Database Software

Guys,

We have an environment with 2 nodes with RAC database version 10.2.0.1. We need to modify the Public IP and VIP of the Oracle CRS. So we did the following steps:

- Alter the VIP

srvctl modify nodeapps -n node1 -A 192.168.1.101/255.255.255.0/eth0
srvctl modify nodeapps -n node2 -A 192.168.1.102/255.255.255.0/eth0

- Alter the Public IP

oifcfg delif -global eth0
oifcfg setif -global eth0/192.168.1.0:public

- Alter the IP´s of the network interfaces
- Update the /etc/hosts

When we start the Oracle CRS, the components starts OK. But when we reboot the second node, the OCFS2 file system didn´t mount. The following errors occurs:

SCSI device sde: 4194304 512-byte hdwr sectors (2147 MB)
sde: cache data unavailable
sde: assuming drive cache: write through
sde: sde1
parport0: PC-style at 0x378 [PCSPP,TRISTATE]
lp0: using parport0 (polling).
lp0: console ready
mtrr: your processor doesn't support write-combining
(2746,0):o2net_start_connect:1389 ERROR: bind failed with -99 at address 192.168.2.132
(2746,0):o2net_start_connect:1420 connect attempt to node rac1 (num 0) at 192.168.2.131:7777 failed with errno -99
(2746,0):o2net_connect_expired:1444 ERROR: no connection established with node 0 after 10 seconds, giving up and returning errors.
(5457,0):dlm_request_join:786 ERROR: status = -107
(5457,0):dlm_try_to_join_domain:934 ERROR: status = -107
(5457,0):dlm_join_domain:1186 ERROR: status = -107
(5457,0):dlm_register_domain:1379 ERROR: status = -107
(5457,0):ocfs2_dlm_init:2007 ERROR: status = -107
(5457,0):ocfs2_mount_volume:1062 ERROR: status = -107
ocfs2: Unmounting device (8,17) on (node 1)

When we did the command to force the mount occurs the errors:

# mount -a


mount.ocfs2: Transport endpoint is not connected while mounting /dev/sdb1 on /ocfs2. Check 'dmesg' for more information on this error.

What occurs is that the OCFS2 is trying to connect with the older Public IP. My question is, how do i change the public IP in the ocfs2 ?

regards,

Eduardo P Niel
OCP Oracle