Re: Socket timeout after making regions persistent in async way

Mon, 19 Dec 2016 11:08:14 -0800

Thank you for that... But all the this is showing me that the connection between the two has been severed. 

I'm a little more interested in what happens before that...
Maybe if this issue is consistently explicable then maybe run with a log-level=fine. 

--Udo


On 12/19/16 09:15, Thacker, Dharam wrote:


Yeah sure!
Client config looks simple. I have not enabled redundant copies. More region is of type REPLICATE_PERSISTENT on server side. Server is just initialized with –Xms = 20G and –Xmx = 30G with ConcMarkSweep GC along with few more tuning parameters which works descent though! 

*_Server logs:_*
[warn 2016/12/19 22:42:13.254 IST <ServerConnection on port 40404 Thread 2> tid=0x5b] Server connection from [identity(XXXXX(11700:loner):63652:3b290e18,connection=2; port=63657]: Unexpected _IOException_:
_java.net.SocketException_: Software caused connection abort: socket write error 

       at java.net.SocketOutputStream.socketWrite0(_Native Method_)
at java.net.SocketOutputStream.socketWrite(_SocketOutputStream.java:109_) 

       at java.net.SocketOutputStream.write(_SocketOutputStream.java:153_)
at org.apache.geode.internal.cache.tier.sockets.Message.flushBuffer(_Message.java:649_)
at org.apache.geode.internal.cache.tier.sockets.Message.sendBytes(_Message.java:627_)
at org.apache.geode.internal.cache.tier.sockets.ChunkedMessage.sendChunk(_ChunkedMessage.java:314_)
at org.apache.geode.internal.cache.tier.sockets.ChunkedMessage.sendChunk(_ChunkedMessage.java:322_)
at org.apache.geode.internal.cache.tier.sockets.command.PutAll80.writeReply(_PutAll80.java:388_)
at org.apache.geode.internal.cache.tier.sockets.command.PutAll80.cmdExecute(_PutAll80.java:339_)
at org.apache.geode.internal.cache.tier.sockets.BaseCommand.execute(_BaseCommand.java:147_)
at org.apache.geode.internal.cache.tier.sockets.ServerConnection.doNormalMsg(_ServerConnection.java:783_)
at org.apache.geode.internal.cache.tier.sockets.ServerConnection.doOneMessage(_ServerConnection.java:913_)
at org.apache.geode.internal.cache.tier.sockets.ServerConnection.run(_ServerConnection.java:1180_)
at java.util.concurrent.ThreadPoolExecutor.runWorker(_ThreadPoolExecutor.java:1142_)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(_ThreadPoolExecutor.java:617_)
at org.apache.geode.internal.cache.tier.sockets.AcceptorImpl$1$1.run(_AcceptorImpl.java:546_) 

       at java.lang.Thread.run(_Thread.java:745_)

*_Client config:_*
<gfe:pool id="event-geode-pool" subscription-enabled="true" server-group="EVENT_GRP"> 

<gfe:localtor...>

</gfe:pool>

<gfe:client-region id="Event"

cache-ref="event-cache"

pool-name="event-geode-pool"

shortcut="PROXY"

key-constraint="java.lang.String"

value-constraint="a.m.n.Event">

</gfe:client-region>

<gfe:cq-listener-container id="cqListenerContainer" cache="event-cache">
<gfe:listener ref="eventListener" query="SELECT * FROM /Event WHERE status='PENDING'"/> 

</gfe:cq-listener-container>

<bean id="eventListener" class="a.m.n.cqlisteners.MyEventListener"/>

Thanks,

Dharam

*From:*Udo Kohlmeyer [mailto:[email protected]]
*Sent:* Monday, December 19, 2016 10:26 PM
*To:* [email protected]
*Subject:* Re: Socket timeout after making regions persistent in async way

Hi there Dharam,
It is possible that persistence has an effect on the performance of the system. But generally socket timeouts occur due to other issues like GC or system load.
What is the server load around the time of the disconnection? Have you enabled redundant copies for your cq? 

