All of lore.kernel.org
 help / color / mirror / Atom feed
* Dell MD3000/multipath - huge numbers of I/O errors on boot
@ 2009-09-25 13:45 John Hughes
  2009-09-25 17:55 ` Moger, Babu
  0 siblings, 1 reply; 4+ messages in thread
From: John Hughes @ 2009-09-25 13:45 UTC (permalink / raw)
  To: dm-devel

I've got a Dell MD3000 SAS cabinet, connected with two SAS cables using 
multipathd/dm-multipath.

Most things work fine, but one problem I have is that the MD3000 wants 
to do failover, it gets very shirty if I access the "ghost" path to a 
LUN without doing some RDAC fancyness.

The problem is that when the kernel discovers each LUN it tries to read 
the partition table, which produces insane numbers of read failures on 
boot.  How can I silence them?

[  236.136435] scsi 4:0:0:14: Direct-Access     DELL     MD3000           0735 PQ: 0 ANSI: 5
[  236.158631] sd 4:0:0:14: [sdav] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[  236.158950] sd 4:0:0:14: [sdav] Write Protect is off
[  236.159035] sd 4:0:0:14: [sdav] Mode Sense: 77 00 10 08
[  236.159456] sd 4:0:0:14: [sdav] Write cache: enabled, read cache: enabled, supports DPO and FUA
[  236.159888] sd 4:0:0:14: [sdav] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[  236.160199] sd 4:0:0:14: [sdav] Write Protect is off
[  236.160285] sd 4:0:0:14: [sdav] Mode Sense: 77 00 10 08
[  236.160679] sd 4:0:0:14: [sdav] Write cache: enabled, read cache: enabled, supports DPO and FUA
[  236.161450] sd 4:0:0:14: [sdav] Attached SCSI disk
[  236.161546] sd 4:0:0:14: Embedded Enclosure Device
[  236.368374] sd 4:0:0:14: Attached scsi generic sg49 type 0
...
[  240.579148] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
[  240.579323] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request [current] 
[  243.792964] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94 ASCQ=0x1
[  245.265769] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
[  245.265940] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request [current] 
[  245.266215] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94 ASCQ=0x1
[  245.825586] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
[  245.825752] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request [current] 
[  245.826017] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94 ASCQ=0x1
... [ MANY MANY ERRORS REMOVED ]
[  303.486849] sd 3:0:1:14: rdac: LUN 14 (unowned)
[  303.494695] sd 4:0:0:14: rdac: LUN 14 (owned)

# multipath -ll 
...
mpath43 (360024e80005b3b23000001b94ab0587f) dm-30 DELL    ,MD3000        
[size=68G][features=1 queue_if_no_path][hwhandler=1 rdac]
\_ round-robin 0 [prio=3][active]
 \_ 4:0:0:14 sdav 66:240 [active][ready]
\_ round-robin 0 [prio=0][enabled]
 \_ 3:0:1:14 sdaf 65:240 [active][ghost]
...

^ permalink raw reply	[flat|nested] 4+ messages in thread

* RE: Dell MD3000/multipath - huge numbers of I/O errors on boot
  2009-09-25 13:45 Dell MD3000/multipath - huge numbers of I/O errors on boot John Hughes
@ 2009-09-25 17:55 ` Moger, Babu
  2009-09-29 16:55   ` John Hughes
  0 siblings, 1 reply; 4+ messages in thread
From: Moger, Babu @ 2009-09-25 17:55 UTC (permalink / raw)
  To: device-mapper development

This is a know problem.. As far as I know these errors mostly come during the boot-up if the scsi rdac handler is not there in the initrd image. You might want to try building scsi_dh_rdac into the initrd image.   

