Wednesday 02 July, 2008 - 09:32
Had problems with the interconnect on padstow1. The eth1 interface failed to come up. Found the following message in /var/log/messages :
ifup: Error, some other host already uses address 192.168.101.1
The reason for this turns out to be that the Vaio had been allocated 192.168.101.1 via DHCP from VMware network.
Editted /etc/sysconfig/network-scripts/ifcfg-eth1 to change the IP address. Had to use service network restart to fix eth1. Also updated /etc/hosts to set the private interconnect address to 192.168.101.3 on both nodes.
There was a subtle problem with the shared OCR disks. padstow1 would report that integrity check would fail (ocrcheck) but ocrdump could extract the OCR data. padstow2 would report no problems. The solution is to start padstow1 first then padstow2. Yet another feature of VMware.
Because this problem was not detected during installation until the configuration assistant stage, I had to manually run the failed commands myself.
Since I forgot to run vipca after the clusterware installation, I now have a single node ASM setup.