Hi Mohammad,
Let me know the outcome of your experiments.
Best Regards,
Anil
On Dec 4, 2012, at 11:21 AM, Mohammad Tariq wrote:
> Hello Anil,
>
> I see. Your logic sounds appealing. I need to take a few test cases
> and test it properly. Thank you for the valuable respone.
>
> Regards,
>
Same answer ; just wait for the StochasticLoadBalancer (currently in trunk)
On Wed, Dec 5, 2012 at 7:05 AM, Stack wrote:
> On Tue, Dec 4, 2012 at 8:57 PM, Liu, Raymond
> wrote:
>
> > Hi
> >
> > I am just curious about the region assignment strategy upon table
> > re-enabling.
> >
I assume you are creating your PrefixFilter with something like:
new PrefixFilter(Bytes.toBytes(val));
In your case are you maybe passing a literal like so:
new PrefixFilter(Bytes.toBytes(42)); ?
If that case toBytes will encode an int. You have to do:
new PrefixFilter(Bytes.toBytes(42L));
Ju
On Tue, Dec 4, 2012 at 8:57 PM, Liu, Raymond wrote:
> Hi
>
> I am just curious about the region assignment strategy upon table
> re-enabling.
> By default, it's random assigned to the region server. I checked
> jira and found HBASE-6143 suggest to make it smarter like take load
>
Hi
I am just curious about the region assignment strategy upon table
re-enabling.
By default, it's random assigned to the region server. I checked jira
and found HBASE-6143 suggest to make it smarter like take load balancing into
account.
While, I am wondering what prev
One last question, Jean-Marc.
Exactly, What version of HBase are you using?
What version of Hadoop are you using?
On 12/04/2012 09:02 PM, Marcos Ortiz wrote:
Regards, Jean-Marc
On 12/04/2012 05:54 PM, Jean-Marc Spaggiari wrote:
Sorry for replying so quickly to myself.
So, here is what I did.
Regards, Jean-Marc
On 12/04/2012 05:54 PM, Jean-Marc Spaggiari wrote:
Sorry for replying so quickly to myself.
So, here is what I did.
I had a table with only "few" lines". about 20 000.
Table was split over 16 regions.
I merged all the regions into one, then asked HBase via the HTML
interfac
Sorry for replying so quickly to myself.
So, here is what I did.
I had a table with only "few" lines". about 20 000.
Table was split over 16 regions.
I merged all the regions into one, then asked HBase via the HTML
interface to split it until I got more than 64 regions.
Then I tried to re-merg
Hi,
While merging many regions, I'm getting this error for some of them:
12/12/04 17:45:16 FATAL util.Merge: Merge failed
java.io.IOException: Files have same sequenceid: 75533866
at org.apache.hadoop.hbase.regionserver.HRegion.merge(HRegion.java:4080)
at org.apache.hadoop.hbase.u
Forcing hbase.master.loadbalance.bytable=true seems to have resolved
my balance issue. Thanks again.
Norbert
On Tue, Dec 4, 2012 at 2:56 PM, Jimmy Xiang wrote:
> Right, that's config. You can turn it on and restart the cluster.
>
> Upstream it is on by default. However, it is turned off by de
Oh! that's so easy! Thanks. I don't know how I missed it. :(
And from the table.jsp code I figured that the Master is used only to
retreive the number of requests for this region, which I don't need.
Thanks,
JM
2012/12/4,ramkrishna vasudevan
> Hi
> Actually you can try using HTable.getRegionLoc
Right, that's config. You can turn it on and restart the cluster.
Upstream it is on by default. However, it is turned off by default in
CDH4.1.2 to be backward compatible.
Thanks,
Jimmy
On Tue, Dec 4, 2012 at 11:28 AM, Norbert Burger
wrote:
> Thanks, Jimmy. Do you mean the config
> "hbase.ma
Thanks for the information.
Please turn on per-table load balancing and tell us what it looks like.
On Tue, Dec 4, 2012 at 11:32 AM, Norbert Burger wrote:
> On Tue, Dec 4, 2012 at 2:10 PM, Ted Yu wrote:
> > Can you give us a little more detail on how much deviation the region
> > counts on regi
On Tue, Dec 4, 2012 at 2:10 PM, Ted Yu wrote:
> Can you give us a little more detail on how much deviation the region
> counts on region servers have ?
