[
https://issues.apache.org/jira/browse/HBASE-1503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12719252#action_12719252
]
ryan rawson commented on HBASE-1503:
------------------------------------
i have this patched in on my cluster, and I just got this:
2009-06-14 04:35:17,194 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
Looking for reference files in:
hdfs://borg13:9000/hbase-1304/table/375231875/default
2009-06-14 04:35:17,196 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
isReference: hdfs://borg13:9000/hbase-1304/table/375
231875/default/7596663133220302018
2009-06-14 04:35:17,196 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
table,\x00\x02\x90\xDF\x00\x00\x01\x11\xFCo\xDB\xC8\
x00\x23\xED\xEA,1244979274858/375231875 no longer has references to
table,\x00\x02\x90\xDF\x00\x00\x01\x11\xFCo\xDB\xC8\x00\x2
3\xED\xEA,1244979185927
2009-06-14 04:35:17,197 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
Looking for reference files in:
hdfs://borg13:9000/hbase-1304/table/375231875/default
2009-06-14 04:35:17,198 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
isReference: hdfs://borg13:9000/hbase-1304/table/375
231875/default/7596663133220302018
2009-06-14 04:35:17,198 DEBUG org.apache.hadoop.hbase.master.BaseScanner:
table,\x00\x02\x90\xDF\x00\x00\x01\x11\xFCo\xDB\xC8\
x00\x23\xED\xEA,1244979274858/375231875 no longer has references to
table,\x00\x02\x90\xDF\x00\x00\x01\x11\xFCo\xDB\xC8\x00\x2
3\xED\xEA,1244979185927
2009-06-14 04:35:17,199 INFO org.apache.hadoop.hbase.master.BaseScanner:
Deleting region table,\x00\x02\x90\xDF\x00\x00\x01\x1
1\xFCo\xDB\xC8\x00\x23\xED\xEA,1244979185927 (encoded=1429689300) because
daughter splits no longer hold references
2009-06-14 04:35:17,199 DEBUG org.apache.hadoop.hbase.regionserver.HRegion:
DELETING region hdfs://borg13:9000/hbase-1304/table/1429689300
the other side of the split still had references like so:
/hbase-1304/table/1765060841/default/2821551828116783466.1429689300
so, still something. or else i forgot to type 'ant'.
> hbase-1304 dropped updating list of store files on flush
> --------------------------------------------------------
>
> Key: HBASE-1503
> URL: https://issues.apache.org/jira/browse/HBASE-1503
> Project: Hadoop HBase
> Issue Type: Bug
> Reporter: stack
> Assignee: Jonathan Gray
> Fix For: 0.20.0
>
> Attachments: HBASE-1503-v1.patch, HBASE-1503-v2.patch
>
>
> Here is how current updateReaders is implemented in StoreScanner:
> {code}
> // Implementation of ChangedReadersObserver
> public void updateReaders() throws IOException {
> if (this.closing.get()) {
> return;
> }
> this.lock.writeLock().lock();
> try {
> // Could do this pretty nicely with KeyValueHeap, but the existing
> // implementation of this method only updated if no existing storefiles?
> // Lets discuss.
> return;
> } finally {
> this.lock.writeLock().unlock();
> }
> }
> {code}
> Fill in missing functionality.
> Count of store files changes on flush and compaction.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
