Thanks , this link is very useful.
to add to that, I just found that the "sar -g " command is very
helpful, it shows paging requests.
On Thu, Sep 8, 2011 at 9:02 AM, Jonathan Ellis wrote:
> What is your bottleneck?
>
> spyced.blogspot.com/2010/01/linux-performance-basics.html
>
> On Thu, Sep 8,
Turned out that wasn't a problem - I put some notes on the ticket.
On Sep 10, 2011, at 6:22 PM, Jeremy Hanna wrote:
> I tried looking through the source to see if the log statements would happen
> regardless but it doesn't look like it. Also I looked at one of the nodes
> via jmx and checked o
I tried looking through the source to see if the log statements would happen
regardless but it doesn't look like it. Also I looked at one of the nodes via
jmx and checked out the System.HintsColumnFamily and there is activity there.
So I created a ticket: https://issues.apache.org/jira/browse/
We just tried to disable hinted handoff by setting:
hinted_handoff_enabled: false
in all the nodes of our cluster and restarting them. When they come back up,
we continue to see things like this:
INFO [HintedHandoff:1] 2011-09-10 22:41:40,813 HintedHandOffManager.java (line
323) Started hinted h
Thanks for the insights. I may first try disabling hinted handoff for one run
of our data pipeline and see if it exhibits the same behavior. Will post back
if I see anything enlightening there.
On Sep 10, 2011, at 5:04 PM, Chris Goffinet wrote:
> You could tail the commit log with `strings` t
You could tail the commit log with `strings` to see what keys are being
inserted.
On Sat, Sep 10, 2011 at 2:24 PM, Jonathan Ellis wrote:
> Two possibilities:
>
> 1) Hinted handoff (this will show up in the logs on the sending
> machine, on the receiving one it will just look like any other write
Of course. Hoping one day I create an issue related to Ec2 that CAN
be reproduced...
https://issues.apache.org/jira/browse/CASSANDRA-3175
On Sat, Sep 10, 2011 at 10:10 PM, Jonathan Ellis wrote:
> Can you create a Jira ticket?
Two possibilities:
1) Hinted handoff (this will show up in the logs on the sending
machine, on the receiving one it will just look like any other write)
2) You have something doing writes that you're not aware of, I guess
you could track that down using wireshark to see where the write
messages a
ok, found it's due to yourkit instrumentation:
http://forums.yourkit.com/viewtopic.php?f=3&t=3299
On Fri, Sep 9, 2011 at 5:50 PM, Yang wrote:
> I have jna-3.2.7 in the classpath, but at the start of server messages, I see:
>
> INFO 17:48:12,321 Unable to link C library. Native methods will be di
Oh and we're running 0.8.4 and the RF is 3.
On Sep 10, 2011, at 3:49 PM, Jeremy Hanna wrote:
> In addition, the mutation stage and the read stage are backed up like:
>
> Pool NameActive Pending Blocked
> ReadStage32 773 0
> RequestRes
In addition, the mutation stage and the read stage are backed up like:
Pool NameActive Pending Blocked
ReadStage32 773 0
RequestResponseStage 0 0 0
ReadRepairStage 0 0 0
Mut
We are experiencing massive writes to column families when only doing reads
from Cassandra. A set of 5 hadoop jobs are reading from Cassandra and then
writing out to hdfs. That is the only thing operating on the cluster. We are
reading at CL.QUORUM with hadoop and have written with CL.QUORUM.
Can you create a Jira ticket?
On Sat, Sep 10, 2011 at 6:52 AM, Sasha Dolgy wrote:
> maybe it's related to this ...
> https://issues.apache.org/jira/browse/CASSANDRA-3114
>
> odd thing is, we haven't "moved" to Ec2Snitch ... been using it for
> quite a long time now ...
>
> On Sat, Sep 10, 2011 at
Hi Peter,
Good call! I went and checked the seed, and indeed it left unchanged when we
copied the config yaml file from the first node to the second node. Thanks!
-- Y.
On Fri, Sep 9, 2011 at 7:47 PM, Peter Schuller
wrote:
> > cluster name for both machines. So in other words, if we want to l
For things like rolling restarts, we do:
disablethrift
disablegossip
(...wait for all nodes to see this node go down..)
drain
2011/9/10 Radim Kolar
> what is recommended node stop method. drain or kill Java process? i haven't
> seen anybody using drain in stop scripts yet
>
> If i kill Java pro
what is recommended node stop method. drain or kill Java process? i
haven't seen anybody using drain in stop scripts yet
If i kill Java process and log flush time is set to 10 seconds, will be
last 10 seconds lost?
maybe it's related to this ...
https://issues.apache.org/jira/browse/CASSANDRA-3114
odd thing is, we haven't "moved" to Ec2Snitch ... been using it for
quite a long time now ...
On Sat, Sep 10, 2011 at 1:42 PM, Sasha Dolgy wrote:
> Upgraded one ring that has four nodes from 0.8.0 to 0.8.5 with o
Upgraded one ring that has four nodes from 0.8.0 to 0.8.5 with only
one minor problem. It relates to Ec2Snitch when running a 'nodetool
ring' from two of the four nodes. the rest are all working fine:
Address DC Rack Status State Load Owns Token
14836224792726
Cassandra makes no attempt to provide cross-row isolation or atomicity.
(Within a row, it technically provides atomicity but not isolation.)
On Fri, Sep 9, 2011 at 9:33 PM, wrote:
> I a thinking about a scenario that goes like this: a node is reading a
> secondary index to reply to a select que
W=Q R=1 is not guaranteed to be consistent, no. you need R=W=Q for that.
On Fri, Sep 9, 2011 at 6:59 PM, Alaa Zubaidi wrote:
> Hi,
> Using 0.7.4, BOP, and RF=3, Write->QUORUM and Read->ONE "NO UPDATES"
> How does get_range_slices using Key Range works? Is it possible to get out
> of date list of
Cassandra ensures that data you put into a column matches its
definition, but if you change its definition later it assumes you know
what you're doing and doesn't scan the entire data set to revalidate.
You probably need to switch back to bytestype until you can fix that.
On Fri, Sep 9, 2011 at 6
21 matches
Mail list logo
