Spot on Aaron!
Of course when I set up the ring the aws nodes could see one another's
internal addresses so I could let the broadcast address default.
I've now used external addresses for all broadcast (and seed) addresses and
it all works fine.
[[ As a matter of interest, the adjusted values in
Depends on your traffic :-)
cassandra-env.sh will try to allocate heap with following formula if
you don't specify MAX_HEAP_SIZE.
1. calculate 1/2 of RAM on your system and cap to 1024MB
2. calculate 1/4 of RAM on your system and cap to 8192MB
3. pick the larger value
So how about to start with t
Hi Dan,
Thanks for answering.
I included the output of cfstats below.
I hope you can say something about our problems with it.
cheers,
Stefan
Keyspace: Keyspace1
Read Count: 60703419
Read Latency: 1.1790332096286043 ms.
Write Count: 105871791
Write Latency: 0.019847457393065166
I recently had to do some shuffling with one of my cassandra nodes because
it was running out of disk space. I did a few things in the process, and
I'm not sure in the end which caused my problem. First I added a second
file path to the data directory in cassandra.yaml. Things still worked
fine
On Wed, Feb 29, 2012 at 5:25 AM, Casey Deccio wrote:
> I recently had to do some shuffling with one of my cassandra nodes because
> it was running out of disk space. I did a few things in the process, and
> I'm not sure in the end which caused my problem. First I added a second
> file path to t
Using cassandra 1.0.7, I got the following, as I was trying to rebuild my
sstables:
$ nodetool -h localhost upgradesstables
Error occured while upgrading the sstables for keyspace MyKeySpace
java.util.concurrent.ExecutionException: java.lang.NullPointerException
at
java.util.concurrent.Fut
Any more feedback on larger deployments of 1.0.*?
We are eager to try out the new features in production, but don't want to
run into bugs as on former 0.7 and 0.8 versions.
Thanks,
Thibaut
On Tue, Jan 31, 2012 at 6:59 AM, Ben Coverston
wrote:
> I'm not sure what Carlo is referring to, but gen
I ran nodetool decommission on a node from a 4 nodes cluster as follows:
nodetool -h 192.168.26.130 decommission
But it never completes & doesn't show any messages, on seeing the logs
I noticed:
java.util.concurrent.RejectedExecutionException: ThreadPoolExecutor has
shut down
at
org.apache.cass
On Wed, Feb 29, 2012 at 5:29 AM, Casey Deccio wrote:
> On Wed, Feb 29, 2012 at 5:25 AM, Casey Deccio wrote:
>
>> I recently had to do some shuffling with one of my cassandra nodes
>> because it was running out of disk space. I did a few things in the
>> process, and I'm not sure in the end whic
On Wed, Feb 29, 2012 at 10:35 AM, Thibaut Britz
wrote:
> Any more feedback on larger deployments of 1.0.*?
>
> We are eager to try out the new features in production, but don't want to
> run into bugs as on former 0.7 and 0.8 versions.
>
> Thanks,
> Thibaut
>
>
>
> On Tue, Jan 31, 2012 at 6:59 AM,
On Sun, Feb 26, 2012 at 8:24 PM, aaron morton wrote:
> All nodes in the cluster need two way communication. Nodes need to talk to
> Gossip to each other so they know they are alive.
>
> If you need to dump a lot of data consider the Hadoop integration.
> http://wiki.apache.org/cassandra/HadoopSupp
Thanks!
We will test it on our test cluster in the coming weeks and hopefully put
it into production on our 200 node main cluster. :)
Thibaut
On Wed, Feb 29, 2012 at 5:52 PM, Edward Capriolo wrote:
> On Wed, Feb 29, 2012 at 10:35 AM, Thibaut Britz
> wrote:
> > Any more feedback on larger deplo
You might check out some of the stuff Netflix does with their Cassandra
backup, and Cassandra ETL tools.:
http://techblog.netflix.com/2012/02/aegisthus-bulk-data-pipeline-out-of.html
http://techblog.netflix.com/2012/02/announcing-priam.html
-Jeremiah
On 02/29/2012 11:04 AM, Alexandru Sicoe wro
First to be clear, I'm not an expert but I suggested "cfstats" because it
can show unhealthy signs. That said, yours looks okay to me: few live
SSTables per column family, reasonable quantity of data... The next things
I'd verify are that (1) the JVM isn't swapping out during these periods of
bad p
This is a good example.
https://gist.github.com/1847261
I couldn't make it work with DynamicComposite though.
Hi,
I have two questions on major compactions (the ones user initiate using
nodetool) and I really appreciate if someone can help.
1. I've noticed that when I run compactions the read latency improves even
more than I expected (which is good :) ) The improvement is so tempting
that I'd like to ru
Thanks a bunch.
On Wed, Feb 29, 2012 at 12:51 PM, juri wrote:
> This is a good example.
>
> https://gist.github.com/1847261
>
> I couldn't make it work with DynamicComposite though.
>
>
>
The error is that the specified colum family doesn't exist. If you
connect with the CLI and describe the keyspace does it show up? Also,
after adding a new column family programmatically you can't use it
immediately, you have to wait for it to propagate. You can use calls to
describe schema to d
DataStax has not recommend to run major compaction now:
http://www.datastax.com/docs/1.0/operations/tuning
But if you can afford it, major compaction will improve read latency as you see.
Major compaction is expensive, so you will not want to run it during
high traffic hours. And you should not
At this point, using LeveledCompaction is a much better way to have good
guarantees about how many sstables your reads will hit (and thus better
latency guarantees) than SizeTiered with periodic major compactions.
On Wed, Feb 29, 2012 at 8:49 PM, Maki Watanabe wrote:
> DataStax has not recommend
Thanks Maki and Tyler.
Re: Q1: I think its the time for me to think about LevelCompaction at this
time. But I'm happy to know I can run major compactions as often as I like
if I can afford.
Re: Q2: Other than the high IO impact if there won't be any data
corruption/consistency issues I think I ca
Jeremiah,
Thanks for the reply.
This is what we have been doing, but it's not reliable as we don't know a
definite time that the schema would get replicated. Is there any way I can
know for sure that changes have propagated.
Then I can block the insertion of data until then.
On Thu, Mar 1, 2012
22 matches
Mail list logo