> In this case should one set ulimit -l to the amount of heap size?
Yes, or under the assumption that you're running Cassandra dedicated
and are not worried about enforcement- just set it to unlimited or
just very high to avoid having to change it if you change the heap
size in the future.
--
/
Unsubscribe
Unsubscribe
unsubscribe
In this case should one set ulimit -l to the amount of heap size?
Thanks,
-Arya Goudarzi
- Original Message -
From: "Peter Schuller"
To: user@cassandra.apache.org
Sent: Saturday, October 9, 2010 1:18:28 AM
Subject: Re: using jna.jar "Unknown mlockall error 0"
> IIRC, mlockall doesn't wo
edit JMX_PORT in conf/cassandra-env.sh
On Thu, Oct 28, 2010 at 11:35 AM, Dmitri Smirnov wrote:
>
> I am running several things on my dev box and one of them takes port 8080
> which I am reluctant to reconfigure.
>
> Cassandra 0.7 trunk 10/14 complains that it need port 8080
>
> How do I tell it t
I am running several things on my dev box and one of them takes port
8080 which I am reluctant to reconfigure.
Cassandra 0.7 trunk 10/14 complains that it need port 8080
How do I tell it to use something else?
I found the below link that I do not have any of these config files,
shall I crea
On Thu, 2010-10-28 at 11:31 -0400, Jose Angel Inda Herrera wrote:
> i need know, cassandra is a DB of graph
No.
--
Eric Evans
eev...@rackspace.com
> Is that Casandra problem?
> I want Cassandra use memory more effective. How can I do that?
I definitely recommend Jonathan's link, but just to specifically
answer this: It's not a Cassandra issue, it's how the JVM will tend to
behave in particular with the CMS garbage collector. While under the
Maybe your7000 is being blocked by iptables
or some firewall or maybe you have it bound ( tag ) to
localhost instead an ip address.
Hope this helps,
Dimitry.
On Thu, Oct 28, 2010 at 5:35 PM, Thibaut Britz <
thibaut.br...@trendiction.com> wrote:
> Hi,
>
> I have the same problem with 0.6.5
Hey Aaron,
> Are there any log messages in the existing nodes or the new one which
> mention each other?
>From the currently running nodes we get the message that the new is up:
INFO [GMFD:1] 2010-10-22 10:22:49,232 Gossiper.java (line 591) Node /
192.168.2.18 is now part of the cluster
INFO [GM
Hi,
I have the same problem with 0.6.5
New nodes will hang forever in bootstrap mode (no streams are being opened)
and the receiver thread just waits for data forever:
INFO [Thread-53] 2010-10-27 20:33:37,399 SSTableReader.java (line 120)
Sampling index for /hd2/cassandra/data/table_xyz/
table
i need know, cassandra is a DB of graph
chees
Use bin/config-converter
2010/10/28 zangds
>
> rt
> 2010-10-28
> --
> zangds
>
--
Best regards.
Dbds
I believe the settings file is now cassandra.yaml, and no longer
storage-conf.xml. There's a conversion tool in the /bin folder that
should help you make the transition.
On Thu, Oct 28, 2010 at 7:03 AM, zangds wrote:
>
> rt
> 2010-10-28
>
> zangds
NEWS.txt
On Thu, Oct 28, 2010 at 4:03 PM, zangds wrote:
>
> rt
> 2010-10-28
>
> zangds
rt
2010-10-28
zangds
The main difficulty is that row mutations are currently represented
only in terms of individual columns. This has implications in the way
the transaction is carried out and represented in the commit log.
Introducing deletable slices is doable, but keeping it within the
semantics of column family o
Yes. I even tried just starting one node only, and then bootstrapping
another node. (However, at the beginning a few days ago, the cluster was
unstable and unresponsive and I had to restart the cluster. Maybe something
went wrong back then.)
Anyway, I will export all the data, and reimport it with
Hi Ryan
I took a sample of one sstable (just flushed, not compacted).
I compared 2 samples of sstables. One that is showing fine false positive
ratios and the problem one.
And yes both look the same to me. Both have the expected 15 buckets per row and
the cardinality of the bitsets are the sa
The best approach is to manually select the tokens, see the Load Balancing
section http://wiki.apache.org/cassandra/Operations Also
Are there any log messages in the existing nodes or the new one which mention
each other?
Is this a production system? Is it still running ?
Sorry there is not
It's unscheduled (https://issues.apache.org/jira/browse/CASSANDRA-494,
as you noted) because there's the obvious workaround of performing a
slice and then deleting each column in the slice. It could happen
during a 0.7 release since it wouldn't require API changes. 0.6 is
basically bugfix-only at
Memtable Data Size is the number of bytes you Column Family is taking up.
Have you read the section on "Repairing missing or inconsistent data" here
http://wiki.apache.org/cassandra/Operations . It describes the nodetool repair
operation used to repair inconsistent data.
Aaron
On 28 Oct 201
I cannot see anything in jira scheduled for 0.7 or 0.8 . Gary may be able to
shed some light on it.
There is a comment at the bottom of CASSANDRA-293 "Range operations cannot be
made to work with the existing implementation of hinted handoff. Further, they
were a bad fit for the transactional u
Gary, Thank you for your comments.
I also have another question in mind:
- If in all nodes "nodetool cfstats" shows that the memtable size is 0. Then
can I be sure that it's safe to assume that all values are consistent?
Regards,
Utku
On Wed, Oct 27, 2010 at 3:24 PM, Gary Dusbabek wrote:
> On
Here is my patch to solve this problem (it very stupid, bot works), please
use it only in test environment, not in production
2010/10/28 Chris Oei
> I guess so. I tried hacking a quick work-around for the "Filename must
> include parent directory", but I got another error (below).
>
> So, since
26 matches
Mail list logo
