Re: truncate command fails
truncate will throw Unavailable when it times out waiting for one of the nodes to return. The full error for the UnavailableException should include some text saying how many messages it received. You can also check the logs on the other machines to see if they are erroring. Finally turn up the log to DEBUG on the node you are running the truncate on and watch for log lines from the MessagingService . Cheers - Aaron Morton Freelance Developer @aaronmorton http://www.thelastpickle.com On 8/02/2012, at 8:10 PM, Patrik Modesto wrote: On Tue, Feb 7, 2012 at 18:35, Brandon Williams dri...@gmail.com wrote: Make sure you have JNA, without it the cost of forking ln to snapshot is expensive. I assume JNA is enabled. From the output.log: INFO 10:16:20,725 JNA mlockall successful Regards, P.
truncate command fails
Hi, I've a strange problem with my test cluster. Calling truncate on a small ColumnFamily on idle cluster of 4 nodes returns UnavailableException after 10s. That 10s is set in rpc_timeout_in_ms. All nodes are up and running, here is nodetool ring: Address DC RackStatus State Load OwnsToken 148873535527910577765226390751398592512 10.0.18.129 GO RAC1Up Normal 7.93 GB 25.00% 21267647932558653966460912964485513216 10.0.18.99 GO RAC1Up Normal 7.44 GB 25.00% 63802943797675961899382738893456539648 10.0.18.98 GO RAC1Up Normal 7.11 GB 25.00% 106338239662793269832304564822427566080 10.0.18.87 GO RAC1Up Normal 7.78 GB 25.00% 148873535527910577765226390751398592512 #using Cassandra 0.8.9 on Debian Squeeze What may be the problem? Regards, Patrik
Re: truncate command fails
On Tue, Feb 7, 2012 at 2:17 AM, Patrik Modesto patrik.mode...@gmail.com wrote: Hi, I've a strange problem with my test cluster. Calling truncate on a small ColumnFamily on idle cluster of 4 nodes returns UnavailableException after 10s. That 10s is set in rpc_timeout_in_ms. Make sure you have JNA, without it the cost of forking ln to snapshot is expensive. -Brandon
Re: truncate command fails
On Tue, Feb 7, 2012 at 18:35, Brandon Williams dri...@gmail.com wrote: Make sure you have JNA, without it the cost of forking ln to snapshot is expensive. I assume JNA is enabled. From the output.log: INFO 10:16:20,725 JNA mlockall successful Regards, P.