Re: SolrCloud Error after leader restarts
Your using ram dir? Sent from my iPhone On Nov 19, 2012, at 1:21 AM, deniz denizdurmu...@gmail.com wrote: Hello, for test purposes, I am running two zookeepers on ports 2181 and 2182. and i have two solr instances running on different machines... For the one which is running on my local and acts as leader: java -Dbootstrap_conf=true -DzkHost=localhost:2181 -jar start.jar and for the one which acts as follower, on a remote machine: java -Djetty.port=7574 -DzkHost=address-of-mylocal:2182 -jar start.jar until this point everything is smooth and i can see the configs on both zookeeper hosts when i connect with zkCli.sh. just to see what happens and check recovery stuff, i have killed the solr which is running on my local and tried to index some files by using the follewer, which was failed... this is normal as writes are routed into the leader... the point that i dont understand is here: when i restart the leader with the same command on terminal, after normal logs, it start showing this Nov 19, 2012 2:15:18 PM org.apache.solr.common.SolrException log SEVERE: SnapPull failed :org.apache.solr.common.SolrException: Index fetch failed : at org.apache.solr.handler.SnapPuller.fetchLatestIndex(SnapPuller.java:400) at org.apache.solr.handler.ReplicationHandler.doFetch(ReplicationHandler.java:297) at org.apache.solr.cloud.RecoveryStrategy.replicate(RecoveryStrategy.java:151) at org.apache.solr.cloud.RecoveryStrategy.doRecovery(RecoveryStrategy.java:405) at org.apache.solr.cloud.RecoveryStrategy.run(RecoveryStrategy.java:220) Caused by: org.apache.lucene.index.IndexNotFoundException: no segments* file found in org.apache.lucene.store.RAMDirectory@1e75e89 lockFactory=org.apache.lucene.store.NativeFSLockFactory@128e909: files: [] at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:741) at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:630) at org.apache.lucene.index.SegmentInfos.read(SegmentInfos.java:343) at org.apache.lucene.index.IndexWriter.init(IndexWriter.java:639) at org.apache.solr.update.SolrIndexWriter.init(SolrIndexWriter.java:75) at org.apache.solr.update.SolrIndexWriter.create(SolrIndexWriter.java:62) at org.apache.solr.update.DefaultSolrCoreState.createMainIndexWriter(DefaultSolrCoreState.java:191) at org.apache.solr.update.DefaultSolrCoreState.getIndexWriter(DefaultSolrCoreState.java:77) at org.apache.solr.handler.SnapPuller.fetchLatestIndex(SnapPuller.java:354) ... 4 more Nov 19, 2012 2:15:18 PM org.apache.solr.common.SolrException log SEVERE: Error while trying to recover:org.apache.solr.common.SolrException: Replication for recovery failed. at org.apache.solr.cloud.RecoveryStrategy.replicate(RecoveryStrategy.java:154) at org.apache.solr.cloud.RecoveryStrategy.doRecovery(RecoveryStrategy.java:405) at org.apache.solr.cloud.RecoveryStrategy.run(RecoveryStrategy.java:220) it fails to recover after shutdown... why does this happen? - Zeki ama calismiyor... Calissa yapar... -- View this message in context: http://lucene.472066.n3.nabble.com/SolrCloud-Error-after-leader-restarts-tp4020985.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: SolrCloud Error after leader restarts
It's generally not a good choice to use ram directory. 4x solrcloud does not work with it no - 5x does, but in any case, ram dir is not persistent. So when you restart Solr you will lose the data. MMap is generally the right dir to use. - Mark On Nov 19, 2012, at 6:52 PM, deniz denizdurmu...@gmail.com wrote: yea, i am using ram. solrcloud is not working with ram directory? - Zeki ama calismiyor... Calissa yapar... -- View this message in context: http://lucene.472066.n3.nabble.com/SolrCloud-Error-after-leader-restarts-tp4020985p4021194.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: SolrCloud Error after leader restarts
i know facts about ramdirectory actually.. just running some perf tests on our dev env right now.. so in case i use ramdir with 5x cloud, it will still not do the recovery? i mean it will not get the data from the leader and fill its ramdir again? - Zeki ama calismiyor... Calissa yapar... -- View this message in context: http://lucene.472066.n3.nabble.com/SolrCloud-Error-after-leader-restarts-tp4020985p4021203.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: SolrCloud Error after leader restarts
On Nov 19, 2012, at 9:11 PM, deniz denizdurmu...@gmail.com wrote: so in case i use ramdir with 5x cloud, it will still not do the recovery? i mean it will not get the data from the leader and fill its ramdir again? Yes, in 5x RAM directory should be able to recover. - Mark
Re: SolrCloud Error after leader restarts
Mark Miller-3 wrote On Nov 19, 2012, at 9:11 PM, deniz lt; denizdurmus87@ gt; wrote: so in case i use ramdir with 5x cloud, it will still not do the recovery? i mean it will not get the data from the leader and fill its ramdir again? Yes, in 5x RAM directory should be able to recover. - Mark thank you so much for your patience with me :) - Zeki ama calismiyor... Calissa yapar... -- View this message in context: http://lucene.472066.n3.nabble.com/SolrCloud-Error-after-leader-restarts-tp4020985p4021209.html Sent from the Solr - User mailing list archive at Nabble.com.
