But this
26.03.2010 22:29 schrieb am "Rob Coli" :
On 3/26/10 1:36 PM, Roland Hänel wrote:
>
> If I was going to write such a tool: do you think the th...
The JMX interface exposes an Attribute which seems appropriate to this use.
It is called "TotalDiskSpaceUsed," and is available on a per-column
Ack... very sorry. I read the original message too quickly.
The fact that neither read-repair nor anti-entropy are working is suspicious
though. Do you think you could paste your config somewhere?
-Original Message-
From: "Stu Hood"
Sent: Friday, March 26, 2010 11:57pm
To: user@cassandr
replication factor == 1 means that there is only one copy of the data. And you
deleted it. Repair depends on the replication factor being greater than 1.
-Original Message-
From: "Jianing Hu"
Sent: Friday, March 26, 2010 9:33pm
To: user@cassandra.apache.org
Subject: Re: Is ReplicationFac
That's not what I saw in my test. I'm probably making some noob
mistakes. Can someone enlighten me? Here's what I did:
1) Bring up a cluster with three servers cs1,2,3, with their initial
token set to 'foo3', 'foo6', and 'foo9', respectively.
ReplicationFactor is set to 2 on all 3.
2) Insert 9 colu
I have found the api.
On Sat, Mar 27, 2010 at 10:13 AM, Jeff Zhang wrote:
> Hi all,
>
> It seems the api of cassandra is a little different from hbase, I am
> looking for the api for list all the columns under one column family ? Is
> there any way to do this ? Thanks.
>
>
>
> --
> Best Regard
Hi all,
It seems the api of cassandra is a little different from hbase, I am looking
for the api for list all the columns under one column family ? Is there any
way to do this ? Thanks.
--
Best Regards
Jeff Zhang
Hi all,
I'd like to use Cassandra to store small files, so I wonder whether
Cassandra has size limitation on value ?
--
Best Regards
Jeff Zhang
Keys don't exist without columns in Cassandra. So the right answer
is, use get or get_slice to check for the column(s) that should be in
the key's row.
On Fri, Mar 26, 2010 at 7:56 PM, Jeff Zhang wrote:
> Hi all,
>
> I'd like to check whether one key exist, currently my solution is to let the
>
On 3/26/10 5:57 PM, Jianing Hu wrote:
In a cluster with ReplicationFactor> 1, if one server goes down, will
new replicas be created on other servers to satisfy the set
ReplicationFactor?
Yes, via Anti-Entropy.
http://wiki.apache.org/cassandra/AntiEntropy
http://wiki.apache.org/cassandra/Archi
In a cluster with ReplicationFactor > 1, if one server goes down, will
new replicas be created on other servers to satisfy the set
ReplicationFactor? I ran some tests that seem to suggest no new
replicas were created. Is that the expected behavior? If so, is there
a way to guarantee that any data a
Hi all,
I'd like to check whether one key exist, currently my solution is to let the
cassandra use the OrderPreservingPartitioner, and facilitate the
get_key_range() API to check key's existence. I wonder whether there's other
ways to do this ? Thanks.
--
Best Regards
Jeff Zhang
The quickest solution is definitely going to be just blowing away the
Hint files from the system keyspace data directories.
On Fri, Mar 26, 2010 at 5:36 PM, Scott White wrote:
> Nope it's always been random.
>
> On Fri, Mar 26, 2010 at 2:13 PM, Jonathan Ellis wrote:
>>
>> Did you switch partitio
Nope it's always been random.
On Fri, Mar 26, 2010 at 2:13 PM, Jonathan Ellis wrote:
> Did you switch partitioner types at some point?
>
> On Fri, Mar 26, 2010 at 2:53 PM, Scott White wrote:
> > I don't know if this is from switching from 0.5 to 0.6-betarc3 just
> recently
> > or from doing a s
2010/3/26 Roland Hänel
> Jonathan,
>
> I agree with your idea about a tool that could 'propose' good token choices
> for optimal load-balancing.
>
> If I was going to write such a tool: do you think the thrift API provides
> the necessary information? I think with the RandomPartitioner you cannot
On 3/26/10 1:36 PM, Roland Hänel wrote:
If I was going to write such a tool: do you think the thrift API
provides the necessary information? I think with the RandomPartitioner
you cannot scan all your rows to actually find out how big certain
ranges of rows are. And even with the OPP (that is the
Did you switch partitioner types at some point?
On Fri, Mar 26, 2010 at 2:53 PM, Scott White wrote:
> I don't know if this is from switching from 0.5 to 0.6-betarc3 just recently
> or from doing a series of bootstrap and removeToken operations but I
> recently started getting ArrayIndexOutOfBound
Jonathan,
I agree with your idea about a tool that could 'propose' good token choices
for optimal load-balancing.
If I was going to write such a tool: do you think the thrift API provides
the necessary information? I think with the RandomPartitioner you cannot
scan all your rows to actually find
Right, that's what I meant, thanks for the correction.