Could you please provide your client configuration for this problem.
Maybe even a snippet of the server logs around the time of the client disconnection. (a minute either side of the client disconnection is a good start). 

--Udo

On 12/19/16 05:22, Thacker, Dharam wrote:

    Hi,

    Has anyone faced below exceptions while cqlistener thread is
    working on? It started coming once after I changes regions to be
    persistent. Am I missing anything related to socket timeout?

    <geode:disk-storeid=/"my_diskstore"/cache-ref=/"geodeCache"/

    compaction-threshold=/"50"/

    auto-compact=/"false"/

    allow-force-compaction=/"true"/

    max-oplog-size=/"512"/

    queue-size=/"10000"/

    time-interval=/"500"/

    write-buffer-size=/"65536"/

    disk-usage-warning-percentage=/"80"/

    disk-usage-critical-percentage=/"98"/>

    <geode:disk-dirlocation=/"mydata"/max-size=/"25600"//>

    </geode:disk-store>

    It tells me that server is unreachable but bouncing client
    connects to server again but after some interval of activity with
    server, again throws back below exception!

    [warn 2016/12/19 18:43:22.608 IST <cqListenerContainer-1>
    tid=0x6c] Pool unexpected socket timed out on client
    connection=Pooled Connection to XXXX:40404:
    Connection[XXXX:40404]@1743344121). Server unreachable: could not
    connect after 1 attempts

    [cqListenerContainer-1] ERROR
    o.s.d.g.l.a.ContinuousQueryListenerAdapter Listener execution
    failed...

    _org.apache.geode.cache.client.ServerConnectivityException_: Pool
    unexpected socket timed out on client connection=Pooled Connection
    to XXXX:40404: Connection[XXXX:40404]@1743344121). Server
    unreachable: could not connect after 1 attempts

           at
    
org.apache.geode.cache.client.internal.OpExecutorImpl.handleException(_OpExecutorImpl.java:819_)

           at
    
org.apache.geode.cache.client.internal.OpExecutorImpl.handleException(_OpExecutorImpl.java:603_)

           at
    
org.apache.geode.cache.client.internal.OpExecutorImpl.execute(_OpExecutorImpl.java:173_)

           at
    
org.apache.geode.cache.client.internal.OpExecutorImpl.execute(_OpExecutorImpl.java:110_)

           at
    org.apache.geode.cache.client.internal.PoolImpl.execute(_PoolImpl.java:697_)

           at
    org.apache.geode.cache.client.internal.PutAllOp.execute(_PutAllOp.java:110_)

           at
    
org.apache.geode.cache.client.internal.ServerRegionProxy.putAll(_ServerRegionProxy.java:643_)

           at
    
org.apache.geode.internal.cache.LocalRegion.basicPutAll(_LocalRegion.java:10166_)

           at
    org.apache.geode.internal.cache.LocalRegion.putAll(_LocalRegion.java:10101_)

           at
    org.apache.geode.internal.cache.LocalRegion.putAll(_LocalRegion.java:10113_)

           at
    
org.springframework.data.gemfire.GemfireTemplate.putAll(_GemfireTemplate.java:201_)

           ... 25 common frames omitted

    Thanks,

    Dharam

    This email is confidential and subject to important disclaimers
    and conditions including on offers for the purchase or sale of
    securities, accuracy and completeness of information, viruses,
    confidentiality, legal privilege, and legal entity disclaimers,
    available at http://www.jpmorgan.com/pages/disclosures/email
    <http://www.jpmorgan.com/pages/disclosures/email>
This email is confidential and subject to important disclaimers and conditions including on offers for the purchase or sale of securities, accuracy and completeness of information, viruses, confidentiality, legal privilege, and legal entity disclaimers, available at http://www.jpmorgan.com/pages/disclosures/email

Reply via email to