lei liu 2013-10-28, 08:57
-Re: Datanode fencing mechanism
Todd Lipcon 2013-10-29, 03:00
Hi Liu Le,
You're correct, that's an oversight that was designed but never
implemented. It's quite a rare circumstance but we should probably
implement the persistent promise as you suggested. Want to have a try at
making a patch for trunk?
On Mon, Oct 28, 2013 at 1:57 AM, lei liu <[EMAIL PROTECTED]> wrote:
> In https://issues.apache.org/jira/browse/HDFS-1972 jira, there is one
> Scenario 3: DN restarts during split brain period
> (this scenario illustrates why I think we need to persistently record the
> promise about who is active)
> - block has 2 replicas, user asks to reduce to 1
> - NN1 adds the block to DN1's invalidation queue, but it's backed up
> behind a bunch of other commands, so doesn't get issued yet.
> - Failover occurs, but NN1 still thinks it's active.
> - DN1 promises to NN2 not to accept commands from NN1. It sends an empty
> deletion report to NN2. Then, it crashes.
> - NN2 has received a deletion report from everyone, and asks DN2 to
> delete the block. It hasn't realized that DN1 is crashed yet.
> - DN2 deletes the block.
> - DN1 starts back up. When it comes back up, it talks to NN1 first
> (maybe it takes a while to connect to NN2 for some reason)
> - ** Now, if we had saved the "promise" as part of persistent state,
> we could ignore NN1 and avoid this issue. Otherwise:
> - NN1 still thinks it's active, and sends a command to DN1 to delete
> the block. DN1 does so.
> - We lost the bloc
> I am use the CDH4.3.1 version, and am reading the DataNode code. I don't
> find the DataNode to save the "promise" as part of persistent state. I
> want to know whether the case 3 is handled in CDH4.3.1 version. If the
> case is hadnled, where is the code?
Software Engineer, Cloudera