Hi Peter,
Thanks for the response.
I haven't explictly set a value for the memtable_flush_after_mins parameter.
Looks like the default is 60minutes.
I will try to play around this value to see if that fixes things.
Thanks again!
On Thu, May 12, 2011 at 11:41 AM, Peter Schuller
My cluster is unbalanced.One have 99 GB Data and other have 87 GB can
any one explain why this is happening.
[root@cassandra2 conf]# /root/cassandra/bin/nodetool -h 10.0.0.4 ring
Address Status Load
Range Ring
On 05/13/2011 11:46 AM, Ali Ahsan wrote:
My cluster is unbalanced.One have 99 GB Data and other have 87 GB can
any one explain why this is happening.
[root@cassandra2 conf]# /root/cassandra/bin/nodetool -h 10.0.0.4 ring
Address Status Load
Range
I haven't explictly set a value for the memtable_flush_after_mins parameter.
Looks like the default is 60minutes.
I will try to play around this value to see if that fixes things.
Is the amount of data in the commit log consistent with what you might
have been writing during 60 minutes?
It seems to work fine but maybe someone here knows that this is not a good
idea
It depends. If you're trying to figure out what generates actual iops
to disk, it won't really tell you since you're counting all I/O
including that which is cached. Since whatever goes down to disk tends
to
we are deploying cassandra over a huge cluster size and we think that inter
node communication can become overload on nodes as the protocol used is TCP.
Similarly gossip was earlier using udp but has been changed to use Tcp.
Is there added advantage for using tcp over udp in cassandra both in
Thanks - yes I agree. Didn't want to judge solely based on this figure.
It should just add to the picture. But since we know access patterns and other
stats like key and row cache hit ratios we hope to be able to make a more
educated guess whats going on.
On May 13, 2011, at 9:08 AM, Peter
What sort of schema changes are you making? can you manage them as a CLI
script under source control ?
You may also be interested in CASSANDRA-2221.
Cheers
Aaron
-
Aaron Morton
Freelance Cassandra Developer
@aaronmorton
http://www.thelastpickle.com
On 12 May 2011, at 20:45,
Hello,
I have written with a few other people a paper for the ACM Symposium
On Cloud Computing. This paper describes the methodology,
infrastructure and configuration used as well as the results obtained
for elasticity and scalability of three noSQL databases, of wich
Cassandra. The paper can be
The number of Completed HH tasks is interesting. AFAIK a task is started when
the node detects another in the cluster has returned. Were you doing some other
restarts around the cluster ?
I don't want to divert from the GC issue, just wondering if something else is
going on as well. Like the
nodetool move decommissions the node then bootstraps it back . I've not checked
the code but decommission will only send the data it needs to the other nodes,
and I believe bootstrap can do that same but I am not sure.
http://wiki.apache.org/cassandra/Operations#Moving_nodes
Aaron
On Fri, May 13, 2011 at 2:46 AM, Ali Ahsan ali.ah...@panasiangroup.com wrote:
My cluster is unbalanced.One have 99 GB Data and other have 87 GB can any
one explain why this is happening.
They are pretty close ... since a row key is pinned to a node - it
is possible that you have a really large
I think you may have a data centre defined in the strategy_options for the
keyspace that does not have any nodes in it. Check the dc names in the create
keyspace statement match the names in your cassandra-topology.propeties file
(assuming you are using the PropertyFileSnitch), or that you have
You need to manually assign the tokens to the nodes
http://wiki.apache.org/cassandra/Operations#Load_balancing
One should be 0 and the other 85070591730234615865843651857942052864
Cheers
-
Aaron Morton
Freelance Cassandra Developer
@aaronmorton
http://www.thelastpickle.com
On
HH will be stored into one of live replica node. It is just a hint,
rather than data to be replicated.
maki
2011/5/12 Anurag Gujral anurag.guj...@gmail.com:
Hi All,
I have two questions:
a) Is there a way to turn on and off hinted handoff per keyspace rather
than for multiple
Actually, I want a way to propagate *any* changes from development to
staging to production, but schema changes are the most important.
Could I use 2221 to propagate schema changes by deleting the schema in the
target cluster, doing show schema in the source cluster, redirecting to a
file, and
UDP is a bad fit for Cassandra; you'd have to implement your own
ordering and robustness on top of it. Better to just use TCP.
On Fri, May 13, 2011 at 2:55 AM, pankajsoni0126
pankajsoni0...@gmail.com wrote:
we are deploying cassandra over a huge cluster size and we think that inter
node
I wrote a small JMX client to invoke getNaturalEndpoints.
It works fine at my test environment, but throws NPE for keyspace we
will use for our application (both 0.7.5).
Does anyone know quick resolution of that before I setting up
cassandra on eclipse to inspect what happens :)
thanks
Exception
cassandra-people,
I'm trying to measure disk usage by cassandra after inserting some columns
in order to plan disk sizes and configurations for future deploys.
