Happy new year !
We have a SuperMicro server with a LSI MegaRAID 9260-4i controller.
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
The server can't boot due to a bug in LSI megaraid module. Boot log ends by a lot of lines:
RESET_GEN2: retry=xxx, hostdiag=a4
I tried Centos 6.6 install disk and it also fails to find disks but Centos 7 install disk boot fine.
It's annoying to be stuck at 2.6.32-431 kernel and I have no idea how to solve this issue :-(
Thanks for any help.
On 05/01/2015 15:14, Philippe BOURDEU d'AGUERRE wrote:
Happy new year !
We have a SuperMicro server with a LSI MegaRAID 9260-4i controller.
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
I have a similar system, and don't have your problem:
# uname -rv 2.6.32-504.3.3.el6.x86_64 #1 SMP Wed Dec 17 01:55:02 UTC 2014 # dmesg | grep LSI scsi4 : LSI SAS based MegaRAID driver scsi 4:2:0:0: Direct-Access LSI MR9260-4i 2.13 PQ: 0 ANSI: 5 Copyright (c) 1999-2008 LSI Corporation # dmesg | grep Supermicro DMI: Supermicro X8DTU/X8DTU, BIOS 2.1b 12/30/2011
The server can't boot due to a bug in LSI megaraid module. Boot log ends by a lot of lines:
RESET_GEN2: retry=xxx, hostdiag=a4
I tried Centos 6.6 install disk and it also fails to find disks but Centos 7 install disk boot fine.
It's annoying to be stuck at 2.6.32-431 kernel and I have no idea how to solve this issue :-(
Thanks for any help.
Le 05/01/2015 16:22, Giles Coochey a écrit :
I have a similar system, and don't have your problem:
# uname -rv 2.6.32-504.3.3.el6.x86_64 #1 SMP Wed Dec 17 01:55:02 UTC 2014 # dmesg | grep LSI scsi4 : LSI SAS based MegaRAID driver scsi 4:2:0:0: Direct-Access LSI MR9260-4i 2.13 PQ: 0 ANSI: 5 Copyright (c) 1999-2008 LSI Corporation # dmesg | grep Supermicro DMI: Supermicro X8DTU/X8DTU, BIOS 2.1b 12/30/2011
For me:
# uname -rv 2.6.32-431.29.2.el6.x86_64 #1 SMP Tue Sep 9 21:36:05 UTC 2014 # dmesg | grep LSI scsi4 : LSI SAS based MegaRAID driver scsi 4:2:0:0: Direct-Access LSI MR9260-4i 2.13 PQ: 0 ANSI: 5 # dmesg | grep Supermicro DMI: Supermicro H8DGU/H8DGU, BIOS 3.0b 05/07/2013
On 1/5/2015 7:14 AM, Philippe BOURDEU d'AGUERRE wrote:
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
works here fine on the 9261, which is an OEM version of the same card with the connectors in a different orientation... you might check your LSI firmware revision.
# uname -rv 2.6.32-504.1.3.el6.x86_64 #1 SMP Tue Nov 11 17:57:25 UTC 2014
# cat /etc/redhat-release CentOS release 6.6 (Final)
# dmesg |grep LSI mpt2sas0: LSISAS2008: FWVersion(14.00.01.00), ChipRevision(0x03), BiosVersion(07.27.00.00) scsi 0:0:1:0: Enclosure LSI CORP SAS2X28 0717 PQ: 0 ANSI: 5 scsi1 : LSI SAS based MegaRAID driver scsi 1:0:8:0: Enclosure LSI CORP SAS2X36 0717 PQ: 0 ANSI: 5 scsi 1:2:0:0: Direct-Access LSI MR9261-8i 2.12 PQ: 0 ANSI: 5 scsi 1:2:1:0: Direct-Access LSI MR9261-8i 2.12 PQ: 0 ANSI: 5 scsi 1:2:2:0: Direct-Access LSI MR9261-8i 2.12 PQ: 0 ANSI: 5
# /opt/MegaRAID/MegaCli/MegaCli64 -ShowSummary -aAll
System Operating System: Linux version 2.6.32-504.1.3.el6.x86_64 Driver Version: 06.803.01.00-rh1 CLI Version: 8.02.16
Hardware Controller ProductName : LSI MegaRAID SAS 9261-8i(Bus 0, Dev 0) SAS Address : 500605b003bbef10 FW Package Version: 12.12.0-0046 Status : Need Attention BBU BBU Type : iBBU Status : Remaining Capacity is Low Enclosure Product Id : SAS2X36 Type : SES Status : OK
Product Id : SGPIO Type : SGPIO Status : OK ......
