Hi,

performing a test installation of OES 1 2-node-cluster with FC-SAN.
Clustering worked fine so far, using NSS cluster volumes.

After this I did a cluster installation of iprint (following the
manuals) and during this created a ext3 cluster resource using EVMS.

IPrint works fine, but since this installation I receive tons of
messages from evms. During bootup I receive messages like:

end_request: I/O error, dev sdb, sector 584888192
Buffer I/O error on device sdb, logical block 73111024
end_request: I/O error, dev sdb, sector 584888192
Buffer I/O error on device sdb, logical block 73111024
end_request: I/O error, dev sdb, sector 0
Buffer I/O error on device sdb, logical block 0
end_request: I/O error, dev sdb, sector 0
Buffer I/O error on device sdb, logical block 0
end_request: I/O error, dev sdb, sector 0

Not only for sdb, also for sdd and sdf, which are the LUNs of the SAN
(two of them for NSS volumes, one for the ext3 irprint volume). In total
I receive approx. 440 of these messages during bootup, and bootup is
much slower than before.


After the systems are booted, when I launch evmsgui, first of all I get
these messages:

From node <name>: Engine: Error code 5 (Input/output error) when reading
the primary copy of feature header on object sdf.
From node <name>: Engine: Error code 5 (Input/output error) when reading
the secondary copy of feature header on object sdf.

These messages are repeated for sdd and sbd. And after this I get the
same messages for the local node. After clicking "OK" button evmsgui is
started and can be used. It lists all devices.

evms-engine.log shows much more error messages, most of them same as
above, but also:

Multipath: read_pv: Error reading LVM PV metadata from object sdf
Multipath: read_pv: Error reading LVM PV metadata from object sdd
Multipath: read_pv: Error reading LVM PV metadata from object sdb


Despite the fact that all volumes seem to work fine, this amount of
error messages can not be ok, I think. Therefor I like to get rid of
them.


Thanks in advance,
Frank