My approach is very straightforward:
clean_data (stop_cassandra rm -rf
/var/lib/cassandra/{dara,commitlog,saved_caches}/*)
I checked file cassandra-topology.properties everything looks good to me.
Please Advise on next steps I can do.
Thanks
Anurag
On Fri, May 13, 2011 at 3:52 AM, aaron morton aa...@thelastpickle.comwrote:
I think you may have a data centre defined in the strategy_options for the
keyspace that
On 5/13/11 10:08 AM, Maki Watanabe wrote:
I wrote a small JMX client to invoke getNaturalEndpoints.
It works fine at my test environment, but throws NPE for keyspace we
will use for our application (both 0.7.5).
Does anyone know quick resolution of that before I setting up
cassandra on eclipse
The number of Completed HH tasks is interesting. AFAIK a task is started when
the node detects another in the cluster has returned. Were you doing some
other restarts around the cluster ?
Not at all. The restarts seem to happen as normal operation.
I don't want to divert from the GC
Is there a way to look at the actual size of memtable? Would that help?
--
View this message in context:
http://cassandra-user-incubator-apache-org.3065146.n2.nabble.com/Commitlog-Disk-Full-tp6356797p6360001.html
Sent from the cassandra-u...@incubator.apache.org mailing list archive at
It is a 64 bit VM. I didn't notice the hs_err_pid.log files since I'm staring
over ssh, so they're in my home directory instead of my working directory.
I've attached one of those below. I don't know much about Java, so I'm not
sure how to interpret this file.
#
# An unexpected error
Hi All,
I have 3 separate cassandra clusters running in multiple data
centers which I want to convert to one cassandra cluster across multiple
data centers
Does anyone tried this? If so what are the steps:
I think I need to do the following:
a)Change cluster name: What is the
Has anyone seen this and know if it is causing an issue or how to fix
it? Anytime I run nodetool ring (on any node) I get this extra token
show up on the top of the list.
Address Status State LoadOwnsToken
113427455640312821154458202477256070484
10.1.2.171 Up
That's the same as the last one. The token space is a circle so the
last one at the list is repeated at the top.
-ryan
On Fri, May 13, 2011 at 9:59 AM, Scott McPheeters
smcpheet...@healthx.com wrote:
Has anyone seen this and know if it is causing an issue or how to fix
it? Anytime I run
Ahh, thanks. I got questioned on that as soon as our test environment came up
and I had no idea.
Thanks for the reply.
Scott
-Original Message-
From: Ryan King [mailto:r...@twitter.com]
Sent: Friday, May 13, 2011 1:03 PM
To: user@cassandra.apache.org
Subject: Re: Ghost token
I am using network Topology strategy for one of my keyspaces.
Thanks
Anurag
On Thu, May 12, 2011 at 6:45 PM, Anurag Gujral anurag.guj...@gmail.comwrote:
Hi All,
I run following command on one of my nodes to move the token
from 0 to 2.
/usr/cassandra/cassandra/bin/nodetool -h
sounds like https://issues.apache.org/jira/browse/CASSANDRA-2353
On Fri, May 13, 2011 at 10:08 AM, Maki Watanabe watanabe.m...@gmail.com wrote:
I wrote a small JMX client to invoke getNaturalEndpoints.
It works fine at my test environment, but throws NPE for keyspace we
will use for our
Hi,
I'm running Cassandra (0.7.4) on a 4 node ring. It was a 3 node ring, but we
ended up expanding it to 4... So then I followed the many suggestions to
rebalance the ring. I found a script that suggested I use:
# ~/nodes_calc.py
How many nodes are in your cluster? 4
node 0: 0
node 1:
I did not drop the keyspace, but your comment lead me to resolution.
I found cassandra-cli is not case sensitive on keyspace. I used
keyspace name FooBar on cassandra-cli, but it was Foobar in correct.
cassandra-cli didn't complain on my mistake, but the JMX interface is
less tolerance.
If I use
Hi All,
I have built a cassandra cluster which is using property snitch
file to specify my network topology to cassandra.The keyspace I have use
network topology strategy.
When I tried to add new node to this cluster,I get the following error:
RROR [main] 2011-05-13 23:45:44,152
You should upgrade to the latest Sun JVM. OpenJDK is almost a year
behind in bug fixes.
On Fri, May 13, 2011 at 11:40 AM, James Cipar jci...@cmu.edu wrote:
It is a 64 bit VM. I didn't notice the hs_err_pid.log files since I'm
staring over ssh, so they're in my home directory instead of my
our write happen in bursts. So often times, clients write data as fast as
they can. Conceivably one can write 5G in one hour.
The other setting that we have is that our replication factor is 3 and we
write using QUORUM. Not sure if that will affect things.
On Fri, May 13, 2011 at 12:04 AM, Peter
5G in one hour is actually very low. Something else is wrong. Peter pointed
to something related to memtable size could be causing this problem, can you
turn down memtable_throughput and see if that helps.
--
View this message in context:
36 matches
Mail list logo