>
> There is a parameter, hbase.regions.slop, with default value of 0.2
> This parameter allows region count to deviate by certain percentage from
Thanks, Jimmy. Do you mean the config
"hbase.master.loadbalance.bytable"? According to [1] and [2], it is
"true" by default.
[1] https://issues.apache.org/jira/secure/attachment/12509174/3373.txt
[2] http://search-hadoop.com/m/M6z7G1PKejw
Norbert
On Tue, Dec 4, 2012 at 2:23 PM, Jimmy Xiang wr
In CDH4.1.2, per table region balancing is turned off by default.
You can change the configuration to turn it on.
Thanks,
Jimmy
On Tue, Dec 4, 2012 at 11:10 AM, Ted Yu wrote:
> Can you give us a little more detail on how much deviation the region
> counts on region servers have ?
>
> There is a
Hello Anil,
I see. Your logic sounds appealing. I need to take a few test cases
and test it properly. Thank you for the valuable respone.
Regards,
Mohammad Tariq
On Tue, Dec 4, 2012 at 11:01 PM, anil gupta wrote:
> Hi Mohammad,
>
> Suppose your rowkey is 345346654. Are you trying to
Can you give us a little more detail on how much deviation the region
counts on region servers have ?
There is a parameter, hbase.regions.slop, with default value of 0.2
This parameter allows region count to deviate by certain percentage from
average region count.
You can tighten the value for th
Hi Mohammad,
Suppose your rowkey is 345346654. Are you trying to match this row on the
basis of prefix 3453?
If yes, then the default PrefixFilter wont work. I am a little skeptic that
prefix like this will work because AFAIK byte[] of 3453 is not guaranteed
to be a prefix of byte[] of 345346654.
We upgraded to CDH4.1.2 (contains HBASE-3373) in one of our
environments. After filling that environment with data, I was
surprised to see that regions were not balanced across regionservers
at the table level. We have restarted all regionservers at least once
here.
In [1], I see Stack's referen
>From the master node hbase master stop should also flush everything for
shutdown.
On Tue, Dec 4, 2012 at 10:22 AM, ramkrishna vasudevan <
ramkrishna.s.vasude...@gmail.com> wrote:
> HBaseAdmin has a flush() api that accepts the table name. This will ensure
> that the data in memory is flushed.
>
HBaseAdmin has a flush() api that accepts the table name. This will ensure
that the data in memory is flushed.
Regards
Ram
On Tue, Dec 4, 2012 at 8:39 PM, Shengjie Min wrote:
> According to Hbase design, Memstore flush is happened automatically behind
> the theme when it reaches the size limit
Thank you for the quick response sir. I was thinking about the same thing,
but just wanted to ask on the list once before moving forward. I'll try it
out.
Regards,
Mohammad Tariq
On Tue, Dec 4, 2012 at 8:02 PM, ramkrishna vasudevan <
ramkrishna.s.vasude...@gmail.com> wrote:
> Ok, what i th
Ok, what i think can be done is try writing a custom filter like
PreFixFilter and use a Comparator that compares long.
In case of SCVF it has a comparator passed to it. So we can implement a
comparator that compares long and pass it to the constructor of SCVF.
Hope this helps.
Regards
Ram
On T
Oh! I tried... But I did a mistake. I missed the "hadoop" part of your
path and tried with the hbase one... Since I also had my libsnappy
copied there, this will have worked too...
So this is another totally valid option to achieve the install.
JM
2012/12/4, surfer :
> good ! never surrender !
>
You have a lot of memory. Do you have some metrics of the failing
servers? Maybe there are "stuck" on a big carbage collector process?
Just my 2¢...
JM
2012/12/4, Arati Patro :
> Hi,
>
> I'm using hbase version 0.94.1 and hadoop version 1.0.3
>
> I'm running HBase + HDFS on a 4 node cluster (48
Hi All
Last week I got a chance to present the secondary indexing solution
what we have done in Huawei at the China Hadoop Conference. You can see the
presentation from
http://hbtc2012.hadooper.cn/subject/track4Anoop%20Sam%20John2.pdf
I would like to hear what others think on th
27 matches
Mail list logo