Thank you for your help.
Le 05/01/2015 19:10, John R Pierce a écrit :
works here fine on the 9261, which is an OEM version of the same card with the connectors in a different orientation... you might check your LSI firmware revision.
My firmware seems to be more up to date. Anyway, I will try to update firmware. I have to check how to do that.
# dmesg |grep LSI scsi4 : LSI SAS based MegaRAID driver scsi 4:2:0:0: Direct-Access LSI MR9260-4i 2.13 PQ: 0 ANSI: 5 # /opt/MegaRAID/MegaCli/MegaCli64 -ShowSummary -aAll
System Operating System: Linux version 2.6.32-431.29.2.el6.x86_64 Driver Version: 06.700.06.00-rh1 CLI Version: 8.07.14
Hardware Controller ProductName : LSI MegaRAID SAS 9260-4i(Bus 0, Dev 0) SAS Address : 500605b008a30fc0 FW Package Version: 12.14.0-0167 Status : Optimal BBU BBU Type : iBBU Status : Healthy Enclosure Product Id : SGPIO Type : SGPIO Status : OK . . .
On 06/01/2015 09:36, Philippe BOURDEU d'AGUERRE wrote:
Thank you for your help.
Le 05/01/2015 19:10, John R Pierce a écrit :
works here fine on the 9261, which is an OEM version of the same card with the connectors in a different orientation... you might check your LSI firmware revision.
My firmware seems to be more up to date. Anyway, I will try to update firmware. I have to check how to do that.
# dmesg |grep LSI scsi4 : LSI SAS based MegaRAID driver scsi 4:2:0:0: Direct-Access LSI MR9260-4i 2.13 PQ: 0 ANSI: 5 # /opt/MegaRAID/MegaCli/MegaCli64 -ShowSummary -aAll
System Operating System: Linux version 2.6.32-431.29.2.el6.x86_64 Driver Version: 06.700.06.00-rh1 CLI Version: 8.07.14
Hardware Controller ProductName : LSI MegaRAID SAS 9260-4i(Bus 0, Dev 0) SAS Address : 500605b008a30fc0 FW Package Version: 12.14.0-0167 Status : Optimal BBU BBU Type : iBBU Status : Healthy Enclosure Product Id : SGPIO Type : SGPIO Status : OK . . .
For info:
# /opt/MegaRAID/MegaCli/MegaCli64 -ShowSummary -aAll
System Operating System: Linux version 2.6.32-504.3.3.el6.x86_64 Driver Version: 06.803.01.00-rh1 CLI Version: 8.04.07
Hardware Controller ProductName : LSI MegaRAID SAS 9260-4i(Bus 0, Dev 0) SAS Address : 500605b0048274e0 FW Package Version: 12.15.0-0189 Status : Optimal BBU BBU Type : iBBU08 Status : Healthy Enclosure Product Id : SGPIO Type : SGPIO Status : OK
Le 06/01/2015 10:36, Philippe BOURDEU d'AGUERRE a écrit :
Anyway, I will try to update firmware.
I have updated firmware to version 12.15.0-0205. The problem remains :-(
Le 05/01/2015 16:14, Philippe BOURDEU d'AGUERRE a écrit :
We have a SuperMicro server with a LSI MegaRAID 9260-4i controller.
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
The server can't boot due to a bug in LSI megaraid module. Boot log ends by a lot of lines:
RESET_GEN2: retry=xxx, hostdiag=a4
I tried Centos 6.6 install disk and it also fails to find disks but Centos 7 install disk boot fine.
It's annoying to be stuck at 2.6.32-431 kernel and I have no idea how to solve this issue :-(
My workaround is to install kernel-lt from elrepo repository. It seems to work well.
