cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Coli <rc...@digg.com>
Subject Re: Gossip yoyo under write load
Date Mon, 15 Nov 2010 19:34:39 GMT
On 11/13/10 11:59 AM, Reverend Chip wrote:
>Swapping could conceivably be a
> factor; the JVM is 32G out of 72G, but the machine is 2.5G into swap
> anyway.  I'm going to disable swap and see if the gossip issues resolve.

Are you using JNA/memlock to prevent the JVM's heap from being swapped?

There's a log message about it at startup time if it fails, which looks 
like :

"
logger_.warn("Unable to lock JVM memory (ENOMEM)."
or
logger.warn("Unknown mlockall error " + errno(e));
"

Trunk also logs if it is successful :
"
logger.info("JNA mlockall successful");
"

See here for background :

https://issues.apache.org/jira/browse/CASSANDRA-1214

=Rob


Mime
View raw message