Two node cluster : HP-UX 11i v2 (HP-UX 11.23) (Boot on local disc)
Service Guard 11.17
Software : Oracle11g
Old disc: HP EVA 4000
Migration to VMAX20K () with LVM (by EMC Engineers)
Issue: before Migration to VMAX the Servers memory consumption was between 70 to 80% after migration the consumption is 95 to 100% and this cause Oracle DB issues :
Process startup failed, error stack:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn5
ORA-27303: additional information: skgpspawn5
someone already encountered this issue?
All diagnostic paths would be welcome.
Fork errors are normally indicative of resource issues
You should check the OS log for errors and contact EMC support who will be able to escalate if partner support is required, e.g. Oracle.
Seems it is not disk I/O related issue from VMAX. Not sure if you can collect the current Oracle AWR during the time when problem happens (better also have AWR before migration if possible for comparison).
Can I have your emc email address, I will forward AWR collection guide to you.
Thank you Allanr and zhaos2
(@zhaos2: I am not an emc employee )
The process that caused the increase in consumption of memory was sendmail, which put its message in the queue and was unable to send them.
The migration to the VMAX was the last action performed on the platform before this anomaly and eventually it is not the cause.
See you next.