08-05-2013 11:05 PM - last edited on 08-06-2013 07:43 PM by Liuqing
We have 2 nodes Oracle Rac cluster with HP-UX 11.31. Both of the nodes are rebooting very frequently and 1node is ahead in this, node 2 boot lesser than node1.
I tried logging a case with HP in the past many times but on the basis of crashdump they will just inform us that
""User requested reset of the system.
Oracle CRS TOC for clusterware integrity..."
Please check with Oracle. Now when we go to Oracle first they say the issue are from the OS side but now they said it looks "Issue is the private network talking between nodes and not the public network".
Any help how to resolve this issues and the server gets so slow also that we need to reboot if it doesn't get rebooted.
Is there any file where we can define the resource utilization by OS that it should use this much memory and can't be overtaken by DB or applications? Below is the memory usage at the time of boot.
Physical Memory = 6283703 pages (23.97 GB)
Free Memory = 81313 pages (317.6 MB)
Average Free Memory = 88717 pages (346.6 MB)
gpgslim = 21504 pages (84 MB)
lotsfree = 433726 pages (1.65 GB)
desfree = 21504 pages (84 MB)
minfree = 10496 pages (41 MB)
Any help how to fix this issues and any file which will let us know if virtual IP assigned to Nodes goes down or not. As oracle says becuase of virtual IP going down it is rebooting.
Advance in thanks for any help.
P.S. This thread has been moved from HP-UX > sysadmin to Databases.
-HP Forum Moderator
08-07-2013 01:13 PM