hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stack <st...@duboce.net>
Subject Re: lost .META. table
Date Tue, 21 Sep 2010 23:25:37 GMT
Ok.  Thanks.  I don't know why its not assigning .META.  Trying
starting the cluster with a different hbase.rootdir then move the
-ROOT- from the new hbase.rootdir in place of the -ROOT- you currently
have.

St.Ack

On Tue, Sep 21, 2010 at 4:17 PM, Stack <stack@duboce.net> wrote:
> Please put the first 1k lines after startup up in pastebin.  The below
> is hard to read.  Its scanning the -ROOT- but there is no mention of
> .META. being assigned out.
>
> St.Ack
>
> On Tue, Sep 21, 2010 at 3:09 PM, Matt Corgan <mcorgan@hotpads.com> wrote:
>> I think this is the relevant section:
>>
>> 2010-09-21 15:42:11,986 INFO org.apache.hadoop.hbase.master.ServerManager:
>> Received start message from: HadoopNode41,60020,1285098132069
>> 2010-09-21 15:42:11,989 INFO org.apache.hadoop.hbase.master.ServerManager:
>> Received start message from: HadoopNode38,60020,1285098131987
>> 2010-09-21 15:42:11,989 DEBUG
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Updated ZNode
>> /hbase/rs/1285098132069 with data 10.214.7.223:60020
>> 2010-09-21 15:42:11,996 DEBUG
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Updated ZNode
>> /hbase/rs/1285098131987 with data 10.209.206.15:60020
>> 2010-09-21 15:42:13,001 DEBUG org.apache.hadoop.hbase.master.RegionManager:
>> Assigning for address: 10.209.206.15:60020, startcode: 1285098131987, load:
>> (requests=0, regions=0, usedHeap=32, maxHeap=5195): total nregions to
>> assign=1, regions to give other servers than this=0, isMetaAssign=true
>> 2010-09-21 15:42:13,001 DEBUG org.apache.hadoop.hbase.master.RegionManager:
>> Assigning address: 10.209.206.15:60020, startcode: 1285098131987, load:
>> (requests=0, regions=0, usedHeap=32, maxHeap=5195) 1 regions
>> 2010-09-21 15:42:13,001 INFO org.apache.hadoop.hbase.master.RegionManager:
>> Assigning region -ROOT-,,0 to HadoopNode38,60020,1285098131987
>> 2010-09-21 15:42:13,085 INFO org.apache.hadoop.hbase.master.ServerManager:
>> Processing MSG_REPORT_OPEN: -ROOT-,,0 from HadoopNode38,60020,1285098131987;
>> 1 of 1
>> 2010-09-21 15:42:13,130 WARN
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Failed to create out of
>> safe mode in ZooKeeper:
>> org.apache.zookeeper.KeeperException$NodeExistsException: KeeperErrorCode =
>> NodeExists for /hbase/safe-mode
>> 2010-09-21 15:42:13,130 INFO
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Node exists; just move
>> on
>> 2010-09-21 15:42:13,130 INFO org.apache.hadoop.hbase.master.RegionManager:
>> exiting safe mode
>> 2010-09-21 15:42:13,132 DEBUG
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: SetData of ZNode
>> /hbase/root-region-server with 10.209.206.15:60020
>> 2010-09-21 15:42:13,133 INFO org.apache.hadoop.hbase.master.BaseScanner:
>> RegionManager.rootScanner scanning meta region {server: 10.209.206.15:60020,
>> regionname: -ROOT-,,0, startKey: <>}