Thanks
Babu Moger 

> -----Original Message-----
> From: dm-devel-bounces@redhat.com [mailto:dm-devel-bounces@redhat.com]
> On Behalf Of John Hughes
> Sent: Friday, September 25, 2009 8:45 AM
> To: dm-devel@redhat.com
> Subject: [dm-devel] Dell MD3000/multipath - huge numbers of I/O errors
> on boot
> 
> I've got a Dell MD3000 SAS cabinet, connected with two SAS cables using
> multipathd/dm-multipath.
> 
> Most things work fine, but one problem I have is that the MD3000 wants
> to do failover, it gets very shirty if I access the "ghost" path to a
> LUN without doing some RDAC fancyness.
> 
> The problem is that when the kernel discovers each LUN it tries to read
> the partition table, which produces insane numbers of read failures on
> boot.  How can I silence them?
> 
> [  236.136435] scsi 4:0:0:14: Direct-Access     DELL     MD3000
> 0735 PQ: 0 ANSI: 5
> [  236.158631] sd 4:0:0:14: [sdav] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [  236.158950] sd 4:0:0:14: [sdav] Write Protect is off
> [  236.159035] sd 4:0:0:14: [sdav] Mode Sense: 77 00 10 08
> [  236.159456] sd 4:0:0:14: [sdav] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [  236.159888] sd 4:0:0:14: [sdav] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [  236.160199] sd 4:0:0:14: [sdav] Write Protect is off
> [  236.160285] sd 4:0:0:14: [sdav] Mode Sense: 77 00 10 08
> [  236.160679] sd 4:0:0:14: [sdav] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [  236.161450] sd 4:0:0:14: [sdav] Attached SCSI disk
> [  236.161546] sd 4:0:0:14: Embedded Enclosure Device
> [  236.368374] sd 4:0:0:14: Attached scsi generic sg49 type 0
> ...
> [  240.579148] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK
> driverbyte=DRIVER_SENSE,SUGGEST_OK
> [  240.579323] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request
> [current]
> [  243.792964] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94
> ASCQ=0x1
> [  245.265769] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK
> driverbyte=DRIVER_SENSE,SUGGEST_OK
> [  245.265940] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request
> [current]
> [  245.266215] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94
> ASCQ=0x1
> [  245.825586] sd 3:0:1:14: [sdaf] Result: hostbyte=DID_OK
> driverbyte=DRIVER_SENSE,SUGGEST_OK
> [  245.825752] sd 3:0:1:14: [sdaf] Sense Key : Illegal Request
> [current]
> [  245.826017] sd 3:0:1:14: [sdaf] <<vendor>> ASC=0x94 ASCQ=0x1ASC=0x94
> ASCQ=0x1
> ... [ MANY MANY ERRORS REMOVED ]
> [  303.486849] sd 3:0:1:14: rdac: LUN 14 (unowned)
> [  303.494695] sd 4:0:0:14: rdac: LUN 14 (owned)
> 
> # multipath -ll
> ...
> mpath43 (360024e80005b3b23000001b94ab0587f) dm-30 DELL    ,MD3000
> [size=68G][features=1 queue_if_no_path][hwhandler=1 rdac]
> \_ round-robin 0 [prio=3][active]
>  \_ 4:0:0:14 sdav 66:240 [active][ready]
> \_ round-robin 0 [prio=0][enabled]
>  \_ 3:0:1:14 sdaf 65:240 [active][ghost]
> ...
> 
> 
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Dell MD3000/multipath - huge numbers of I/O errors on boot
  2009-09-25 17:55 ` Moger, Babu
@ 2009-09-29 16:55   ` John Hughes
  2009-09-29 17:22     ` Moger, Babu
  0 siblings, 1 reply; 4+ messages in thread
From: John Hughes @ 2009-09-29 16:55 UTC (permalink / raw)
  To: device-mapper development

Moger, Babu wrote:
> This is a know problem.. As far as I know these errors mostly come during the boot-up if the scsi rdac handler is not there in the initrd image. You might want to try building scsi_dh_rdac into the initrd image.
Yay!  that works.  Now I get a nice quiet(ish) boot:

[    0.000000] Initializing cgroup subsys cpuset
...
[    1.292136] SCSI subsystem initialized
[    1.294173] rdac: device handler registered
...
[   30.781404] scsi3 : ioc0: LSISAS1068 B0, FwRev=000a3300h, Ports=1, MaxQ=366, IRQ=38
[   30.820324] scsi 3:0:0:0: Direct-Access     DELL     MD3000           0735 PQ: 0 ANSI: 5
[   30.843415] scsi 3:0:0:0: rdac: LUN 0 (owned)
[   30.843838] sd 3:0:0:0: [sdb] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[   30.844158] sd 3:0:0:0: [sdb] Write Protect is off
[   30.844251] sd 3:0:0:0: [sdb] Mode Sense: 77 00 10 08
[   30.844929] sd 3:0:0:0: [sdb] Write cache: enabled, read cache: enabled, supports DPO and FUA
[   30.845678] sd 3:0:0:0: [sdb] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[   30.846020] sd 3:0:0:0: [sdb] Write Protect is off
[   30.846114] sd 3:0:0:0: [sdb] Mode Sense: 77 00 10 08
[   30.846532] sd 3:0:0:0: [sdb] Write cache: enabled, read cache: enabled, supports DPO and FUA
[   30.846649]  sdb: unknown partition table
[   30.857018] sd 3:0:0:0: [sdb] Attached SCSI disk
[   30.857114] sd 3:0:0:0: Embedded Enclosure Device
[   31.229896] sd 3:0:0:0: Attached scsi generic sg3 type 0
[   31.234390] scsi 3:0:0:1: Direct-Access     DELL     MD3000           0735 PQ: 0 ANSI: 5
[   31.256709] scsi 3:0:0:1: rdac: LUN 1 (unowned)
[   31.257125] sd 3:0:0:1: [sdc] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[   31.257444] sd 3:0:0:1: [sdc] Write Protect is off
[   31.257539] sd 3:0:0:1: [sdc] Mode Sense: 77 00 10 08
[   31.257909] sd 3:0:0:1: [sdc] Write cache: enabled, read cache: enabled, supports DPO and FUA
[   31.258338] sd 3:0:0:1: [sdc] 142323712 512-byte hardware sectors: (72.8 GB/67.8 GiB)
[   31.258651] sd 3:0:0:1: [sdc] Write Protect is off
[   31.258744] sd 3:0:0:1: [sdc] Mode Sense: 77 00 10 08
[   31.259108] sd 3:0:0:1: [sdc] Write cache: enabled, read cache: enabled, supports DPO and FUA
[   31.259223]  sdc:<2>ldm_validate_partition_table(): Disk read failed.
[   31.264296] Dev sdc: unable to read RDB block 0
[   31.264389]  unable to read partition table
[   31.264521] sd 3:0:0:1: [sdc] Attached SCSI disk
[   31.264614] sd 3:0:0:1: Embedded Enclosure Device
[   31.528977] sd 3:0:0:1: Attached scsi generic sg4 type 0

