[
https://issues.apache.org/jira/browse/HBASE-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12870905#action_12870905
]
ryan rawson commented on HBASE-2606:
------------------------------------
these keys are in the 'middle file':
K: item,,1274235159640/info:server/9223372036854775807/DeleteColumn/vlen=0 V:
K: item,,1274235159640/info:server/1274556669196/Put/vlen=17 V:
10.211.55.4:60020
K:
item,,1274235159640/info:serverstartcode/9223372036854775807/DeleteColumn/vlen=0
V:
K: item,,1274235159640/info:serverstartcode/1274556669196/Put/vlen=8 V:
\x00\x00\x01\x28\xC1\x7E\xA2j
In this case we have a DeleteColumn on info:server and info:serverstartcode at
the timestamp "9223372036854775807" == Long.MAX_VALUE.
So any further updates would be masked by these deletes.
> NativeException: org.apache.hadoop.hbase.client.NoServerForRegionException:
> No server address listed in .META. for region item,,1274235159640
> ----------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-2606
> URL: https://issues.apache.org/jira/browse/HBASE-2606
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.20.4
> Environment: Linux x86_64, java 1.6.0_20, hbase 0.20.4, CDH3
> Reporter: Sam Pullara
> Fix For: 0.20.5
>
> Attachments: info.tar.gz
>
>
> Though the region is listed when you attempt to access it you get an error
> that it is not assigned to a region. In the log you find this repeating over
> and over:
> 2010-05-23 16:10:08,958 INFO org.apache.hadoop.hbase.master.ServerManager: 1
> region servers, 0 dead, average load 5.0
> 2010-05-23 16:10:09,519 INFO org.apache.hadoop.hbase.master.BaseScanner:
> RegionManager.rootScanner scanning meta region {server: 10.211.55.4:60020,
> regionname: -ROOT-,,0, startKey: <>}
> 2010-05-23 16:10:09,529 INFO org.apache.hadoop.hbase.master.BaseScanner:
> RegionManager.rootScanner scan of 1 row(s) of meta region {server:
> 10.211.55.4:60020, regionname: -ROOT-,,0, startKey: <>} complete
> 2010-05-23 16:10:52,178 INFO org.apache.hadoop.hbase.master.BaseScanner:
> RegionManager.metaScanner scanning meta region {server: 10.211.55.4:60020,
> regionname: .META.,,1, startKey: <>}
> 2010-05-23 16:10:52,188 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
> Current assignment of item,,1274235159640 is not valid; serverAddress=,
> startCode=0 unknown.
> 2010-05-23 16:10:52,193 INFO org.apache.hadoop.hbase.master.BaseScanner:
> RegionManager.metaScanner scan of 3 row(s) of meta region {server:
> 10.211.55.4:60020, regionname: .META.,,1, startKey: <>} complete
> 2010-05-23 16:10:52,193 INFO org.apache.hadoop.hbase.master.BaseScanner: All
> 1 .META. region(s) scanned
> 2010-05-23 16:10:53,139 INFO org.apache.hadoop.hbase.master.RegionManager:
> Assigning region item,,1274235159640 to hbase,60020,1274655730505
> 2010-05-23 16:10:53,143 INFO org.apache.hadoop.hbase.master.ServerManager:
> Processing MSG_REPORT_OPEN: item,,1274235159640 from
> hbase,60020,1274655730505; 1 of 1
> 2010-05-23 16:10:53,143 DEBUG org.apache.hadoop.hbase.master.HMaster:
> Processing todo: PendingOpenOperation from hbase,60020,1274655730505
> 2010-05-23 16:10:53,143 INFO
> org.apache.hadoop.hbase.master.RegionServerOperation: item,,1274235159640
> open on 10.211.55.4:60020
> 2010-05-23 16:10:53,146 INFO
> org.apache.hadoop.hbase.master.RegionServerOperation: Updated row
> item,,1274235159640 in region .META.,,1 with startcode=1274655730505,
> server=10.211.55.4:60020
> Doing a flush and a major_compaction of the table resets it. I have a VM
> snapshot that has the behavior and it is reproducible at will. Attached are
> the log files from the .META. region at jdcryans request.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
