hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bill Graham <billgra...@gmail.com>
Subject Re: Data is always written to one node
Date Tue, 15 Mar 2011 06:32:19 GMT
On Mon, Mar 14, 2011 at 8:54 PM, Stack <stack@duboce.net> wrote:
> On Mon, Mar 14, 2011 at 4:09 PM, Bill Graham <billgraham@gmail.com> wrote:
>> Anyway, it's been about a week and all regions for the table are still
>> on 1 node. I see messages like this in the logs every 5 minutes:
>>
>> 2011-03-14 15:59:03,148 INFO
>> org.apache.hadoop.hbase.master.LoadBalancer: Skipping load balancing.
>> servers=4 regions=62 average=15.5 mostloaded=16 leastloaded=16
>>
>
> That says that you have 4 servers and a total of 62 regions.  The
> regions are spread around the cluster at about 16 regions per server.
> Is this not the case?

Yes, that's the case. I was just including that point out that the
cluster is balanced w.r.t. regions per RS.

>
>
>> It seems the total regions are evenly balanced, but individual tables
>> are not.
>
> We don't pay attention to the region a table comes from, just the
> regions themselves.
>
>  HBASE-3586  which will make a showing in 0.90.2 should help.  It goes
> a bit of footwork to make sure we're more random than we currently
> are.

Thanks, that's exactly the issue I'm seeing.

>
> St.Ack
>

Mime
View raw message