We are migrating from a NetWare cluster against an EMC CX500 SAN to a new OES2SP1 LX cluster running against a CX4-120.
The OES2SP1 LX cluster was originally configured to run against some spare space (LUNS) we had in the CX500, whilst we awaited the delivery and configuration of the CX4-120. The nodes were built, cluster formed and test resources created. We have dual HBA's, switches, etc. and have configured MPIO, using multipath.conf to specify friendly names, etc. - all good.
The CX4-120 has now been configured, so I am looking to re-create the SBD on one of it's luns and move other stuff about.

However, I am concerned about some messages I am seeing the /var/log/messages:

Jun 22 18:36:00 sblx-1 kernel: Additional sense: Logical unit not ready, manual intervention required
Jun 22 18:36:00 sblx-1 kernel: end_request: I/O error, dev sd<x>, sector 0

These scroll through, for seemingly all devices, for about a minute during system boot and also when I use nssmu, etc.

I can see single instances of the luns, with their 'friendly names' in nssmu and I have initialized and shared the luns, but the messages still keep scrolling through and I am having trouble re-creating the SBD on one of the new luns (sep. post in clustering forum)

Can anyone give me some insight into what these messages mean? Perhaps mpio isn't configured correctly? Superficially it looks OK, and I have been able to create an nss based pool cluster resource and fail that over to all 3 nodes, but I am concerned about the messages and the fact I can't re-create the sbd...!

All info welcome!

Many thanks


Many than