[Linux-cluster] semi-OT: Where to report (i)SCSI-issues?

[Paolo Marini]

I had to manually add a line with

sleep 30

in the /etc/init.d/iscsi script BEFORE the iscsi login, in order to take
into account the time needed for my bonding (2 ethernet) interface to come
up. These solves the problem and was the only modification I had to do.

cheers, Paolo

> Hi,
>
> http://www.redhat.com/mailman/listinfo
>
> doesn't list a place to report SCSI-problems.
> Anyway, in RHEL5.1, I have a problem in that if a logical volume (LVM)
> is located on an iSCSI device, it will not be detected on boot-up,
> because the network (needed for iSCSI, somehow) starts up after lvm(8).
>
> Either I missed it or the documentation lacks a workaround for this.
> (_netdev doesn't help).
> No tips from the KB. I created an additional init.d-script that does a
> vgscan+vgchange -ay after the iscsi comes up.
> BTW: the iscsi-initiator README actually conflicts with reality:
> configuration databases are in /var/lib, not /etc, as the README suggests.
>
> There's a bugzilla for a similar problem (with shutdown), but it only
> deals with shutdowns. I wonder how they get up their LVMs in the first
> place...
>
> Sorry if this is offtopic here, but I didn't know where to report it
> (except for bugzilla, but maybe there is an (un-) documented workaround.
>
>
>
> cheers,
> Rainer
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>