^ permalink raw reply	[flat|nested] 4+ messages in thread

* RE: Dell MD3000/multipath - huge numbers of I/O errors on boot
  2009-09-29 16:55   ` John Hughes
@ 2009-09-29 17:22     ` Moger, Babu
  0 siblings, 0 replies; 4+ messages in thread
From: Moger, Babu @ 2009-09-29 17:22 UTC (permalink / raw)
  To: device-mapper development

Good.. Good.. Thanks for the update...

--Babu Moger 

> -----Original Message-----
> From: dm-devel-bounces@redhat.com [mailto:dm-devel-bounces@redhat.com]
> On Behalf Of John Hughes
> Sent: Tuesday, September 29, 2009 11:56 AM
> To: device-mapper development
> Subject: Re: [dm-devel] Dell MD3000/multipath - huge numbers of I/O
> errors on boot
> 
> Moger, Babu wrote:
> > This is a know problem.. As far as I know these errors mostly come
> during the boot-up if the scsi rdac handler is not there in the initrd
> image. You might want to try building scsi_dh_rdac into the initrd
> image.
> Yay!  that works.  Now I get a nice quiet(ish) boot:
> 
> [    0.000000] Initializing cgroup subsys cpuset
> ...
> [    1.292136] SCSI subsystem initialized
> [    1.294173] rdac: device handler registered
> ...
> [   30.781404] scsi3 : ioc0: LSISAS1068 B0, FwRev=000a3300h, Ports=1,
> MaxQ=366, IRQ=38
> [   30.820324] scsi 3:0:0:0: Direct-Access     DELL     MD3000
> 0735 PQ: 0 ANSI: 5
> [   30.843415] scsi 3:0:0:0: rdac: LUN 0 (owned)
> [   30.843838] sd 3:0:0:0: [sdb] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [   30.844158] sd 3:0:0:0: [sdb] Write Protect is off
> [   30.844251] sd 3:0:0:0: [sdb] Mode Sense: 77 00 10 08
> [   30.844929] sd 3:0:0:0: [sdb] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [   30.845678] sd 3:0:0:0: [sdb] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [   30.846020] sd 3:0:0:0: [sdb] Write Protect is off
> [   30.846114] sd 3:0:0:0: [sdb] Mode Sense: 77 00 10 08
> [   30.846532] sd 3:0:0:0: [sdb] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [   30.846649]  sdb: unknown partition table
> [   30.857018] sd 3:0:0:0: [sdb] Attached SCSI disk
> [   30.857114] sd 3:0:0:0: Embedded Enclosure Device
> [   31.229896] sd 3:0:0:0: Attached scsi generic sg3 type 0
> [   31.234390] scsi 3:0:0:1: Direct-Access     DELL     MD3000
> 0735 PQ: 0 ANSI: 5
> [   31.256709] scsi 3:0:0:1: rdac: LUN 1 (unowned)
> [   31.257125] sd 3:0:0:1: [sdc] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [   31.257444] sd 3:0:0:1: [sdc] Write Protect is off
> [   31.257539] sd 3:0:0:1: [sdc] Mode Sense: 77 00 10 08
> [   31.257909] sd 3:0:0:1: [sdc] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [   31.258338] sd 3:0:0:1: [sdc] 142323712 512-byte hardware sectors:
> (72.8 GB/67.8 GiB)
> [   31.258651] sd 3:0:0:1: [sdc] Write Protect is off
> [   31.258744] sd 3:0:0:1: [sdc] Mode Sense: 77 00 10 08
> [   31.259108] sd 3:0:0:1: [sdc] Write cache: enabled, read cache:
> enabled, supports DPO and FUA
> [   31.259223]  sdc:<2>ldm_validate_partition_table(): Disk read
> failed.
> [   31.264296] Dev sdc: unable to read RDB block 0
> [   31.264389]  unable to read partition table
> [   31.264521] sd 3:0:0:1: [sdc] Attached SCSI disk
> [   31.264614] sd 3:0:0:1: Embedded Enclosure Device
> [   31.528977] sd 3:0:0:1: Attached scsi generic sg4 type 0
> 
> 
> 
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-09-29 17:22 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-09-25 13:45 Dell MD3000/multipath - huge numbers of I/O errors on boot John Hughes
2009-09-25 17:55 ` Moger, Babu
2009-09-29 16:55   ` John Hughes
2009-09-29 17:22     ` Moger, Babu

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.