MSG_REPORT_CLOSE but region still thought to be out on closed server
--------------------------------------------------------------------
Key: HBASE-1007
URL: https://issues.apache.org/jira/browse/HBASE-1007
Project: Hadoop HBase
Issue Type: Bug
Environment: TRUNK on hadoop 0.18.x.
Reporter: stack
Fix For: 0.19.0
See how region streamitems,LE�,1227049197721 is opened then closed but
subsequent if I try to scan, I get stuck because no one is serving this region.
{code}
2008-11-18 15:00:01,967 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_SPLIT: streamitems,LE�,1226969493750:
streamitems,LE�,1226969493750 split; daughters: streamitems,LE�,1227049197721,
streamitems,O+O,1227049197721 from XX.XX.XX.104:60020
2008-11-18 15:00:16,967 INFO org.apache.hadoop.hbase.master.RegionManager:
assigning region streamitems,O+O,1227049197721 to server XX.XX.XX.103:60020
2008-11-18 15:00:16,968 INFO org.apache.hadoop.hbase.master.RegionManager:
assigning region streamitems,LE�,1227049197721 to server XX.XX.XX.103:60020
2008-11-18 15:00:19,969 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_PROCESS_OPEN: streamitems,LE�,1227049197721 from
XX.XX.XX.103:60020
2008-11-18 15:00:19,970 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_OPEN: streamitems,O+O,1227049197721 from XX.XX.XX.103:60020
2008-11-18 15:00:19,970 INFO
org.apache.hadoop.hbase.master.ProcessRegionOpen$1:
streamitems,O+O,1227049197721 open on XX.XX.XX.103:60020
2008-11-18 15:00:19,970 INFO
org.apache.hadoop.hbase.master.ProcessRegionOpen$1: updating row
streamitems,O+O,1227049197721 in region .META.,,1 with startcode 1227047108563
and server XX.XX.XX.103:60020
2008-11-18 15:00:22,971 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_OPEN: streamitems,LE�,1227049197721 from XX.XX.XX.103:60020
2008-11-18 15:00:22,971 INFO
org.apache.hadoop.hbase.master.ProcessRegionOpen$1:
streamitems,LE�,1227049197721 open on XX.XX.XX.103:60020
2008-11-18 15:00:22,971 INFO
org.apache.hadoop.hbase.master.ProcessRegionOpen$1: updating row
streamitems,LE�,1227049197721 in region .META.,,1 with startcode 1227047108563
and server XX.XX.XX.103:60020
2008-11-18 15:09:20,870 INFO org.apache.hadoop.hbase.master.RegionManager:
Skipping region streamitems,LE�,1227049197721 because it is already closing.
2008-11-18 15:09:26,873 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_CLOSE: streamitems,LE�,1227049197721 from XX.XX.XX.103:60020
2008-11-18 15:10:32,547 INFO org.apache.hadoop.hbase.master.RegionManager:
Skipping region streamitems,O+O,1227049197721 because it is already closing.
2008-11-18 15:10:38,549 INFO org.apache.hadoop.hbase.master.ServerManager:
Received MSG_REPORT_CLOSE: streamitems,O+O,1227049197721 from XX.XX.XX.103:60020
{code}
Unfortunately, logs were at INFO level so can't see why its losing.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
