[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Noble Paul updated SOLR-7191: - Attachment: SOLR-7191.patch removed some unnecessary changes > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey >Assignee: Shalin Shekhar Mangar > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Noble Paul updated SOLR-7191: - Attachment: SOLR-7191.patch > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey >Assignee: Shalin Shekhar Mangar > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Noble Paul updated SOLR-7191: - Attachment: SOLR-7191.patch patch updated to trunk > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey >Assignee: Shalin Shekhar Mangar > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > SOLR-7191.patch, SOLR-7191.patch, lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Damien Kamerman updated SOLR-7191: -- Attachment: SOLR-7191.patch I've had a first look at porting the patch I did for Solr 4.10 to the Solr trunk (1708905). I created 6,000 collections (3 nodes; 2 x replicas) and re-started the 3 nodes, and all green in 24min. The bottleneck was ZkController.register() calling zkStateReader.updateClusterState() for each collection. I moved this call up to the end of CoreContainer.load(). Comments appreciated. > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey >Assignee: Shalin Shekhar Mangar > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > SOLR-7191.patch, lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Damien Kamerman updated SOLR-7191: -- Attachment: SOLR-7191.patch Thanks Shawn, Here's a patch for branch_5x that does the following: 1. Load cores in fixed threadPool. Cores are registered in background threads, so no reason to load all cores at once! 2. Register cores in corename order with a fixed 128 threadPool. This is to not flood the DistributedQueue. > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, SOLR-7191.patch, > lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Damien Kamerman updated SOLR-7191: -- Attachment: SOLR-7191.patch This fix takes some load off the DistributedQueue. On shutdown the node will publish a 'down' state for the entire node and overseer will do the rest. Registration is done in a fixed size (128) threadPool in corename order. > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey > Labels: performance, scalability > Attachments: SOLR-7191.patch, SOLR-7191.patch, > lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Damien Kamerman updated SOLR-7191: -- Attachment: SOLR-7191.patch Patch against lucene_solr_4_10_4. > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey > Labels: performance, scalability > Attachments: SOLR-7191.patch, > lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Updated] (SOLR-7191) Improve stability and startup performance of SolrCloud with thousands of collections
[ https://issues.apache.org/jira/browse/SOLR-7191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Shawn Heisey updated SOLR-7191: --- Attachment: lots-of-zkstatereader-updates-branch_5x.log Above I mentioned a full recovery that I expected to take half a day, testing with the 5.0.0 release. I killed that and started over AGAIN, this time with branch_5x. I'm attaching a log excerpt from branch_5x (during the creation of 4000 collections) showing what happens between the receipt of two CREATE calls to the collections admin. After the first create call, all the stuff for the create happens, and then a massive number of ZkStateReader lines are logged, indicating that an update is happening. The collection (mycoll3221) that was created would be the 3223rd collection on the system, and the log shows 3198 update lines from zkStateReader, then the CREATE call for the next collection. This is happening on EVERY collection create. Why would every existing collection need to be touched when a new collection is created? > Improve stability and startup performance of SolrCloud with thousands of > collections > > > Key: SOLR-7191 > URL: https://issues.apache.org/jira/browse/SOLR-7191 > Project: Solr > Issue Type: Bug > Components: SolrCloud >Affects Versions: 5.0 >Reporter: Shawn Heisey > Labels: performance, scalability > Attachments: lots-of-zkstatereader-updates-branch_5x.log > > > A user on the mailing list with thousands of collections (5000 on 4.10.3, > 4000 on 5.0) is having severe problems with getting Solr to restart. > I tried as hard as I could to duplicate the user setup, but I ran into many > problems myself even before I was able to get 4000 collections created on a > 5.0 example cloud setup. Restarting Solr takes a very long time, and it is > not very stable once it's up and running. > This kind of setup is very much pushing the envelope on SolrCloud performance > and scalability. It doesn't help that I'm running both Solr nodes on one > machine (I started with 'bin/solr -e cloud') and that ZK is embedded. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
