Home » Server Options » RAC & Failsafe » Issue with RAC VIP . (ORACLE 10g)
Issue with RAC VIP . [message #485432] Wed, 08 December 2010 00:39 Go to next message
parvees
Messages: 1
Registered: December 2010
Junior Member
Dear All,

I have a database setup using Oracle 10g. two nodes, with real IPs and two VIPS (RAC). as usual both have listeners which is using both the VIP and in the application , i am calling the lister. and based on software load balancing it is serving both the DBs. everything was working perfect.

Yesterday , our network team undergone firewall failover , at that time , for few seconds gateway wasnt reachable. Firewall failover forced two or three request time out. Due to which DB1 lost gateway reachability , i suppose and it moved its VIP to DB2 and now both VIPS are served by DB2 and DB1 is being idle now after that .

lister is serving VIPs which is active on single DB server . This issue happened exactly at same time when the firewall got failover

My question is , why it affected only DB1 , is there any mechanism through which this scenario wont happen in future. I have no clue why is this happening.

please assist.

Thanking everyone in advance.

-Parvees

Please find the log from the server below

2010-12-07 00:29:16.343: [ CRSRES][1266029] In stateChanged, ora.gia-ora-01.vip target is ONLINE
2010-12-07 00:29:16.357: [ CRSRES][1266029] ora.gia-ora-01.vip on gia-ora-01 went OFFLINE unexpectedly
2010-12-07 00:29:16.357: [ CRSRES][1266029] StopResource: setting CLI values
2010-12-07 00:29:16.385: [ CRSRES][1266029] Attempting to stop `ora.gia-ora-01.vip` on member `gia-ora-01`
2010-12-07 00:29:16.987: [ CRSRES][1266029] Stop of `ora.gia-ora-01.vip` on member `gia-ora-01` succeeded.
2010-12-07 00:29:16.988: [ CRSRES][1266029] ora.gia-ora-01.vip RESTART_COUNT=0 RESTART_ATTEMPTS=0
2010-12-07 00:29:16.992: [ CRSRES][1266029] ora.gia-ora-01.vip failed on gia-ora-01 relocating.
2010-12-07 00:29:17.331: [ CRSRES][1266029] StopResource: setting CLI values
2010-12-07 00:29:17.338: [ CRSRES][1266029] Attempting to stop `ora.gia-ora-01.LISTENER_GIA_GIA-ORA-01.lsnr` on member `gia-ora-01`
2010-12-07 00:30:10.280: [ COMMCRS][46]clsc_receive: (101cbb990) error 2

2010-12-07 00:33:03.780: [ CRSRES][1266029] Stop of `ora.gia-ora-01.LISTENER_GIA_GIA-ORA-01.lsnr` on member `gia-ora-01` succeeded.
2010-12-07 00:33:03.781: [ CRSRES][1266029] StopResource: setting CLI values
2010-12-07 00:33:03.801: [ CRSRES][1266029] Attempting to stop `ora.gia-ora-01.LISTENER_GIAKHDB_GIA-ORA-01.lsnr` on member `gia-ora-01`
2010-12-07 00:36:53.125: [ CRSRES][1266029] Stop of `ora.gia-ora-01.LISTENER_GIAKHDB_GIA-ORA-01.lsnr` on member `gia-ora-01` succeeded.
2010-12-07 00:36:53.126: [ CRSRES][1266029] StopResource: setting CLI values
2010-12-07 00:36:53.133: [ CRSRES][1266029] Attempting to stop `ora.gia-ora-01.LISTENER_ASM_GIA-ORA-01.lsnr` on member `gia-ora-01`
2010-12-07 00:40:39.455: [ CRSRES][1266029] Stop of `ora.gia-ora-01.LISTENER_ASM_GIA-ORA-01.lsnr` on member `gia-ora-01` succeeded.
2010-12-07 00:40:39.510: [ CRSRES][1266029] Attempting to start `ora.gia-ora-01.vip` on member `gia-ora-02`
2010-12-07 00:40:43.167: [ CRSRES][1266029] Start of `ora.gia-ora-01.vip` on member `gia-ora-02` succeeded.
2010-12-07 00:40:43.258: [ CRSRES][1266416] CRS-1002: Resource 'ora.gia-ora-01.vip' is already running on member 'gia-ora-02'
Re: Issue with RAC VIP . [message #485483 is a reply to message #485432] Wed, 08 December 2010 09:06 Go to previous messageGo to next message
mkounalis
Messages: 147
Registered: October 2009
Location: Dallas, TX
Senior Member
I believe you need to read rac concepts docs. The network on the db1 node failed so the VIP moved to the secondary. Seems exactly what should happen. When the network got restored you can migrate the VIP back to the db1 node.
Re: Issue with RAC VIP . [message #485484 is a reply to message #485483] Wed, 08 December 2010 09:06 Go to previous messageGo to next message
mkounalis
Messages: 147
Registered: October 2009
Location: Dallas, TX
Senior Member
Also, which db version and patch set are you running?
Re: Issue with RAC VIP . [message #511811 is a reply to message #485484] Wed, 15 June 2011 04:11 Go to previous message
kapardi
Messages: 19
Registered: May 2011
Location: Pune
Junior Member

hello Parvees

When a node fails, the VIP associated with it is automatically failed over to some other node. When this occurs, the following things happen.
(1) VIP detects public network failure which generates a FAN event.
(2) the new node re-arps the world indicating a new MAC address for the IP.
(3) connected clients subscribing to FAN immediately receive ORA-3113 error or equivalent. Those not subscribing to FAN will eventually time out.
(4) New connection requests rapidly traverse the tnsnames.ora address list skipping over the dead nodes, instead of having to wait on TCP-IP timeouts
Previous Topic: Sun Cluster 3.2u1 with oracle 11g
Next Topic: RAC Installation
Goto Forum:
  


Current Time: Thu Mar 28 16:46:37 CDT 2024