-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
provider removed the deployment as it was unable to (re-)deploy it after receiving updated manifest file, leaving bid/lease open/active #121
Comments
Another case with the
The log file also contained:
More of the log file https://gist.githubusercontent.com/andy108369/b159c06b4f0d57e13f1915a0b3d94a5f/raw/dc56a3dd240970bd64dcb6d14f4ef23e1493440b/err.log The deployment was terminated, manifest & namespace were removed by the akash-provider. Recovering from the issue
Logs - The logs of a failure after
Did repeat the above steps, except for:
Logs - The logs of a successful result after
|
Have occurred again on the Hurricane provider.
I've nuked the deployments and have redeployed them from scratch now:
|
Issue happens after the provider has been rebuilt from scratch (including the VM's / OS / newer kubespray/K8s)
(I've sent you the provider logs in Slack) |
New guess for I'll try to disable them, reboot the worker node and see whether this issue occurs again. refs. |
Can't reproduce this manually.
I believe the enabled unattended upgrades were the root cause of the issue #131 Neither I've had any reports for this issue. |
Re-opening as am still seeing this error (
Provider logs:
|
followed-up by #152 |
provider removed the deployment as it was unable to (re-)deploy it after receiving updated manifest file, leaving bid/lease open/active
provider-services
0.4.6
00:42
provider restarts due toaccount sequence mismatch error
(which is expected mechanism 1 2 to tackle the case when provider would just hang (stop bid on new order requests until manually restarted))00:48
provider fails to update (redeploy) the deployment (send-manifest) (only the image version was bumped, the day before it worked without the issues with the same provider-services v0.4.6)Provider removed the deployment manifest and its namespace after it failed to deploy it upon receiving, leaving bid/lease in the active/open state.
I've manually recovered the lease this way.
Logs
The text was updated successfully, but these errors were encountered: