Skip to contentSkip to conversation

CCISS driver and disk failure...

DISCUSSION David Brown 2 participants 3 messages Feb 4, 2003 – Feb 6, 2003

lore
Share & tools
lore
#0 lore raw

Hi:

Does anyone know of an easy way to get messages (via syslog or otherwise) when a member disk of a CCISS SMART-2 raid array fails? Grepping through drivers/block/cciss.c didn't yield any obvious printk's. My gut feeling is that one could get the disk failure information through one of the CCISS_PASSTHRU ioctls(); I saw some reference to a similar call in code for monitoring the cpqarray driver.

HP appears to have some sort of management suite, but it appears to require X11 server-side, which isn't an option on this machine.

Is there an easy way to get disk failure information from the CCISS driver, or should I continue relying on the pretty LEDs? :)

Thanks in advance,

- Dave

#1 lore raw

Hi David-

The only way we're able to do this on proliants is to use the *hack* closed-source HP kernel modules cpqasm and cpqevt. We then run their 30+ userspace daemons that monitor the system via these 2 kernel modules. Slap their hacked up ucd-snmpd on top of that and you get disk status monitoring via SNMP on the machine.

On a redhat machine, that's the following RPMs from HP:
cmafdtn
cmanic
cmastor
cmasvr
cpqhealth
ucd-snmp (from HP, of course)

It's a nasty mess, though, so if the cciss author had the time to put disk fail logging into the driver, that'd be amazingly cool.

-Matt

Quoted 27 lines · David Brown

Hi:

Does anyone know of an easy way to get messages (via syslog or otherwise) when a member disk of a CCISS SMART-2 raid array fails? Grepping through drivers/block/cciss.c didn't yield any obvious printk's. My gut feeling is that one could get the disk failure information through one of the CCISS_PASSTHRU ioctls(); I saw some reference to a similar call in code for monitoring the cpqarray driver.

HP appears to have some sort of management suite, but it appears to require X11 server-side, which isn't an option on this machine.

Is there an easy way to get disk failure information from the CCISS driver, or should I continue relying on the pretty LEDs? :)

Thanks in advance,

- Dave

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#2 lore raw

Hi David-

I noticed that the disk array component of this stuff actually does not require the kernel modules, which is good news. So, you should be able to install the 'cmafdtn', 'cmastor', 'cpqhealth' and 'ucd-snmp' RPMs from compaq and monitor your drives via SNMP without using their darn kernel modules. I assume that it's communicating with the array via the cciss driver in some way.

Hope this helps.

-Matt

Quoted 57 lines · Matt C

Hi David-

The only way we're able to do this on proliants is to use the *hack* closed-source HP kernel modules cpqasm and cpqevt. We then run their 30+ userspace daemons that monitor the system via these 2 kernel modules. Slap their hacked up ucd-snmpd on top of that and you get disk status monitoring via SNMP on the machine.

On a redhat machine, that's the following RPMs from HP:
cmafdtn
cmanic
cmastor
cmasvr
cpqhealth
ucd-snmp (from HP, of course)

It's a nasty mess, though, so if the cciss author had the time to put disk fail logging into the driver, that'd be amazingly cool.

-Matt

Quoted 19 lines · David Brown

Hi:

Does anyone know of an easy way to get messages (via syslog or otherwise) when a member disk of a CCISS SMART-2 raid array fails? Grepping through drivers/block/cciss.c didn't yield any obvious printk's. My gut feeling is that one could get the disk failure information through one of the CCISS_PASSTHRU ioctls(); I saw some reference to a similar call in code for monitoring the cpqarray driver.

HP appears to have some sort of management suite, but it appears to require X11 server-side, which isn't an option on this machine.

Is there an easy way to get disk failure information from the CCISS driver, or should I continue relying on the pretty LEDs? :)

Thanks in advance,

Quoted 8 lines

- Dave

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/