Re: Soliciting thoughts on possible read optimization

On Tue, Aug 10, 2010 at 8:43 PM, Arya Asemanfar wrote: > I mentioned this today to a couple folks at Cassandra Summit, and thought > I'd solicit some more thoughts here. > Currently, the read stage includes checking row cache. So if your concurrent > reads is N and you have N reads reading from di

Re: Soliciting thoughts on possible read optimization

On Wed, Aug 11, 2010 at 11:37 AM, Ryan King wrote: > On Tue, Aug 10, 2010 at 8:43 PM, Arya Asemanfar > wrote: >> I mentioned this today to a couple folks at Cassandra Summit, and thought >> I'd solicit some more thoughts here. >> Currently, the read stage includes checking row cache. So if your

File write errors but cassandra isn't crashing

Due to administrative error one of the hosts in the cluster lost permission to write to it's data directory. So I started seeing errors in the log, however, the server continued serving traffic. It wasn't able to compact and do other write operations but it didn't crash. I was wondering wether that

Re: Using a separate commit log drive was 4x slower

Thanks for all the feedback, I'll be back in 2 weeks and pick up then. -JD On Tue, Aug 10, 2010 at 3:45 PM, Peter Schuller wrote: > > Yeah, it has a BBU, and it is charged and on.. > > Very odd behavior, I'm stumped. > > I advise double-checking raid volume settings and ensuring that policy > is

the 0.7 beta process

Thanks for the 0.7 beta  1 release. I know this is a how long is a piece of string question, but anyone got a feel for how long the beta / release candidate process may take? According to git hub 0.6 beta 1 was tagged  on march 14 and 0.6 on april 12 I'm probably going to jump ship from 0.6 to 0.7

a plea not to remove rowsize warning

Hello all, I recently posted on list about a situation where two of my nodes from my 16 node were garbage collecting and at ooming. I was able to move my xmx from 9gb to 11gb to see that rather then my memory saw tooth. I would saw tooth around 4 gb before memory shot up like a rocket. After digg

whirr/cassandra support

Looks like Cassandra is now supported by the whirr incubator project now. Whirr is an apache incubator project that runs cloud services in a cloud provider (ec2, rackspace cloud, etc.) neutral way: https://cwiki.apache.org/confluence/display/WHIRR/Index One of the first use cases is Cassandra -

Corrupt file after restarting in 0.7

When I kill cassandra and restart I keep seeing the following error message. Is this something I should be concerned about? I though killing cassandra via kill $PID was somewhat "safe"??? Thanks Switching to ParallelGC to avoid CMS/CompressedOops incompatibility INFO 17:05:08,680 DiskAccessMod

0.7 CLI w/TSocket

"org.apache.thrift.protocol.TProtocolException: Missing version in readMessageBegin, old client?" Is the CLI not supported when using TSocket? I don't believe this was the same in 0.6. Can someone explain the differences between TFramedTransport vs TSocket. I tried searching but I couldn't f

Reload/Update schema 0.7

How is this accomplished? I tried using the org.apache.cassandra.service.StorageService.loadSchemaFromYAML() method but I am receiving the following error. java.util.concurrent.ExecutionException: org.apache.cassandra.config.ConfigurationException: Cannot load from XML on top of pre-existin

Re: Corrupt file after restarting in 0.7

On Wed, Aug 11, 2010 at 7:09 PM, Mark wrote: > When I kill cassandra and restart I keep seeing the following error > message. Is this something I should be concerned about? > No, there's just a race condition as evidenced by these lines: INFO 17:05:08,933 Deleted /var/lib/cassandra/data/system/

Re: Corrupt file after restarting in 0.7

On 8/11/10 8:02 PM, Brandon Williams wrote: On Wed, Aug 11, 2010 at 7:09 PM, Mark > wrote: When I kill cassandra and restart I keep seeing the following error message. Is this something I should be concerned about? No, there's just a race condition as

Re: Soliciting thoughts on possible read optimization

https://issues.apache.org/jira/browse/CASSANDRA-1379 On Tue, Aug 10, 2010 at 8:43 PM, Arya Asemanfar wrote: > I mentioned this today to a couple folks at Cassandra Summit, and thought > I'd solicit some more thoughts here. > Currently, the read stage includes checking row cache. So if your concur

Re: the 0.7 beta process

Seems like 0.7beta1 is of a similar baked-ness to 0.6beta1 to me. On Wed, Aug 11, 2010 at 3:29 PM, Aaron Morton wrote: > Thanks for the 0.7 beta  1 release. > I know this is a how long is a piece of string question, but anyone got a > feel for how long the beta / release candidate process may tak

Re: 0.7 CLI w/TSocket

you have to use an up to date CLI, the old one used broken options w/ its framed mode On Wed, Aug 11, 2010 at 6:39 PM, Mark wrote: > "org.apache.thrift.protocol.TProtocolException: Missing version in > readMessageBegin, old client?" > > Is the CLI not supported when using TSocket? I don't believe