Hi All, Found some documents suggesting we should increase the maxClientCnxns count from 60 [default] to 300. Increasing this count, solved my problem.
1. https://community.cloudera.com/t5/Support-Questions/Zookeeper-Issue/m-p/25651 2. https://community.cloudera.com/t5/Support-Questions/setting-quot-maxClientCnxns-quot-option-in-zookeeper/m-p/341370#M233508 On Wed, Jun 26, 2024 at 11:06 AM Saksham Gupta <saksham.gu...@indiamart.com> wrote: > Hi All, > Pinging again for some assistance. > > On Tue, Jun 25, 2024 at 12:32 PM Saksham Gupta < > saksham.gu...@indiamart.com> wrote: > >> Hi All, >> We are facing timeouts while connecting with zookeeper for indexing. >> >> *Exception Details:* org.apache.solr.common.SolrException: >> java.util.concurrent.TimeoutException: Could not connect to ZooKeeper >> 10.128.193.20:2181,10.128.193.21:2181,10.128.193.22:2181 within 15000 ms >> >> On checking this further, we found that the total number of connections >> on the zookeeper cluster was greater than 60. Upon backtracking further, we >> found that we recently added a new collection on zookeeper [cluster of 3], >> and the new collection contains 59 shards on which indexing is happening. >> On stopping indexing on the new collection, everything was fine, and >> indexing got fine on the other collections, as the number of connections on >> zookeeper reduced to less than 60. >> >> My solr cluster consists of 16 nodes, with 5 collections and >70 shards. >> >> Any help on how to resolve this issue and perform indexing on all the >> collections. Is simply increasing the number of maxClientCnxns [in >> zookeeper config] the standard way, and if in future I add even more >> collections, then to what extent should we increase this limit? >> >
