Sean Mackrory commented on HDFS-13135:

So this test more or less reproduces what I was seeing. I'm still trying to get 
more info about the workload that did this, because it seems insane, but the 
same RPC client ID was appending to a file, and then deleting it, and upon 
starting back up we got a NullPointerException because there was a lease for an 
inode that didn't exist anymore.

I'm uncertain about whether or not the fix is correct here: a lot of the code 
this is calling is completely new to me, so it's entirely possible there are 
side effects I haven't considered (like whether or not this causes data that 
should not be cleaned up because it's needed for the s0 snapshot to get 

> Lease not deleted when deleting INodeReference
> ----------------------------------------------
>                 Key: HDFS-13135
>                 URL: https://issues.apache.org/jira/browse/HDFS-13135
>             Project: Hadoop HDFS
>          Issue Type: Bug
>            Reporter: Sean Mackrory
>            Assignee: Sean Mackrory
>            Priority: Major
>         Attachments: HDFS-13135.001.patch
> In troubleshooting an occurrence of HDFS-13115, it seemed that there was 
> another underlying root cause that should also be addressed. There was an 
> INodeReference that was deleted and the lease on it was not subsequently 
> deleted because it was never added to the reclaim context.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org

Reply via email to