[jira] [Created] (SOLR-5935) SolrCloud hangs under certain conditions

["Rafał Kuć (JIRA)" <ji...@apache.org>]

Rafał Kuć created SOLR-5935:
-------------------------------

             Summary: SolrCloud hangs under certain conditions
                 Key: SOLR-5935
                 URL: https://issues.apache.org/jira/browse/SOLR-5935
             Project: Solr
          Issue Type: Bug
    Affects Versions: 4.6.1
            Reporter: Rafał Kuć
            Priority: Critical


As discussed in a mailing list - let's try to find the reason while under certain conditions SolrCloud can hang.

I have an issue with one of the SolrCloud deployments. Six machines, a collection with 6 shards with a replication factor of 3. It all runs on 6 physical servers, each with 24 cores. We've indexed about 32 million documents and everything was fine until that point.

Now, during performance tests, we run into an issue - SolrCloud hangs
when querying and indexing is run at the same time. First we see a
normal load on the machines, than the load starts to drop and thread
dump shown numerous threads like this:

Thread 12624: (state = BLOCKED)
 - sun.misc.Unsafe.park(boolean, long) @bci=0 (Compiled frame; information may be imprecise)
 - java.util.concurrent.locks.LockSupport.park(java.lang.Object) @bci=14, line=186 (Compiled frame)
 - java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await() @bci=42, line=2043 (Compiled frame)
 - org.apache.http.pool.PoolEntryFuture.await(java.util.Date) @bci=50, line=131 (Compiled frame)
 - org.apache.http.pool.AbstractConnPool.getPoolEntryBlocking(java.lang.Object, java.lang.Object, long, java.util.concurrent.TimeUnit, org.apache.http.pool.PoolEntryFuture) @bci=431, line=281 (Compiled frame)
 - org.apache.http.pool.AbstractConnPool.access$000(org.apache.http.pool.AbstractConnPool, java.lang.Object, java.lang.Object, long, java.util.concurrent.TimeUnit, org.apache.http.pool.PoolEntryFuture) @bci=8, line=62 (Compiled frame)
 - org.apache.http.pool.AbstractConnPool$2.getPoolEntry(long, java.util.concurrent.TimeUnit) @bci=15, line=176 (Compiled frame)
 - org.apache.http.pool.AbstractConnPool$2.getPoolEntry(long, java.util.concurrent.TimeUnit) @bci=3, line=169 (Compiled frame)
 - org.apache.http.pool.PoolEntryFuture.get(long, java.util.concurrent.TimeUnit) @bci=38, line=100 (Compiled frame)
 - org.apache.http.impl.conn.PoolingClientConnectionManager.leaseConnection(java.util.concurrent.Future, long, java.util.concurrent.TimeUnit) @bci=4, line=212 (Compiled frame)
 - org.apache.http.impl.conn.PoolingClientConnectionManager$1.getConnection(long, java.util.concurrent.TimeUnit) @bci=10, line=199 (Compiled frame)
 - org.apache.http.impl.client.DefaultRequestDirector.execute(org.apache.http.HttpHost, org.apache.http.HttpRequest, org.apache.http.protocol.HttpContext) @bci=259, line=456 (Compiled frame)
 - org.apache.http.impl.client.AbstractHttpClient.execute(org.apache.http.HttpHost, org.apache.http.HttpRequest, org.apache.http.protocol.HttpContext) @bci=344, line=906 (Compiled frame)
 - org.apache.http.impl.client.AbstractHttpClient.execute(org.apache.http.client.methods.HttpUriRequest, org.apache.http.protocol.HttpContext) @bci=21, line=805 (Compiled frame)
 - org.apache.http.impl.client.AbstractHttpClient.execute(org.apache.http.client.methods.HttpUriRequest) @bci=6, line=784 (Compiled frame)
 - org.apache.solr.client.solrj.impl.HttpSolrServer.request(org.apache.solr.client.solrj.SolrRequest, org.apache.solr.client.solrj.ResponseParser) @bci=1175, line=395 (Interpreted frame)
 - org.apache.solr.client.solrj.impl.HttpSolrServer.request(org.apache.solr.client.solrj.SolrRequest) @bci=17, line=199 (Compiled frame)
 - org.apache.solr.client.solrj.impl.LBHttpSolrServer.request(org.apache.solr.client.solrj.impl.LBHttpSolrServer$Req) @bci=132, line=285 (Interpreted frame)
 - org.apache.solr.handler.component.HttpShardHandlerFactory.makeLoadBalancedRequest(org.apache.solr.client.solrj.request.QueryRequest, java.util.List) @bci=13, line=214 (Compiled frame)
 - org.apache.solr.handler.component.HttpShardHandler$1.call() @bci=246, line=161 (Compiled frame)
 - org.apache.solr.handler.component.HttpShardHandler$1.call() @bci=1, line=118 (Interpreted frame)
 - java.util.concurrent.FutureTask$Sync.innerRun() @bci=29, line=334 (Interpreted frame)
 - java.util.concurrent.FutureTask.run() @bci=4, line=166 (Compiled frame)
 - java.util.concurrent.Executors$RunnableAdapter.call() @bci=4, line=471 (Interpreted frame)
 - java.util.concurrent.FutureTask$Sync.innerRun() @bci=29, line=334 (Interpreted frame)
 - java.util.concurrent.FutureTask.run() @bci=4, line=166 (Compiled frame)
 - java.util.concurrent.ThreadPoolExecutor.runWorker(java.util.concurrent.ThreadPoolExecutor$Worker) @bci=95, line=1145 (Compiled frame)
 - java.util.concurrent.ThreadPoolExecutor$Worker.run() @bci=5, line=615 (Interpreted frame)
 - java.lang.Thread.run() @bci=11, line=724 (Interpreted frame)

I've checked I/O statistics, GC working, memory usage, networking and
all of that - those resources are not exhausted during the test.

Hard autocommit is set to 15 seconds with openSearcher=false and
softAutocommit to 4 hours. We have a fairly high query rate, but until
we start indexing everything runs smooth.

I've attached four thread dumps, stack_1 to stack_4. They were gathered incrementally - stack_1 and stack_2 are when Solr was still able to respond, stack_3 is Solr barely alive and stack_4 is Solr not responding at all.

If more information is needed I can provide those.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org