CentOS - Aug 2010 - Automatically detecting LUNs without a reboot

Hi everybody.

I have an HP StorageWorks array connected to one of my servers, 
connected via an LSI controller (lspci reports: SCSI storage controller: 
LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS (rev 08)) 
on an HP DL385 running CentOS 5.5 x86_64.

I have set up multipath and all is working but, when I create a snaphost 
and export it as a new LUN from the HP web interface, it does not appear 
as  a new disk. If I reboot, I get the LUN to appear as a new SCSI 
device (sd[a-z]).

Is there any way I can force automatic detection? One of the reasons to 
use snapshots is to be able to back up the system without downtime, so 
not being able to detect the LUNs makes the whole set up a bit pointless.

I have tried kudzu, the rescan-scsi-bus.sh script and everything else I 
have found online, with no luck. The LSI tools only seem to be available 
for i386, and I would like to avoid adding the 32bit environment, if I can.

Any help would be greatly appreciated.

Thanks,

Gabriel Tabares

centos-bounces at centos.org wrote on 31.08.2010 11:43:50:
> Gabriel Tabares <gabriel.tabares at roboreus.com> 
> Gesendet von: centos-bounces at centos.org
> 
> 31.08.2010 11:44
> 
> Bitte antworten an
> CentOS mailing list <centos at centos.org>
> 
> An
> 
> CentOS mailing list <centos at centos.org>
> 
> Kopie
> 
> Thema
> 
> [CentOS] Automatically detecting LUNs without a reboot
> 
> Hi everybody.
> I have an HP StorageWorks array connected to one of my servers,
> connected via an LSI controller (lspci reports: SCSI storage controller:
> LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS (rev 08))
> on an HP DL385 running CentOS 5.5 x86_64.
> I have set up multipath and all is working but, when I create a snaphost
> and export it as a new LUN from the HP web interface, it does not appear
> as  a new disk. If I reboot, I get the LUN to appear as a new SCSI
> device (sd[a-z]).
> Is there any way I can force automatic detection? One of the reasons to
> use snapshots is to be able to back up the system without downtime, so
> not being able to detect the LUNs makes the whole set up a bit 
pointless.
> I have tried kudzu, the rescan-scsi-bus.sh script and everything else I
> have found online, with no luck. The LSI tools only seem to be available
> for i386, and I would like to avoid adding the 32bit environment, if I 
can.
> Any help would be greatly appreciated.
> Thanks,
> Gabriel Tabares
> _______________________________________________
> CentOS mailing list
> CentOS at centos.org
> http://lists.centos.org/mailman/listinfo/centos
Hi,
I do a rescan (online) on multipath with 
echo "- - -" > /sys/class/scsi_host/host0/scan
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan
echo "- - -" > /sys/class/scsi_host/host3/scan
echo "- - -" > /sys/class/scsi_host/host4/scan
echo "- - -" > /sys/class/scsi_host/host5/scan
echo "- - -" > /sys/class/scsi_host/host6/scan

depends on the amount of controllers.
 
Gru? 
Andreas Reschke
________________________________________________________________

Unix/Linux-Administration
Andreas.Reschke at behrgroup.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.centos.org/pipermail/centos/attachments/20100831/eef09971/attachment-0002.html>

>   Hi everybody.
>
> I have an HP StorageWorks array connected to one of my servers,
> connected via an LSI controller (lspci reports: SCSI storage controller:
> LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS (rev 08))
> on an HP DL385 running CentOS 5.5 x86_64.
>
> I have set up multipath and all is working but, when I create a snaphost
> and export it as a new LUN from the HP web interface, it does not appear
> as  a new disk. If I reboot, I get the LUN to appear as a new SCSI
> device (sd[a-z]).
>
> Is there any way I can force automatic detection? One of the reasons to
> use snapshots is to be able to back up the system without downtime, so
> not being able to detect the LUNs makes the whole set up a bit pointless.
>
> I have tried kudzu, the rescan-scsi-bus.sh script and everything else I
> have found online, with no luck. The LSI tools only seem to be available
> for i386, and I would like to avoid adding the 32bit environment, if I
> can.
>
> Any help would be greatly appreciated.
>
> Thanks,
>
> Gabriel Tabares
rescan-scsi-bus.sh from sg3-utils should work.