# uname -rv 3.10.64-1.el6.elrepo.x86_64 #1 SMP Fri Jan 9 10:40:17 EST 2015
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Philippe BOURDEU d'AGUERRE Sent: den 14 januari 2015 14:16 To: centos@centos.org Subject: Re: [CentOS] Hardware raid LSI Megaraid not working since Centos 6.6
Le 05/01/2015 16:14, Philippe BOURDEU d'AGUERRE a écrit :
We have a SuperMicro server with a LSI MegaRAID 9260-4i controller.
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
The server can't boot due to a bug in LSI megaraid module. Boot log ends by a lot of lines:
RESET_GEN2: retry=xxx, hostdiag=a4
I tried Centos 6.6 install disk and it also fails to find disks but Centos 7 install disk boot fine.
It's annoying to be stuck at 2.6.32-431 kernel and I have no idea how to solve this issue :-(
My workaround is to install kernel-lt from elrepo repository. It seems to work well.
# uname -rv 3.10.64-1.el6.elrepo.x86_64 #1 SMP Fri Jan 9 10:40:17 EST 2015
When I tinkered with CentOS and one of those LSI-cards a while ago, CentOS couldn't find the card until the card's BIOS was flashed with the latest firmware available at the time. After the flash, the CentOS stock-kernel found the card out-of-the-box.
YMMV though. -- //Sorin
That's weird, because starting with 6.6 we could finally ditch inserting kmods at installation time, the new kernel in 6.6 supported everything. I believe the model we have (among others) is 9341-4i; I would imagine it's pretty similar with what you have. You should open perhaps a bugzilla issue with redhat about this.
Lucian
-- Sent from the Delta quadrant using Borg technology!
Nux! www.nux.ro
----- Original Message -----
From: "Philippe BOURDEU d'AGUERRE" bda@aime-toulouse.fr To: centos@centos.org Sent: Wednesday, 14 January, 2015 13:16:07 Subject: Re: [CentOS] Hardware raid LSI Megaraid not working since Centos 6.6
Le 05/01/2015 16:14, Philippe BOURDEU d'AGUERRE a écrit :
We have a SuperMicro server with a LSI MegaRAID 9260-4i controller.
Since Centos 6.6 update (kernel 2.6.32-504), the controller doesn't initialize at boot. Reverting to kernel 2.6.32-431 allows server to boot fine.
The server can't boot due to a bug in LSI megaraid module. Boot log ends by a lot of lines:
RESET_GEN2: retry=xxx, hostdiag=a4
I tried Centos 6.6 install disk and it also fails to find disks but Centos 7 install disk boot fine.
It's annoying to be stuck at 2.6.32-431 kernel and I have no idea how to solve this issue :-(
My workaround is to install kernel-lt from elrepo repository. It seems to work well.
# uname -rv 3.10.64-1.el6.elrepo.x86_64 #1 SMP Fri Jan 9 10:40:17 EST 2015
-- Philippe BOURDEU d'AGUERRE AIME - Campus de l'INSA http://www.aime-toulouse.fr/ 135 av. de Rangueil Tél +33 561 559 885 31077 TOULOUSE Cedex 4 - FRANCE Fax +33 561 559 870 _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Le 14/01/2015 14:43, Nux! a écrit :
You should open perhaps a bugzilla issue with redhat about this.
Is it possible to post in Redhat bugzilla without a Redhat support contract ? Even when using Centos ?
On 01/14/2015 08:00 AM, Philippe BOURDEU d'AGUERRE wrote:
Le 14/01/2015 14:43, Nux! a écrit :
You should open perhaps a bugzilla issue with redhat about this.
Is it possible to post in Redhat bugzilla without a Redhat support contract ? Even when using Centos ?
Yes, anyone can file a bugzilla report. It's separate from the paid customer support ticketing system. There's no SLA or obligation to respond to bugzilla as there is for paid support tickets.
-
Giles Coochey -
Jim Perrin -
John R Pierce -
Nux! -
Philippe BOURDEU d'AGUERRE -
Sorin Srbu