AnsweredAssumed Answered

handling disk failure if this disk contains cldb.internal volume

Question asked by dimamah on Jul 2, 2013
Latest reply on May 19, 2016 by maprcommunity
Branched to a new discussion
I had experienced recently a disk failure in the cluster (M3) when unfortunately that disk contained the 
cldb.internal volume thus failing the CLDB. 

I had to shutdown the CLDB node and install a new CLDB on another node containing the cldb.internal volume.

To learn from this case I have few questions : 

 1. I want to minimize the down time to minimal in the future (hopefully not having to install a new CLDB). 
 How can i overcome this problem of failure in the disk containing the cldb.internal?
 2. I see that this internal volume isn't mounted at all times, is there a way to control on which disk it will reside?

**UPDATE** : 
I'll summarize the findings. 
The CLDB internal volume (CID1) is for all practical reasons the CLDB itself and thus if a disk containing this volume fails - the CLDB fails. 
The only way to overcome this (in M3) is to re-install the CLDB on a different node containing a replication of the internal volume. 
In addition, there is no way we can manually choose the location of the internal volume.