NetRaid Errors

[Ronen Shurer]

I got the following errors on the Netraid Event Log, but i cannot understand what do they mean?

Log Started.
- Sun Mar 23 11:39:31 2003
NOTIFY:Check Condition on Ch 0 ID 5 with the following sense key - Sun Mar 23 11:39:31 2003
f0 00 05 00 00 00 00
0e 00 00 00 00 20 00
NOTIFY:Check Condition on Ch 0 ID 5 with the following sense key - Sun Mar 23 11:39:31 2003
f0 00 05 00 00 00 00
0e 00 00 00 00 20 00
NOTIFY:Check Condition on Ch 0 ID 5 with the following sense key - Sun Mar 23 11:39:31 2003
f0 00 05 00 00 00 00
0e 00 00 00 00 20 00
NOTIFY:Check Condition on Ch 0 ID 5 with the following sense key - Sun Mar 23 11:39:31 2003
f0 00 05 00 00 00 00
0e 00 00 00 00 20 00

I am using an LH3 netserver with an onboard netraid, a 3si netraid card, and an fcarray card, on the same server.

[Alicia White]

Well, I can't help you interpret the sense key codes. But, I think you need to look very seriously at your drive cage as the source of this problem.

On the HP NetServer LH series servers ID5 on each drive cage/channel is reserved for the SAF-TE processor. This device on the backplane monitors the condition of the drive cage (voltage, temp, etc.). I used to do tech support for HP NetServers and I don't think I ever saw a sense key error on the SAF-TE processor before. But, there are a few things you can try to trouble-shoot this problem.

Make sure the firmware on the drive cage backplane is up to date. You can download the firmware file (lh3fwhsb.exe) at:

http://h20004.www2.hp.com/soar_rnotes/bsdmatrix/matrix50448en_US.html

While you're on that page, you should also download the files for the latest BIOS (lh3bios_us.exe) & NetRAID firmware (fw_all_nr_lh3_d0303.exe). It's always a good idea to keep your firmware & BIOS all up to date. You can also download the latest firmware for the 3si on that page, too (fw_all_nr_3si_c0203.exe).

If after updating the NetRAID controllers' firmware and backplane firmware, the problem doesn't go away, I think you may need a new drive cage, which is the only way to replace the hard drive backplane.

Hopefully, your server is still under warranty so you can get a free replacement from HP if that's necessary. But, I believe the LH3s are all now out of warranty.

Good luck,

Alicia

[kris rombauts]

Hi Ronan,

this is nothing to worry about and you will see that each time you open up the Netraid assistant and look at the log file, the same entries will be added with the current date and time. This is due to a "limitation" (or bug ..) of how Netraid Assistant cooperates with the legacy Netraid controllers. Each time you open the log file it goes out and reads the error log contents from the controller but it adds its own date/timestamp and not the date/timestamp when the error got logged at in the controllers nvram. So this confuses us a bit in a way that you think the error is brand new, whereas it could have been in the log for weeks, months etc etc.

Enough about this, now to the error. In fact as Alicia pointed out this is generated by the SAFTE processor on the disk backplane and the SCSI sense key means this :

05: ILLEGAL REQUEST. Indicates that there was an illegal parameter in the command descriptor block or in the additional parameters supplied as data for some commands (FORMAT UNIT, SEARCH DATA,etc.). If the device server detects an invalid parameter in the command descriptor block, then it shall terminate the command
without altering the medium. If the device server detects an invalid parameter in the additional parameters supplied as data,
then the device server may have already altered the medium.

20 00: Invalid Command Operation Code
The controller received a command from the initiator that it does not support.


So it means someone has send a SCSI command to the SAFTE processor (at SCSI ID=5) like a write command i.e. but this SCSI processor is not a disk device and does not support such SCSI commands, hence it replies with this SCSI sense code to inform the one who send it that this is not something he "understands".

This can be send from older rev's of TopTools agents or some special utilities that scan the SCSI bus i.e.

Not sure since when you noticed this the first time but it's probably in their since a long time. The only way to clear this from reoccuring in the logfile is unfortunately to erase the Netraid controllers configuration and reconfigure the raid config again with the exact same parameters or rereading from the disk (without initializing of course) but i would not do this just to get rid of these log entries now you know what they mean.

HTH

Kris

[Marino Meloni_1]

f0 00 05 00 00 00 00 0e 00 00 00 00 20 00
You should take the word 2 as a Sense Key, the 12 as a ASK (Additional Sense Key) the 13 as ASKQ (Additional Sense Key Kualifier)

The meaning of sense key 05: ILLEGAL REQUEST

The meaning of ASK/ASKQ: INVALID COMMAND OPERATION CODE

NOTIFY:Check Condition on Ch 0 ID 5 means that some software have sent a non valid request to a device in Id 5, that on the lh3 is the SaFTe processor (the one who manage the the scsi bus on the hotswap cage), it is not a hdd.

Check if you use some supervisor software that can try to send request to this processor thinking it is a HDD

Marino

[Hans Lindgren]

Hello,
I have another sense code which I can't decipher:
NOTIFY: Check Condition on Ch 0 ID 2 with the following sense key - Wed Jul 06 10:45:42 2005
f0 00 03 07 a7 57 8f
18 00 00 00 00 11 02

This sense key started to appear on May 16th.
The server is a NetServer LH 3000 with an integrated NetRaid controller

Thank you, Hans

[kris rombauts]

Hans,

it looks like this disk drive has medium problems. If you have the toptools agents installed they should also log you some entries.

This disk looks like it is sick, depends how many times this entry gets logged in the Windows System eventlog if the necessary s/w is installed to copy these errors into the Windows log also. (Netraid monitor service and toptools i.e.)



3h MEDIUM ERROR. Indicates that the command terminated with a nonrecovered error condition that was probably caused by a flaw in the medium or an error in the recorded data. This sense key may also be returned if the device server is unable to distinguish between a flaw in the medium and a specific hardware failure.

"11 02" means error is to long to correct.


So i would make sure (depends on the frequency of this error in the Windows logs)
to prepare for a disk replacement.

You can also run a consistency check if you don't doso regualry and if you have a raid level other then raid0 so the whole array gets scanned for consistency, this is a kinf of health check really.


HTH

Kris

[calbendin]

Hello,
I have another sense code which I can't decipher:
Rebuild Started on CHNL 0 TARG 2. - Mon Jul 18 13:14:53 2005
NOTIFY:Check Condition on Ch 0 ID 2 with the following sense key - Mon Jul 18 13:15:01 2005
f0 00 04 00 00 04 7c
28 00 00 00 00 03 00
This sense key appear when we change a disk an order rebuilt.

The server is a NetServer LH 6000 U and the disks are in raid 5

Thanks

[kris rombauts]

pls check this thread :



http://forums.itrc.hp.com/service/forums/questionanswer.do?threadId=940655