From: Huang, Peter (peter.huang@attws.com)
Date: Wed Sep 10 2003 - 21:26:00 EDT
Hi, Dear Expert,
I am new in this sun cluster area,
I have two nodes(Netra 1405), one D1000 6 disks, 4 SCSI cables.
disk 4 is mirror for disk1 and disk 3 is spare,
disk 5 is mirror for disk 2 and disk 6 is spare.
disk 4 also serves as quorum disk.
running sun cluster 2.2 clusters in both nodes,
node1 is primary node.
1.if I pulled out the SCSI cable 3, the cluster will be down.
2.if I pulled out the SCSI cable either 2 or 4, the cluster will be OK.
3.if I pulled out the SCSI cable 1, node1 will try to fail over to
node2, but failover not succeed, then node2 reboot,
and cluster is up.
Could you shed some light on me?
why test case 1 cluster will be down? I believe node2 still can access
quorum disk and should become primary node?
why test case 3 failover not succeed, the node2 reboot, and cluster is
up?
why test case 2 the cluster is up and only saw disk syn, not data
services interrupted?
I am trying to simulate the SCSI controller broken or SCSI cable unplug
or broken to see HA services.
Your help is much appreciated!
Peter Huang
AT&T Wireless Mobile Multimedia Services
425-580-6733(Desk)
206-604-4612(Cell)
[demime 1.01b removed an attachment of type application/msword which had a name of SCSI_cable.doc]
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:27:06 EDT