Hi everybody:
I've got an index of 20300 documents, with a size of 1.4G. I'm using
nutch-0.8 and hadoop 0.4. To access this index I use two machines. In the
first one I store the physical index and the command: "./bin/nutch server
port index" is running. The second machine runs a Tomcat that transfers user
queries to the first machine and shows results to the end user. After
checking everything is correct and both machines communicate to each other
I launch Jmeter in order to stress the architecture. My problem is that 1%
of the queries produce an error in the first machine giving this trace:
2006-12-21 12:01:28,171 WARN ipc.Server - handler output error
java.lang.NullPointerException
at
org.mortbay.http.nio.SocketChannelOutputStream.write(SocketChannelOutputStream.java:94)
at
java.io.BufferedOutputStream.flushBuffer(BufferedOutputStream.java:65)
at java.io.BufferedOutputStream.flush(BufferedOutputStream.java:123)
at java.io.DataOutputStream.flush(DataOutputStream.java:106)
at org.apache.hadoop.ipc.Server$Handler.run(Server.java:487)
So, I dont know if this problem means that "./bin/nutch server port index"
can't handle a big number of concurrent queries... or if this problem can be
solved increasing the number of handlers (10 is the default) when creating
a new RPC.Server in org.apache.nutch.searcher.DistributedSearch$Server (line
72 server=RPC.getServer(bean, port, 10, true, conf))
Thanks in advance.
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Nutch-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/nutch-general
