[ 
https://issues.apache.org/jira/browse/HADOOP-2251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12544917
 ] 

Jim Kellerman commented on HADOOP-2251:
---------------------------------------

This problem happened in Hadoop-Patch build 1143

{code}
    [junit] 2007-11-23 01:23:19,499 INFO  [HMaster] 
org.apache.hadoop.hbase.HMaster$PendingServerShutdown.process(HMaster.java:2004):
 process shutdown of server 140.211.11.75:52601: logSplit: true, rootChecked: 
true, rootRescanned: false, numberOfMetaRegions: 1, onlineMetaRegions.size(): 1
    [junit] 2007-11-23 01:23:19,500 DEBUG [HMaster] 
org.apache.hadoop.hbase.HMaster$PendingServerShutdown.process(HMaster.java:2052):
 process server shutdown scanning root region cancelled because 
rootRegionLocation is null
    [junit] 2007-11-23 01:23:19,500 DEBUG [HMaster] 
org.apache.hadoop.hbase.HMaster.run(HMaster.java:1098): Put 
PendingServerShutdown of 140.211.11.75:52601 back on queue
    [junit] 2007-11-23 01:23:19,501 DEBUG [HMaster] 
org.apache.hadoop.hbase.HMaster.run(HMaster.java:1082): Main processing loop: 
PendingServerShutdown of 140.211.11.75:52601
    [junit] 2007-11-23 01:23:19,501 INFO  [HMaster] 
org.apache.hadoop.hbase.HMaster$PendingServerShutdown.process(HMaster.java:2004):
 process shutdown of server 140.211.11.75:52601: logSplit: true, rootChecked: 
true, rootRescanned: false, numberOfMetaRegions: 1, onlineMetaRegions.size(): 1
    [junit] 2007-11-23 01:23:19,502 DEBUG [HMaster] 
org.apache.hadoop.hbase.HMaster$PendingServerShutdown.process(HMaster.java:2052):
 process server shutdown scanning root region cancelled because 
rootRegionLocation is null
    [junit] 2007-11-23 01:23:19,503 DEBUG [HMaster] 
org.apache.hadoop.hbase.HMaster.run(HMaster.java:1098): Put 
PendingServerShutdown of 140.211.11.75:52601 back on queue
{code}


> [hbase] master won't go down if root region was not found
> ---------------------------------------------------------
>
>                 Key: HADOOP-2251
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2251
>             Project: Hadoop
>          Issue Type: Bug
>          Components: contrib/hbase
>            Reporter: stack
>            Priority: Minor
>
> psaab couldn't shut down his master... was getting reams of the below:
> {code}
> 2007-11-21 13:45:30,421 INFO  hbase.HMaster - process shutdown of server 
> 38.99.76.15:60020: logSplit: true, rootChecked: true, rootRescanned: false, 
> numberOfMetaRegions: 0, onlineMetaRegions.size(): 0
> 2007-11-21 13:45:30,421 DEBUG hbase.HMaster - process server shutdown 
> scanning root region cancelled because rootRegionLocation is null
> 2007-11-21 13:45:30,421 DEBUG hbase.HMaster - Put PendingServerShutdown of 
> 38.99.76.15:60020 back on queue
> 2007-11-21 13:45:30,421 DEBUG hbase.HMaster - Main processing loop: 
> PendingServerShutdown of 38.99.76.21:60020
> 2007-11-21 13:45:30,422 INFO  hbase.HMaster - process shutdown of server 
> 38.99.76.21:60020: logSplit: true, rootChecked: true, rootRescanned: false, 
> numberOfMetaRegions: 0, onlineMetaRegions.size(): 0
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - process server shutdown 
> scanning root region cancelled because rootRegionLocation is null
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - Put PendingServerShutdown of 
> 38.99.76.21:60020 back on queue
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - Main processing loop: 
> PendingServerShutdown of 38.99.76.31:60020
> 2007-11-21 13:45:30,422 INFO  hbase.HMaster - process shutdown of server 
> 38.99.76.31:60020: logSplit: true, rootChecked: true, rootRescanned: false, 
> numberOfMetaRegions: 0, onlineMetaRegions.size(): 0
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - process server shutdown 
> scanning root region cancelled because rootRegionLocation is null
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - Put PendingServerShutdown of 
> 38.99.76.31:60020 back on queue
> 2007-11-21 13:45:30,422 DEBUG hbase.HMaster - Main processing loop: 
> PendingServerShutdown of 38.99.76.17:60020
> ..
> {code}
> Looks like a shutdown soon after startup so should be reproducible.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to