On Mon,= Mar 4, 2013 at 2:44 PM, <roshanp@gmail.com> wrote:

Thanks Eric, Corey got it to work = easily with just the WRITE permission granted to root.

So if root does not have write permission to METADATA, then how does t= he accumulo server write to the METADATA table. Is there a server specific = user that is not visible to us?

Also, one thing we tried to do is to remove the link in= zookeeper from /accumulo/<instanceId>/tables/<tableId> and the= re was an authorization issue. Is there a specific zookeeper permission tha= t accumulo uses when connecting to the zookeepers?

Just curious :)

Roshan

On Mar 4, 2013, at 11:21 AM, Eric Newton= <eric.newton= @gmail.com> wrote:

Offline the table. =A0I= t may take some time for it to settle down. =A0Start any failed tservers.

Shutdown the accumulo =A0garbage collector:

=A0$ pkill -f =3Dgc

Grant the root user the write permissions the !METADATA= table:

shell> grant -u root -t !METADATA =A0Ta= ble.WRITE

Find your table id:

shell> tables -l

Now use th= e id to construct a delete command:

shell> dele= temany -c file -s id; -e id<

It's important that you use your table id, and= not "id" in the above command.
When you= get tired of typing yes to each file, stop the shell and re-run it like th= is:

shell> deletemany -f -c file -s id; -e id<<= /div>

Now, go and move the directory in hdfs:
=
$ hadoop fs -mv /accumulo/tables/id /files-from-dead-table

You can bulk import the directories in /files-from-dead= -table after you bring the table back online with some appropriate splits.<= /div>

The accumulo garbage collector will complain about= missing files, so expect those as warnings.

-Eric

On Mon, Mar 4, 2013 at 10:53 AM,= Corey Nolet <cnolet@texeltek.com> wrote:

We have a sharded-event table that failed mi= serably when we accidentally tried to merge all of the tablets together. Wh= en starting accumulo, the monitor page says the event table (once having 43= k tablets) now has 5 tablets and 1.05B rows. There are 14.5k unassigned tab= lets, The tablet servers each have response times ranging from 10s to 1m un= til eventually they all die. We thought that it may have been our ulimit on= the accumulo master being set to 1024 but raising it to 65535 didn't s= eem to have any immediate effects.

The Accumulo shell freezes when we try to drop the event table and we'v= e gotten a little experimental (trying to remove the references to the R fi= les in the !METADATA table manually, removing the reference to the event ta= ble in zookeeper), etc..) Our experiments have mostly ended in permissions = issues.

With that, do you guys have any good techniques/tools for unlinking a buste= d and unresponsive table? All of the other tables/tablets seem to be doing = just fine.

Thanks in advance!