It might also be http2. Ive seen numerous issues with http2 performance (which will hopefully be fixed soon). This can be tested with “-Dsolr.http1=true”
- Houston On Fri, May 24, 2024 at 1:25 PM Eric Pugh <ep...@opensourceconnections.com> wrote: > I’d be curious if other folks have seen this either anecdotally or even > better through some performance testing! > > > On May 24, 2024, at 9:12 AM, Tim Funk <funk...@apache.org> wrote: > > > > solr9 turns security on manager by default. This has a high probability > of > > a performance impact.. > > > > disable via adding to startup > > export SOLR_SECURITY_MANAGER_ENABLED=false > > > > -Tim > > > > On Thu, May 23, 2024 at 11:23 AM elisabeth benoit < > elisaelisael...@gmail.com> > > wrote: > > > >> Hello, > >> > >> We've just migrated one of our services from solr 7.3.1 to solr 9.2.1 in > >> production. > >> > >> We have 6 servers, each with one solr instance, all conf is automated > with > >> puppet, git, etc. so they all have same conf, but they dont behave the > >> same way. We have two servers with solr response time much faster, from > >> 50ms to 75ms. > >> > >> This week we rebooted all our servers, and now we have three solr > instances > >> running faster than the three other ones instead of two. > >> > >> We've check everything we could think of: memory, GC, caches, > everything is > >> the same. > >> > >> We're not in solr cloud, but with leader conf. > >> > >> Just writing in case someone has an idea about this unstable behaviour. > >> > >> > > _______________________ > Eric Pugh | Founder | OpenSource Connections, LLC | 434.466.1467 | > http://www.opensourceconnections.com < > http://www.opensourceconnections.com/> | My Free/Busy < > http://tinyurl.com/eric-cal> > Co-Author: Apache Solr Enterprise Search Server, 3rd Ed < > https://www.packtpub.com/big-data-and-business-intelligence/apache-solr-enterprise-search-server-third-edition-raw> > > This e-mail and all contents, including attachments, is considered to be > Company Confidential unless explicitly stated otherwise, regardless of > whether attachments are marked as such. > >
