknown " time="2023-08-31T09:28:17.525847659Z" level=error msg="[container-management][update_operation.go:546][pkg:github][func:com/eclipse-kanto/container-management/containerm/updateagent] could not start container [hello-world] \n\t Error: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: time=\"2023-08-31T09:28:17Z\" level=fatal msg=\"no sandbox present for cd0be8fb-e533-4cce-bbc4-f0db9d0775d8\": unknown "The text was updated successfully, but these errors were encountered: