hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sam William <sa...@stumbleupon.com>
Subject Re: Deleting a table when disable 'failed'
Date Fri, 30 Oct 2015 19:02:48 GMT

I was able to solve this problem with Stack's help. The thing is, znode under '/hbase/table94'
also had to be deleted before the master restart.  The table came back online and we were
able to disable and drop it.


> On Oct 29, 2015, at 13:13, Ted Yu <yuzhihong@gmail.com> wrote:
> Which release of hbase are you using ?
> Can you pastebin master log snippet related to this table ?
> Thanks
> On Thu, Oct 29, 2015 at 11:32 AM, Sam William <sampd@stumbleupon.com> wrote:
>> Hi,
>>  I have been trying to clean up of cluster  clearing out a lot of unused
>> tables.  There is this one table that had grown too big (4800 regions on a
>> 50 node cluster). The disable command failed. But is_enabled returns true
>> and is_disabled  returns false. However, when I try to disable the table
>> again, I get the TableNotEnabled exception.  I have cleared the znode under
>> /hbase/table and restarted the master.  The table is listed on the master's
>> UI with 0 online regions.  The znode for that table under /hbase/table94
>> says 'DISABLED'.  When I try to run hbck -repair, I see messages that say
>> these regions are not deployed on any region server and "Trying to assign
>> region". But it looks like it never succeeds in assigning any of the
>> regions.   I just want to get rid of this table and bring the cluster back
>> to a consistent state .  Is there way to do this without bringing the
>> cluster down ?

View raw message