>> 2010-09-21 15:42:13,223 INFO org.apache.hadoop.hbase.master.BaseScanner:
>> RegionManager.rootScanner scan of 0 row(s) of meta region {server:
>> 10.209.206.15:60020, regionname: -ROOT-,,0, startKey: <>} complete
>> 2010-09-21 15:42:13,466 INFO org.apache.hadoop.hbase.master.ServerManager:
>> Received start message from: HadoopNode40,60020,1285098133450
>> 2010-09-21 15:42:13,468 DEBUG
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Updated ZNode
>> /hbase/rs/1285098133450 with data 10.192.66.240:60020
>> 2010-09-21 15:42:13,828 INFO org.apache.hadoop.hbase.master.ServerManager:
>> Received start message from: HadoopNode39,60020,1285098133654
>> 2010-09-21 15:42:13,830 DEBUG
>> org.apache.hadoop.hbase.zookeeper.ZooKeeperWrapper: Updated ZNode
>> /hbase/rs/1285098133654 with data 10.206.26.95:60020
>> 2010-09-21 15:43:10,256 INFO org.apache.hadoop.hbase.master.ServerManager: 4
>> region servers, 0 dead, average load 0.25
>> 2010-09-21 15:43:10,341 INFO org.apache.hadoop.hbase.master.BaseScanner:
>> RegionManager.rootScanner scanning meta region {server: 10.209.206.15:60020,
>> regionname: -ROOT-,,0, startKey: <>}
>> 2010-09-21 15:43:10,345 INFO org.apache.hadoop.hbase.master.BaseScanner:
>> RegionManager.rootScanner scan of 0 row(s) of meta region {server:
>> 10.209.206.15:60020, regionname: -ROOT-,,0, startKey: <>} complete
>> 2010-09-21 15:43:10,354 INFO org.apache.hadoop.hbase.master.BaseScanner: All
>> 0 .META. region(s) scanned
>> 2010-09-21 15:43:56,351 INFO org.apache.zookeeper.ZooKeeper: Initiating
>> client connection, connectString=10.249.107.223:2181 sessionTimeout=60000
>> watcher=org.apache.hadoop.hbase.client.HConnectionManager$ClientZKWatcher@72898540
>> 2010-09-21 15:43:56,352 INFO org.apache.zookeeper.ClientCnxn: Attempting
>> connection to server /10.249.107.223:2181
>> 2010-09-21 15:43:56,362 INFO org.apache.zookeeper.ClientCnxn: Priming
>> connection to java.nio.channels.SocketChannel[connected local=/
>> 10.249.107.223:56828remote=/10.249.107.223:2181]
>> 2010-09-21 15:43:56,380 INFO org.apache.zookeeper.ClientCnxn: Server
>> connection successful
>>
>>
>> On Tue, Sep 21, 2010 at 6:05 PM, Stack <stack@duboce.net> wrote:
>>
>>> Ok.  So .META. is there.  What is in your master log on startup?  Its
>>> supposed to assign out -ROOT- and .META. (when .META. is assigned
>>> it'll add row to -ROOT-).  I'd say hold off on any of the trickery
>>> described in the first mail till we see snippet from your master log.
>>> Lets figure why not assigning .META. before we do any messing.
>>>
>>> Thanks,
>>> St.Ack
>>>
>>> On Tue, Sep 21, 2010 at 2:52 PM, Matt Corgan <mcorgan@hotpads.com> wrote:
>>> > Thanks for responding Stack.  In a 4 slave cluster (rf=3) cluster we
>>> > disabled an hdfs/hbase node which also seemed to make a second
>>> regionserver
>>> > crash.  I'm not sure exactly what went wrong.
>>> >
>>> > There is data in the .META. directory
>>> >
>>> > [root@Hadoop Master(hadoopmaster):/mnt/hadoop-0.20.2] bin/hadoop fs -lsr
>>> > /hbase/.META.