On Fri, Mar 26, 2010 at 1:11 PM, Brandon Williams wrote:
> On Fri, Mar 26, 2010 at 3:08 PM, Scott White wrote:
>
>> Yep I believe those are inserts per second. Take the last line:
>>
>> "811653,1666,250"
>>
>> I believe that's telling you t
On Fri, Mar 26, 2010 at 3:08 PM, Scott White wrote:
> Yep I believe those are inserts per second. Take the last line:
>
> "811653,1666,250"
>
> I believe that's telling you that during that 10 second interval you did
> 1666 inserts but your overall insert rate is 811653/250 = 3246.612
> inserts/s
Yep I believe those are inserts per second. Take the last line:
"811653,1666,250"
I believe that's telling you that during that 10 second interval you did
1666 inserts but your overall insert rate is 811653/250 = 3246.612
inserts/sec.
Timeouts may be due to your machine(s) being fully saturated?
Ok I ran the stress test with out of box settings -- 50 threads and 1M row
inserts. It seems to get as high as 4400 ops per second and as low as 968.
Am I reading these correctly as inserts per second?
These are results below. But is also generates timeouts and failures in the
python code like
I don't know if this is from switching from 0.5 to 0.6-betarc3 just recently
or from doing a series of bootstrap and removeToken operations but I
recently started getting ArrayIndexOutOfBoundsException exceptions (centered
around reading UTF from SSTableSliceIterator) on one of the machines in my
c
Priyanka,
I think our listserv might be dropping your attachment. Please send it
directly to me at benstande...@gmail.com.
-Ben
On Fri, Mar 26, 2010 at 10:41 AM, Priyanka Sharma wrote:
> Hi
>
> I am Priyanka Sharma, master student at Vrije University, Amsterdam. My
> major is "parallel and d
On Fri, Mar 26, 2010 at 10:45 AM, malcolm smith <
malsm...@treehousesystems.com> wrote:
> I've been getting a feel for the performance elements of Cassandra using
> version 0.51. I've done similar tests on HBase before, but Cassandra has
> some very appealing aspects that I would like to pursue.
No. The reason we're using mmap in the first place is that it's much
better at "allowing the OS to do the caching."
You just have too much data for the OS to cache effectively; making
Cassandra set that aside to cache key locations can help because it's
much more ram-efficient.
On Fri, Mar 26, 2
so just to close this out ... before mmap files, i would allow the OS to do the
caching using its I/O cache, but now since mmap files take up a majority of my
RAM, i need to cache more to maintain performance.
is that a fair statement?
From: Jonathan Ell
nodetool ring
http://wiki.apache.org/cassandra/NodeProbe
On Fri, Mar 26, 2010 at 10:37 AM, Y Aw wrote:
> Yes it does...
>
> Is there an easy way to know if a node is down or cannot reply to queries (a
> simple telnet command) ?
>
>
>
>
>
> 2010/3/25 Jeremy Dunck
>>
>> On Thu, Mar 25, 2010 at 1:
I've been getting a feel for the performance elements of Cassandra using
version 0.51. I've done similar tests on HBase before, but Cassandra has
some very appealing aspects that I would like to pursue.
However I'm not seeing the what seems like the common level of performance
others are seeing.
Yes it does...
Is there an easy way to know if a node is down or cannot reply to queries (a
simple telnet command) ?
2010/3/25 Jeremy Dunck
> On Thu, Mar 25, 2010 at 1:20 PM, Y Aw wrote:
> > Hi all,
> > I have a question about load-balancing.
>
> http://wiki.apache.org/cassandra/FAQ#node_c
On Fri, Mar 26, 2010 at 7:40 AM, Henrik Schröder wrote:
> For each indexvalue we insert a row where the key is indexid + ":" +
> indexvalue encoded as hex string, and the row contains only one column,
> where the name is the object key encoded as a bytearray, and the value is
> empty.
It's a uniq
>
> So all the values for an entire index will be in one row? That
> doesn't sound good.
>
> You really want to put each index [and each table] in its own CF, but
> until we can do that dynamically (0.7) you could at least make the
> index row keys a tuple of (indexid, indexvalue) and the column n
Different subnet isn't a problem from Cassandra's point of view, but
it might be if your network is doing something funky.
Did you check the logs on one of the machines that isn't in the ring?
On Fri, Mar 26, 2010 at 4:17 AM, Jeff Zhang wrote:
> Hi all,
>
> I have 6 different machines, 4 are in
On Fri, Mar 26, 2010 at 11:12 AM, ROGER PUIG GANZA wrote:
> Hi everybody.
>
> When will Cassandra 0.6.0 be released aproximately?
>
Within a couple of weeks (hopefully)
> Is it fine developing with the current beta release and put it on
> production when the final stable version is out?
>
Ab
Thanks! This has solved the problem!
On Wed, Mar 24, 2010 at 5:50 PM, gabriele renzi wrote:
> On Wed, Mar 24, 2010 at 3:36 PM, Oleg Mürk wrote:
> > Hi Jonathan,
> >
> > On Wed, Mar 24, 2010 at 4:32 PM, Jonathan Ellis
> wrote:
> >>
> >> probably 0.5.1 is allowing an invalid query and erroring o
Hi everybody.
When will Cassandra 0.6.0 be released aproximately?
Is it fine developing with the current beta release and put it on production
when the final stable version is out?
Thanks
Roger Puig Ganza
Hi all,
I have 6 different machines, 4 are in one subnet and the other two are in
another subnet. The following is the ip address of the 6 machines.
10.148.219.12
10.148.219.15
10.148.219.11
10.148.219.71
10.148.224.199
10.148.224.194
I make the same configuration on each machine, and finally fo
36 matches
Mail list logo
