You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@lucene.apache.org by GitBox <gi...@apache.org> on 2021/03/01 17:02:21 UTC
[GitHub] [lucene-solr-operator] HoustonPutman opened a new issue #227: Errors often occur while deleting Solr Pods
HoustonPutman opened a new issue #227:
URL: https://github.com/apache/lucene-solr-operator/issues/227
Its very common for Kubernetes to spew the following error when trying to take down a Solr pod:
> Exec lifecycle hook ([solr stop -p 8983]) for Container "solrcloud-node" in Pod "..." failed - error: command 'solr stop -p 8983' exited with 137: , message: "Sending stop command to Solr running on port 8983 ... waiting up to 180 seconds to allow Jetty process 12 to stop gracefully."
This is likely because Solr and Kubernetes have a different idea of the amount of time to wait before trying to forcefully kill a pod. Solr defaults to a time of 180 seconds, as shown in the message above. The Solr Operator defaults to giving the Solr statefulset a `terminationGracePeriodSeconds` of 10. Therefore Kubernetes will try to kill the pod much earlier than Solr thinks it has time to wait.
We should make sure that the `terminationGracePeriodSeconds` of the statefulSet and the `SOLR_STOP_WAIT` are always in sync with each other (not necessarily the exact same value), and that the default is more reasonable than 10 seconds.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org
[GitHub] [lucene-solr-operator] HoustonPutman closed issue #227: Errors often occur while deleting Solr Pods
Posted by GitBox <gi...@apache.org>.
HoustonPutman closed issue #227:
URL: https://github.com/apache/lucene-solr-operator/issues/227
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org
[GitHub] [lucene-solr-operator] HoustonPutman commented on issue #227: Errors often occur while deleting Solr Pods
Posted by GitBox <gi...@apache.org>.
HoustonPutman commented on issue #227:
URL: https://github.com/apache/lucene-solr-operator/issues/227#issuecomment-788111634
It should be noted, in recent versions of Solr the `bin/solr stop` command can use additional time over the `SOLR_STOP_WAIT`. So we should make sure that we give a buffer between `SOLR_STOP_WAIT` and `terminationGracePeriodSeconds` so that kubernetes doesn't try to kill Solr while `bin/solr stop` is still working. A good initial value for this buffer might be 5 seconds.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org