>>> > drwxr-xr-x   - root supergroup          0 2010-09-20 16:44
>>> > /hbase/.META./1028785192
>>> > -rw-r--r--   3 root supergroup        932 2010-09-20 14:25
>>> > /hbase/.META./1028785192/.regioninfo
>>> > drwxr-xr-x   - root supergroup          0 2010-09-21 15:41
>>> > /hbase/.META./1028785192/info
>>> > -rw-r--r--   3 root supergroup       5018 2010-09-21 15:41
>>> > /hbase/.META./1028785192/info/4694688683869779325
>>> > -rw-r--r--   3 root supergroup          0 2010-09-21 14:48
>>> > /hbase/.META./1028785192/info/6832908118284006198
>>> > -rw-r--r--   3 root supergroup     228130 2010-09-21 14:44
>>> > /hbase/.META./1028785192/info/7773537252698623585
>>> > drwxr-xr-x   - root supergroup          0 2010-09-21 14:44
>>> > /hbase/.META./compaction.dir
>>> >
>>> >
>>> > but it appears the -ROOT- region is empty.  If i scan '-ROOT-' in the
>>> hbase
>>> > shell i get:
>>> >
>>> > hbase(main):001:0> scan '-ROOT-'
>>> > ROW                          COLUMN+CELL
>>> > 0 row(s) in 0.0840 seconds
>>> >
>>> >
>>> > even though it appears -ROOT- exists in hdfs.  Any idea why ROOT would
>>> not
>>> > get loaded from hdfs after a master restart?
>>> >
>>> > [root@Hadoop Master(hadoopmaster):/mnt/hadoop-0.20.2] bin/hadoop fs -lsr
>>> > /hbase/-ROOT-
>>> > drwxr-xr-x   - root supergroup          0 2010-09-21 14:50
>>> > /hbase/-ROOT-/70236052
>>> > -rw-r--r--   3 root supergroup        618 2010-09-21 14:50
>>> > /hbase/-ROOT-/70236052/.regioninfo
>>> > drwxr-xr-x   - root supergroup          0 2010-09-21 15:41
>>> > /hbase/-ROOT-/70236052/info
>>> > -rw-r--r--   3 root supergroup        489 2010-09-21 15:41
>>> > /hbase/-ROOT-/70236052/info/308400263036654323
>>> > -rw-r--r--   3 root supergroup        462 2010-09-21 13:43
>>> > /hbase/-ROOT-/70236052/info/8250773031047005400
>>> >
>>> >
>>> > I'll just follow your suggestion of generating a new .META. and copying
>>> > over.  Thanks,
>>> >
>>> > Matt
>>> >
>>> >
>>> > On Tue, Sep 21, 2010 at 5:20 PM, Stack <stack@duboce.net> wrote:
>>> >
>>> >> On Tue, Sep 21, 2010 at 1:07 PM, Matt Corgan <mcorgan@hotpads.com>
>>> wrote:
>>> >> > Is there a way to create a new .META. table if I lost it?  For
>>> example,
>>> >> I'm
>>> >> > getting this exception everywhere:
>>> >> >
>>> >> > org.apache.hadoop.hbase.TableNotFoundException: .META.
>>> (NativeException)
>>> >> >
>>> >>
>>> >> How'd you do that?  Thats pretty impressive.
>>> >>
>>> >> >
>>> >> > The add_table script doesn't appear to handle .META., giving this
>>> error:
>>> >> >
>>> >> > java.lang.IllegalArgumentException: Illegal first character <46>
at 0.
>>> >> > User-space table names can only start with 'word characters': i.e.
>>> >> > [a-zA-Z_0-9]: .META.
>>> >> >
>>> >> >
>>> >> > And the hbase shell will only create user-space table names.
>>>  Restarting
>>> >> the
>>> >> > hbase master also does not appear to re-create it.
>>> >> >
>>> >> > I'm using 0.20.6.
>>> >> >
>>> >> Whats the master log look like when you start up hbase?  Whats it say
>>> >> about missing .META.?
>>> >>
>>> >> If you do a
>>> >>
>>> >> $ bin/hadoop fs -lsr /hbase/.META.
>>> >>
>>> >> ... is there anything in your filesystem?
>>> >>
>>> >> If nothing there, one thing you could do is run an hbase elsewhere.
>>> >> Let it create a new "/hbase" dir and then copy the .META. table from
>>> >> here up to the filesystem with data (Run add_table.rb after .META. is
>>> >> in place to repopulate its rows).
>>> >>
>>> >> St.Ack
>>> >>
>>> >
>>>
>>
>

Mime
View raw message