[
https://issues.apache.org/jira/browse/HBASE-1084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12658910#action_12658910
]
Andrew Purtell commented on HBASE-1084:
---------------------------------------
I don't disagree with Thibaut's point of view, but it's a question of how much
work it would be. Is it wise to bring down HDFS local and start working on it?
Maybe the problem is on the Datanode side of the connection. We can't say at
this point. Hacks to dfsclient would not be accepted by upstream if they just
work around a problem, so the point would then be moot, unless we're going to
fix anything anywhere in the DFS plumbing. I suggest that is out of scope.
Anyway, this issue is about working around what will hopefully be a temporary
condition. Issues about the underlying cause(s) can be filed upstream as more
is learned about the particulars of the problem(s).
> Reinitializable DFS client
> --------------------------
>
> Key: HBASE-1084
> URL: https://issues.apache.org/jira/browse/HBASE-1084
> Project: Hadoop HBase
> Issue Type: Improvement
> Components: io, master, regionserver
> Reporter: Andrew Purtell
> Fix For: 0.20.0
>
>
> HBase is the only long lived DFS client. Tasks handle DFS errors by dying.
> HBase daemons do not and instead depend on dfsclient error recovery
> capability, but that is not sufficiently developed or tested. Several issues
> are a result:
> * HBASE-846: hbase looses its mind when hdfs fills
> * HBASE-879: When dfs restarts or moves blocks around, hbase regionservers
> don't notice
> * HBASE-932: Regionserver restart
> * HBASE-1078: "java.io.IOException: Could not obtain block": allthough file
> is there and accessible through the dfs client
> * hlog indefinitely hung on getting new blocks from dfs on apurtell cluster
> * regions closed due to transient DFS problems during loaded cluster restart
> These issues might also be related:
> * HBASE-15: Could not complete hdfs write out to flush file forcing
> regionserver restart
> * HBASE-667: Hung regionserver; hung on hdfs: writeChunk,
> DFSClient.java:2126, DataStreamer socketWrite
> HBase should reinitialize the fs a few times upon catching fs exceptions,
> with backoff, to compensate. This can be done by making a wrapper around all
> fs operations that releases references to the old fs instance and makes and
> initializes a new instance to retry. All fs users would need to be fixed up
> to handle loss of state around fs wrapper invocations: hlog, memcache
> flusher, hstore, etc.
> Cases of clear unrecoverable failure (are there any?) should be excepted.
> Once the fs wrapper is in place, error recovery scenarios can be tested by
> forcing reinitialization of the fs during PE or other test cases.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
