warning in stm- how do I get rid of it?

[Mark Vollmers]

Hi, all. I've been getting this error emailed to me since I updated to 11.0 about the server not liking our RAID drive. I think it just thinks that the firmware is bad or whatever, but the drive is working fine (the same as it was under 10.20). I will attach the email to root, since it has a little more that what shows up in the syslog. I get this daily. How can I get it to stop sending me this error. The drive is a third party (Artecon, now owned by Dot Hill) and I have the latest firmware from them on the drive. I think that they have stopped working with the drive as far as updates, although they still support it. Ideas, anyone? Thanks.

Mark

[Vito Sarducci]

Go to this site to investigate.

http://docs.hp.com/hpux/content/hardware/ems/scsi.htm#100776

Im sure this will help.

Vito

[Mark Vollmers]

Vito-

I already looked there, and it just says under the recommended action:

Probable Cause / Recommended Action: The error most likely indicates that the device is not fully supported by the current driver. This may or may not cause problem in operation of the device.

This really doesn't help me a whole lot. I'm looking for a way to get stm to either shut up and accept the device or to just shut up.

Mark

[John Bolene]

Looks like the raid is sending nonsense sense bytes.

They may have had meaning under 10.20, but from your log, 11 does not like them.

Only solution I see, since the manufacturer sounds like they will not upgrade software, is to live with it or change the hardware to something else.

[Vito Sarducci]

Mark,

I would go into STM and disable logging from that device. There is an option to active and de-activate reporting of errors.

In the utility tools option, you can change your options to only report errors and warnings or only errors. Since this is a warning, i would try using this option to report only errors and see if this resolves the issue.

[Mark Vollmers]

Vito-

Is there a way to only disable logging for that ONE warning? I would really hate to disable the logging for everything and then end up missing another problem. Thanks.

Mark

[Vito Sarducci]

HP will be reluctant to resolving the error message your receiving only because this is a third party drive and software driver.

Now they will tell you to reduce the warning level on that device and this might be your only alternative. Give it a shot and see what level of reporting is showing up after making the change and watching it the next few days of operations.

If this eliminates your achtungs coming from other devices, then I would switch it back to normal reporting. And seek another opinion...

Vito

[Vito Sarducci]

The other issue is if you are suspecting other problems arising from this device, then I would recommend replacing it with an HP supported device and software driver certified by HP. This would eliminate your problem all together.

Vito

[Mark Vollmers]

How do I disable logging for that device. The only place I see is to go under Options > general and click to report only errors. Is there another place?
As far as replacing the drive, I doubt that the boss will kick out 20K for a new drive. Kinda tough to justify for a warning...

Mark

[Vito Sarducci · Accepted solution]

go into STM, options, tool options, utility and generate tool activity log. This is where I would just generate errors and warnings. or higher to just errors. I personally dont worry about warnings until they become errors.

I hope this helps.

Vito

[Deshpande Prashant]

Hi Mark
Try using /etc/opt/resmon/lbin/monconfig
to modify the monitored events and their severity.

Thanks.
Prashant Deshpande.

[Christopher McCray_1]

If you are certain that this is an erroneous message, you can do this:

EMS uses a file in /var/stm/data/tools/monitor called disabled_instances to tell it to ignore a particular instance(s) of monitored devices. So what you would do is
1)run /etc/opt/resmon/lbin/monconfig
2) (K)ill monitoring
3) vi the disabled_instances file and add the line to the end:
/storage/events/disks/default/8_0.0.0 (in the case of your one disk)
4)rerun monconfig, (E)enable monitoring
5) (C)heck monitors; the disk you specified should show up with no actions descriptions for it.

I would try to also find out if the disk in question has fallen off the hcl.
I hope this helps.
Regards
Chris

[Christopher McCray_1]

An alternative is to turn off the particular event:
1)run /etc/opt/resmon/lbin/monconfig, (K)ill monitoring.
2)vi the /var/stm/config/tools/monitor/default_disk_em.clcfg (after making a copy) file and find the event # (on the notification)and change the TRUE to a FALSE.
3)rerun monconfig, (E)nable monitoring, (C)heck monitors
4)run /etc/opt/resmon/lbin/set_fixed -l to check the monitor's state as UP or DOWN. run /etc/opt/resmon/lbin/set_fixed -n \* to bring the down ones up.

Unfortunately, the only alternatives are to either turn off all monitoring for a device, explained in my first reply, or to turn off a specific event for all associated devices. You will have to decide which is the better of two evils. Alternatively, loading the most current version of Online diagnostics may solve your problem, assuming you haven't already.

Good luck,
Chris

[Mark Vollmers]

Thanks for all the help everyone. I will have to figure out which path to take here (turn off everything, turn off the device, or turn off the event). I'm thinking event, since it is just a driver warning, but we'll see. Again, thanks.

mark