Hi all,

When adding a cluster to our network, we are getting slow responses from 
what appears to be getdestiny/nextdestiny on our xCAT 2.3 cluster, and I 
was wondering if any enhancements / changes have been made in 2.6 to 
this behavior?

The steps of replicating the problem (and how we add a cluster):

* Rack and cable up, power on
* Run getmacs
* Add cluster nodes to other required tables (we have a custom script 
that does this)
* makedhcp for the cluster
* nodeset cluster runcmd=bmcsetup
* Power cycle nodes

At this point, as the nodes begin asking the management node for 
getdestiny/nextdestiny, the management node begins to become very slow 
in responding to the getdestiny/nextdestiny requests. We are only 
talking about 30 to 35 nodes at a time, and it happens when the nodes 
boot nbfs and get their commands to run bmcsetup. We also see a perl 
process consistently using 100% of a CPU core during this time. What is 
the process of how these nodes are pulling this data? Obviously it's 
getting it from the chain table, but what could cause this response to 
be so resource hungry and slow?

Also, we are looking to upgrade from 2.3 to 2.6. Has there been any 
changes to this process that could assist in correcting this behavior?


Thanks!



------------------------------------------------------------------------------
Storage Efficiency Calculator
This modeling tool is based on patent-pending intellectual property that
has been used successfully in hundreds of IBM storage optimization engage-
ments, worldwide.  Store less, Store more with what you own, Move data to 
the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
_______________________________________________
xCAT-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/xcat-user

Reply via email to