Hi, The Solr cluster is running on a large ec2 instance. Some more details on the environment.
Server : AWS-EC2 instance Zookeeper : Standalone external zookeeper. Solr index : On AWS - EFS Zookeeper data directory : Pointing to a folder on ec2 instance The core deletion is not happening on Solr nodes restart but only on system reboot. Your inputs will be very helpful. Thanks, Modassar On Tue, Jan 25, 2022 at 12:22 PM Modassar Ather <modather1...@gmail.com> wrote: > Thanks for the reply Mike. > > The zookeeper data directory is set to a different location and not to > /tmp. > > The deletion is happening only on system reboot. > > Thanks, > Modassar > > On Tue, Jan 25, 2022 at 1:33 AM Mike Drob <md...@mdrob.com> wrote: > >> Make sure you’re not using /tmp for your data directory. >> >> Do you see deletions on system reboot only or also process restart? >> >> On Mon, Jan 24, 2022 at 12:37 PM Modassar Ather <modather1...@gmail.com> >> wrote: >> >> > Hi, >> > >> > Please provide your suggestions to fix the issue. Is there some >> > configuration which I am missing so the nodes are getting deleted? >> > I tried recreating new cores whenever the Solr server restarted but that >> > has issues like nodes not coming up on certain exceptions. >> > >> > Thanks, >> > Modassar >> > >> > On Thu, Jan 20, 2022 at 10:27 AM Shawn Heisey <apa...@elyograg.org> >> wrote: >> > >> > > On 1/19/2022 9:37 PM, Modassar Ather wrote: >> > > > The clusterstate.json file is empty. >> > > >> > > It's completely normal for recent Solr versions to have an empty >> > > clusterstate.json file. The empty file is still used by Solr's >> > > interaction with zookeeper, but the actual clusterstate is now kept >> on a >> > > per-collection basis in state.json files. I think that >> > > clusterstate.json will no longer be used in a future Solr version. >> > > >> > > Thanks, >> > > Shawn >> > > >> > >> >
