RAC 10.1.0.3 VIP and instance availability, what is expected behaveoir
- From: steen bartholdy <steenbdkoraclel@xxxxxxxx>
- To: freelist oracle <oracle-l@xxxxxxxxxxxxx>
- Date: Mon, 13 Mar 2006 16:31:08 +0100 (CET)
hi
2 nodes RAC on AIX instance and CRS version 10.1.0.3
We have a setup where we are independent of the VIP adresses for production
use.
We had an incedent where 1switch failed where both VIP adresses is connected,
not smart but true.
I observed this from syslog :
CRS tries 3 times to move each VIP adress to the other side.
>> ifconfig delete and alias commands <<
This happendend on both sides.
When this didnt work ( hardware malfunction on switch ) shutdown started on
plab2
>>
Mon Mar 13 13:26:36 2006
Shutting down instance: further logons disabled
...
Mon Mar 13 13:26:39 2006
Shutting down instance (transactional local)
...
Mon Mar 13 13:26:55 2006
Reconfiguration started (old inc 2, new inc 3)
List of nodes:
1
Global Resource Directory frozen
* dead instance detected - domain 0 invalid = TRUE
Update rdomain variables
Communication channels reestablished
Master broadcasted resource hash value bitmaps
Non-local Process blocks cleaned out
Mon Mar 13 13:26:55 2006
<<
so before this shutdown complete before plab1 started and completed shutdown :
>>
Mon Mar 13 13:26:43 2006
Shutting down instance: further logons disabled
...
Mon Mar 13 13:26:50 2006
ALTER DATABASE DISMOUNT
Completed: ALTER DATABASE DISMOUNT
<<
is it expected behaveoir that both sides shutdown ? is the reason for the
shutdown the missing VIP's ?
Where do I see if CRS has detected that the interconnect was down ?
Thanks In advance
SteenB, Denmark
Other related posts:
- » RAC 10.1.0.3 VIP and instance availability, what is expected behaveoir
