Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2378283956 This is a low priority thing; feel free to abandon this until we get to Jetty 12 and JDK 17. In the mean time, we know maxConnectionPerDestination is misconfigured to be too low; will be in

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
HoustonPutman commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2377364632 Thanks for getting those numbers, very interesting that it had no effect. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitH

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2376651527 ``` Benchmark (maxConnectionPerHost) (nodeCount) (numReplicas) (numShards) (preGenerate) (useSmallDocs) Mode Cnt Score Error Units CloudIndexing.index

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2376350232 > Yes; our versions are in `versions.props` I inquired about the Jetty version because I posted a question regarding the cancel_stream error on the Jetty forum, and they informed me

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-237661 There should be way to provide some sort of setting by which user can control this setting and then can play with it. We do have `maxConnectionPerDestination` but that is being used by htt

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-26 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2376310523 Latest Benchmark with different values for maxConnectionPerDestination ``` Benchmark (maxConnectionPerHost) (nodeCount) (numReplicas) (numShards) (preGenerate)

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-24 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2371983637 Yes; our versions are in `versions.props` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-24 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2371907560 We are using jetty 10, right? https://github.com/jetty/jetty.project/issues/10485 -- This is an automated message from the Apache Git Service. To respond to the message, please log on

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-20 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2364337057 > There is a possibility that Jetty is cancelling stream because one of the shard went into recovery! I'm surprised / suspicious app level concerns/errors (Solr returning an HTTP erro

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-20 Thread via GitHub
HoustonPutman commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2364152108 > This one looks different! > > ``` > Client exception => org.apache.solr.common.SolrException: can not use FieldCache on a field w/o docValues unless it is indexed uninverti

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-20 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2364140639 We can test the setup where we are trying to index document and shard is down (rewriting the zookeeper cluster state)and see If we are getting the same `cancel_stream` error. Also I

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-20 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2364134679 Actually I am still not convinced that issue is related to `maxConnectionPerDestination` setting. Because today When I ran it again and again, I found the issue with higher values as well

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-19 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2361850229 Wow; good find! If `setMaxConnectionsPerDestination(1)` then what? I wonder what Jetty is supposed to do when this limit is approached -- like is cancel_stream_error expected in some sense

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-19 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2361774989 MaxConnectionsPerDestination 128 ``` Benchmark (nodeCount) (numReplicas) (numShards) (preGenerate) (useSmallDocs) Mode Cnt Score Error Units Cl

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-19 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2361669105 One thing I noticed that the moment I increased the maxConnectionPerDestination to 8 and keep on increasing till 64, I did'nt see any exception, able to run the full benchmark without any

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-19 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2361655395 MaxConnectionsPerDestination 8 ``` Benchmark (nodeCount) (numReplicas) (numShards) (preGenerate) (useSmallDocs) Mode Cnt Score Error Units

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-19 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2361058171 Maybe you are unfamiliar with Log4j but look at log4j2-bench.xml in this module. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to Gi

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-09-18 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2359947816 Was there anyway to enable more logs for benchmark? I wonder If it's too much overhead with the current Jetty configuration. ``` org.apache.solr.client.solrj.impl.BaseHttpSolrClient$R

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-08-21 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2303631157 I recall encountering a somewhat similar error in the IndexFetcher class when we transitioned to HTTP2. The IndexFetcher class was prematurely closing the client connection and sending a G

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-08-15 Thread via GitHub
dsmiley commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2291627819 `crave run -- ./jmh.sh CloudIndexing -jvmArgsAppend -Dsolr.bench.seed=823716671780822 -p useSmallDocs=true -p nodeCount=1 -p numShards=1 -p numReplicas=1 -I 8` The benchmark module u

Re: [PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-08-15 Thread via GitHub
iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2291372990 Could you please walk me through the process of running this specific benchmark which is giving the error, such as a command, or anything which helps me re-produce it on my system. --

[PR] Benchmark: HTTP2 fix cancel_stream_error [solr]

2024-08-01 Thread via GitHub
dsmiley opened a new pull request, #2612: URL: https://github.com/apache/solr/pull/2612 Fixes Jetty HTTP2 cancel_stream_error by removing the flow control limit. I was encountering this issue with CloudIndexing, esp when using single-node, single-replica, single-shard -- like 30K ops