Re: more regionservers does not improve performance

Matt, Yes, I did. What I observed is that the map job proceeds about 3-4x faster for a while. But then I observed long pauses partway through the job, and overall run time was only reduced only modestly, way from 50 minutes to 40 minutes. Just to summarize the issue, my mapper jobs seem to

coprocessors and JMX

I have a Coprocessor, what's the best way to hook it into HBase's JMX setup so that I can get stats on the coprocessor? Thanks, Grant

Re: bulk load

Yep. Bulk-loads are an extremely useful way of loading data. That would be 2 jobs since those are 2 tables. For more info on bulk loading, seeŠ http://hbase.apache.org/book.html#arch.bulk.load On 10/14/12 10:58 AM, yutoo yanio yutoo.ya...@gmail.com wrote: hi i want to bulk load my data,

Hbase 0.92 Example

Could anyone provide me with an example on hbase 0.92 on getSum() function, bec i am unable to implement it. Thanks

Re: Hbase 0.92 Example

Hi Dalia, Which getSum() method are you talking about? Please be specific. Thanks, Anil Gupta On Sun, Oct 14, 2012 at 2:44 PM, Dalia Sobhy dalia.mohso...@hotmail.comwrote: Could anyone provide me with an example on hbase 0.92 on getSum() function, bec i am unable to implement it. Thanks

Re: Hbase 0.92 Example

Method in AggregateImplementation class. On 2012-10-14, at 10:05 PM, anil gupta anilgupt...@gmail.com wrote: Hi Dalia, Which getSum() method are you talking about? Please be specific. Thanks, Anil Gupta On Sun, Oct 14, 2012 at 2:44 PM, Dalia Sobhy dalia.mohso...@hotmail.comwrote:

Rest Client remote connection

Hi, I have yet another problem with HBase rest client/server thing. Through Java Rest Client, if I call scan function, an IOException occurs, saying scan request failed with 400. Here is the situation: I am running my application on my laptop, which has hbase 0.94.1. On the other hand,

Re: Hbase 0.92 Example

Hi anil, Whats R,S ? Sent from my iPhone On 2012-10-14, at 10:24 PM, anil gupta anilgupt...@gmail.com wrote: Hi Dalia, On the client-side you dont need to use that method directly. Instead, use the following method : org.apache.hadoop.hbase.client.coprocessor.AggregationClient.sum(byte[]

Re: Rest Client remote connection

First, if you don't need mismatched client and server versions of HBase you'll do yourself a favor by avoiding that - just as a general suggestion. Next, a 400 response means your request was not properly formatted. We will need more information to understand why. What was the request as sent by

Re: Rest Client remote connection

With a more detailed investigation, I found out that the problem was about setting a filter for the scan instance. In my application, I create FilterList and put a KeyOnlyFilter in it. Then I use Scan::setFilter(Filter) method. If I remove this statement, then there is no problem. Does Rest

Re: Rest Client remote connection

What REST does wrt. filters is pass through a filter specification from client to server. The filter API has been changing for each HBase major version so it hasn't made sense (yet) to make this a first class part of the REST representations, thus version independent. (Otherwise retired filter

Re: Rest Client remote connection

Thank you so much for your answers Andrew. Erman On 10/14/2012 4:28 PM, Andrew Purtell wrote: What REST does wrt. filters is pass through a filter specification from client to server. The filter API has been changing for each HBase major version so it hasn't made sense (yet) to make this a

Re: Hbase 0.92 Example

R- Cell value data type S- Promoted data type On Sun, Oct 14, 2012 at 3:39 PM, Dalia Sobhy dalia.mohso...@hotmail.comwrote: Hi anil, Whats R,S ? Sent from my iPhone On 2012-10-14, at 10:24 PM, anil gupta anilgupt...@gmail.com wrote: Hi Dalia, On the client-side you dont need to use

Re: more regionservers does not improve performance

It could be network bound, especially if you have decently size values (~500B+). HBase can be rough on the network because each value travels from client to regionserver, and then makes 2 additional network hops in the WAL, and then an additional 2 hops in the memstore flush, plus ongoing

Re: more regionservers does not improve performance

Sorry for jumping in late here. What's you compaction queue size over time? It might be that your IO system just cannot keep up with the load. HBase will buffer data in the memstore, but eventually this data has to make it to disk, then eventually you get a lot of storefiles that need to be

Re: more regionservers does not improve performance

Thanks Matt, I have 10 regions per regionserver (100 splits over 10 regionservers), and yes they all seem to almost stop at the same time. I'll try splitting the table into fewer regions as you suggest. Where do I set the memstore flush size? Sorry pretty new to this. Jon On Sun, Oct 14, 2012

RE: hmaster and regionserver died

Check your GC configurations. Seems to that a Full GC has happened and the Zookeeper thought that to be session expiry. Regards Ram -Original Message- From: Xiang Hua [mailto:bea...@gmail.com] Sent: Saturday, October 13, 2012 6:20 PM To: user@hbase.apache.org Subject: hmaster and

RE: Delete by timestamp?

Also just see the discussions over the JIRA which will help you to come out with more specific usecases that you want to implement. The example over there will surely help you out. Regards Ram -Original Message- From: Ted Yu [mailto:yuzhih...@gmail.com] Sent: Sunday, October 14, 2012

RE: Debugging Coprocessor code in Eclipse

Hi Anil We also do a lot of stuff with coprocessors MasterObservers, RegionObservers and WALObservers. Just start your master and RS in debug mode and connect remotely from eclipse. This should be fine. Whenever the code goes to the RegionObserver or any observers automatically you will be able