Short answer: no.

We're planning to fix that at some point, and as a stopgap you could backup
the namenode state (fsimage, edits) every once in a while, limiting the
damage, but as for the data that's lost - your only hope is to try and
extract it somehow from the dead disk.

Yoram


> -----Original Message-----
> From: Jagadeesh [mailto:[EMAIL PROTECTED] 
> Sent: Tuesday, October 24, 2006 11:57 PM
> To: [email protected]
> Subject: Hadoop Master Node failed
> 
> Hi,
> 
> I have a storage cluster setup with 30 nodes using Hadoop and 
> it was working
> fine without any issues. But the hard drive of the machine 
> running master
> node failed yesterday and in no way I can retrieve data from that.
> 
> I would like to know whether I can retrieve data using the 
> other 29 nodes
> that I have in the cluster.
> 
> Thanks
> Jugs
> 
> 
> 

Reply via email to