On 8/22/2018 7:09 PM, mmb1234 wrote:
The original post clarified the current state. "we have about 75 cores with
"transientCacheSize" set to 32". If transientCacheSize is increased to match
current cores, we'll differ the issue. It's going to hit 100's cores per
solr instance shortly.
Majority of the cores do not get ingest request so it would be waste to
spend memory resources and that is why all solr.xml for cores is:
The way things are now, you're going to want to have your cache size set
to the number of cores that will get simultaneous updates *plus* enough
more that you can run simultaneous queries on other cores at the same time.
You might need beefier machines that can handle more, so you can either
stop using transient cores altogether or have a much bigger
transientCacheSize.
I created an issue. I can't make any promises, but there might be some
things we can look at to help out with your real production setup. The
2-core reproduction setup might never be something that will work right
- you need upfront warning about that.
https://issues.apache.org/jira/browse/SOLR-12691
Thanks,
Shawn