SolrCloud Error after leader restarts
Hello, for test purposes, I am running two zookeepers on ports 2181 and 2182. and i have two solr instances running on different machines... For the one which is running on my local and acts as leader: java -Dbootstrap_conf=true -DzkHost=localhost:2181 -jar start.jar and for the one which acts as follower, on a remote machine: java -Djetty.port=7574 -DzkHost=address-of-mylocal:2182 -jar start.jar until this point everything is smooth and i can see the configs on both zookeeper hosts when i connect with zkCli.sh. just to see what happens and check recovery stuff, i have killed the solr which is running on my local and tried to index some files by using the follewer, which was failed... this is normal as writes are routed into the leader... the point that i dont understand is here: when i restart the leader with the same command on terminal, after normal logs, it start showing this Nov 19, 2012 2:15:18 PM org.apache.solr.common.SolrException log SEVERE: SnapPull failed :org.apache.solr.common.SolrException: Index fetch failed : at org.apache.solr.handler.SnapPuller.fetchLatestIndex(SnapPuller.java:400) at org.apache.solr.handler.ReplicationHandler.doFetch(ReplicationHandler.java:297) at org.apache.solr.cloud.RecoveryStrategy.replicate(RecoveryStrategy.java:151) at org.apache.solr.cloud.RecoveryStrategy.doRecovery(RecoveryStrategy.java:405) at org.apache.solr.cloud.RecoveryStrategy.run(RecoveryStrategy.java:220) Caused by: org.apache.lucene.index.IndexNotFoundException: no segments* file found in org.apache.lucene.store.RAMDirectory@1e75e89 lockFactory=org.apache.lucene.store.NativeFSLockFactory@128e909: files: [] at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:741) at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:630) at org.apache.lucene.index.SegmentInfos.read(SegmentInfos.java:343) at org.apache.lucene.index.IndexWriter.init(IndexWriter.java:639) at org.apache.solr.update.SolrIndexWriter.init(SolrIndexWriter.java:75) at org.apache.solr.update.SolrIndexWriter.create(SolrIndexWriter.java:62) at org.apache.solr.update.DefaultSolrCoreState.createMainIndexWriter(DefaultSolrCoreState.java:191) at org.apache.solr.update.DefaultSolrCoreState.getIndexWriter(DefaultSolrCoreState.java:77) at org.apache.solr.handler.SnapPuller.fetchLatestIndex(SnapPuller.java:354) ... 4 more Nov 19, 2012 2:15:18 PM org.apache.solr.common.SolrException log SEVERE: Error while trying to recover:org.apache.solr.common.SolrException: Replication for recovery failed. at org.apache.solr.cloud.RecoveryStrategy.replicate(RecoveryStrategy.java:154) at org.apache.solr.cloud.RecoveryStrategy.doRecovery(RecoveryStrategy.java:405) at org.apache.solr.cloud.RecoveryStrategy.run(RecoveryStrategy.java:220) it fails to recover after shutdown... why does this happen? - Zeki ama calismiyor... Calissa yapar... -- View this message in context: http://lucene.472066.n3.nabble.com/SolrCloud-Error-after-leader-restarts-tp4020985.html Sent from the Solr - User mailing list archive at Nabble.com.