Hi all, I have a Cluster of 2 Solr 8.11 nodes running in cloud mode and 1 zookeeper node, this is used for our testing environment, the servers are in Amazon Ec2 and are t3.xlarge (4 cores and 16gb of ram), heap has been set to 8g.
Everything has been working very well since I installed these, several months ago, I have 6 different with around 11M records each (each collection correspond to one of our testing environments). Indexing and searching has been fast as we don't have a lot of traffic on those, but suddenly, yesterday we started having very high CPU spikes (main solr process up to 400% CPU), this is causing both a slow down on searches, but an incredibly long delay when indexing, for example when we index a record, it takes about 10 minutes for those records to be available on search, our commit strategy has been hard commit every 60 seconds and soft commit every 1 second. As far as I know, we haven't installed anything on those servers, I've tried restarting both nodes and they keep having the same issue. I'm fairly new to Solr and SolrCloud, specially maintaining servers so I'm not sure what to do or what information to provide, but I would appreciate any help you can provide! The curious thing is that before this started happening, we were using t3.large instances (2 cores and 8gb of ram) with a 4g heap with no issues, we decided to upgrade because the CPU spikes were bringing down the solr processes. Thanks in advance for any possible help. Matias Laino | DIRECTOR OF PASSARE REMOTE DEVELOPMENT matias.la...@passare.com<mailto:matias.la...@passare.com> | +54 11-6357-2143 [Image]
