Sounds like you should contact aws about it since it’s not a solr issue if the qtimes haven’t increased in the solr logs. And again, don’t load balance but that’s my personal opinion
> On Dec 13, 2022, at 6:50 AM, Pradeep <kumarks.prad...@gmail.com> wrote: > > Hi, > > I cant change it to NLB at this moment, firstly why timeout we need to > understand no clue at this moment. It works fine if i increase timeout from > 60 secs to 4000 secs. Its same code working in classic load balancer but > with ALB we have this issue. > > > Thanks, > Pradeep > >> On Tue, 13 Dec, 2022, 11:08 Deepak Goel, <deic...@gmail.com> wrote: >> >> Can you try 'Network Load Balancer" in aws? >> >> >> Deepak >> "The greatness of a nation can be judged by the way its animals are treated >> - Mahatma Gandhi" >> >> +91 73500 12833 >> deic...@gmail.com >> >> Facebook: https://www.facebook.com/deicool >> LinkedIn: www.linkedin.com/in/deicool >> >> "Plant a Tree, Go Green" >> >> Make In India : http://www.makeinindia.com/home >> >> >> On Tue, Dec 13, 2022 at 12:21 AM Shawn Heisey <elyog...@elyograg.org> >> wrote: >> >>> On 12/12/22 08:14, Pradeep wrote: >>>> We have migrated Apache solr aws load balancer from classic to >>> Application >>>> load balancer. We havent changed anything in solr other than migrating >>> load >>>> balancer but solr taking more time while committing in Solr. >>>> >>>> Fyi, we are processing 200 records every batch, this is failing at last >>>> stage of commit in solr (deleting stale records in solr). But same >> logic >>>> working fine from many years in classic load balancer but after >> migration >>>> to ALB this is happenning. >>> >>> The message subject mentions optimize but the message body does not. >>> >>> Is optimize involved? An optimize of a large index is always going to >>> be slow. >>> >>> Solr does not have any kind of "deleting stale records" that I have ever >>> heard of, so that must be something your code, which is more difficult >>> for us to troubleshoot because we are not going to have any idea how to >>> get that code to give useful logs without a lot more information about >>> that code. >>> >>> If you're encountering an error from Solr then solr.log should have at >>> least some info about what happened and that might be enough info to >>> figure out the problem. >>> >>> If changing the load balancer caused issues, then you may need to talk >>> to someone who understands how to troubleshoot the LB. >>> >>> Basically we need a lot more information about what's happening to have >>> any hope of helping you. >>> >>> Thanks, >>> Shawn >>> >>> >>