Please see
http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/html/Online_Storage_Reconfiguration_Guide/scanning_for_new_devices.html.

Regards

Alexander

On 31/08/2010 11:07, Andreas Reschke wrote:
>
> centos-bounces at centos.org wrote on 31.08.2010 11:43:50:
>
> > Gabriel Tabares <gabriel.tabares at roboreus.com>
> > Gesendet von: centos-bounces at centos.org
> >
> > 31.08.2010 11:44
> >
> > Bitte antworten an
> > CentOS mailing list <centos at centos.org>
> >
> > An
> >
> > CentOS mailing list <centos at centos.org>
> >
> > Kopie
> >
> > Thema
> >
> > [CentOS] Automatically detecting LUNs without a reboot
> >
> > Hi everybody.
>
> > I have an HP StorageWorks array connected to one of my servers,
> > connected via an LSI controller (lspci reports: SCSI storage
controller:
> > LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS (rev
08))
> > on an HP DL385 running CentOS 5.5 x86_64.
>
> > I have set up multipath and all is working but, when I create a
snaphost
> > and export it as a new LUN from the HP web interface, it does not
appear
> > as  a new disk. If I reboot, I get the LUN to appear as a new SCSI
> > device (sd[a-z]).
>
> > Is there any way I can force automatic detection? One of the reasons
to
> > use snapshots is to be able to back up the system without downtime, so
> > not being able to detect the LUNs makes the whole set up a bit 
> pointless.
>
> > I have tried kudzu, the rescan-scsi-bus.sh script and everything else
I
> > have found online, with no luck. The LSI tools only seem to be
available
> > for i386, and I would like to avoid adding the 32bit environment, if 
> I can.
>
> > Any help would be greatly appreciated.
>
> > Thanks,
>
> > Gabriel Tabares
>
> > _______________________________________________
> > CentOS mailing list
> > CentOS at centos.org
> > http://lists.centos.org/mailman/listinfo/centos
>
> Hi,
> I do a rescan (online) on multipath with
> echo "- - -" > /sys/class/scsi_host/host0/scan
> echo "- - -" > /sys/class/scsi_host/host1/scan
> echo "- - -" > /sys/class/scsi_host/host2/scan
> echo "- - -" > /sys/class/scsi_host/host3/scan
> echo "- - -" > /sys/class/scsi_host/host4/scan
> echo "- - -" > /sys/class/scsi_host/host5/scan
> echo "- - -" > /sys/class/scsi_host/host6/scan
>
> depends on the amount of controllers.
Tried that. I only have one controller:

[root at db02 ~]# cat /proc/partitions
major minor  #blocks  name

  104     0  292935982 cciss/c0d0
  104     1  265217053 cciss/c0d0p1
  104     2    2988090 cciss/c0d0p2
    8     0  683593744 sda
    8     1  683589816 sda1
    8    16  683593744 sdb
    8    17  683589816 sdb1
    8    32  683593744 sdc
    8    33  683589816 sdc1
[root at db02 ~]# echo "- - -" > /sys/class/scsi_host/host0/scan
[root at db02 ~]# cat /proc/partitions
major minor  #blocks  name

  104     0  292935982 cciss/c0d0
  104     1  265217053 cciss/c0d0p1
  104     2    2988090 cciss/c0d0p2
    8     0  683593744 sda
    8     1  683589816 sda1
    8    16  683593744 sdb
    8    17  683589816 sdb1
    8    32  683593744 sdc
    8    33  683589816 sdc1

Still not appearing at all.
>
> Gru?
> Andreas Reschke
> ________________________________________________________________
>
> Unix/Linux-Administration
> Andreas.Reschke at behrgroup.com
>
>
> _______________________________________________
> CentOS mailing list
> CentOS at centos.org
> http://lists.centos.org/mailman/listinfo/centos