fix: set restart limits to 0 to prevent being marked as failed #1952
+30
−6
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
samrose
requested changes
Dec 2, 2025
Collaborator
samrose
left a comment
samrose
left a comment
There was a problem hiding this comment.
When we ultimately merge this, we should bump the versions in ansible/vars.yml to create a release for these changes. This way, it will be a distinct change instead of bundled with other changes.
Contributor
Author
|
Hi @samrose - I've just updated the branch. Any updates on this? |
cstockton
force-pushed
the
cs/gotrue-start-limit-fix
branch
2 times, most recently
from
e097bf1 to
3ef31ba
Compare
added 2 commits
December 8, 2025 10:28
The systemd default is 10s / 5 for these values with a DefaultRestartUSec of 100ms. Most services set a RestartSec limit of 3, under most circumstances it takes 15s to restart 5 times so the limit of 10s is not exceeded. However if other system processes (salt, cloud init) restart it explicitly, or recovering system services within the --before chain trigger a restart the limit can be exceeded causing it to be marked as failed. Since no services mark gotrue.service as required it will remain offline until the next explicit restart is issued. Setting these values to 0 with Restart=always and RestartSec=3 will prevent gotrue from being marked as failed.
I've noticed all !oneshot services set a `RestartSec` of `3s` and we use the systemd defaults of `StartLimitBurst=5` and `StartLimitInterval=10s`. Together this forms a property that under typical conditions a service will be restarted indefinitely until it comes back up due to `(3s * 5) > 10s`, but it is still possible for a service to enter a failed state under some scenarios. This change defensively sets them to 0/0 to keep them in restart loops.
cstockton
force-pushed
the
cs/gotrue-start-limit-fix
branch
from
3ef31ba to
c89c805
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The systemd default is 10s / 5 for these values with a DefaultRestartUSec of 100ms. Most services set a RestartSec limit of 3, under most circumstances it takes 15s to restart 5 times so the limit of 10s is not exceeded. However if other system processes (salt, cloud init) restart it explicitly, or recovering system services within the --before chain trigger a restart the limit can be exceeded causing it to be marked as failed. Since no services mark gotrue.service as required it will remain offline until the next explicit restart is issued.
Setting these values to 0 with Restart=always and RestartSec=3 will prevent gotrue from being marked as failed.