Error messages in Tru64 Unix

[Ronny_7]

Hi,

System is A4100 running Tru64 Unix V4.0D.
There are two third party Judeboxes connected to this system.

Currently there are no problem accessing these two judeboxes but the messages keep logging these messages,

vmunix: [2/1/0] scmn_ReportError: RZxx: test unit ready failed, sense key = 0x2

Can these be ignored or is there any way to stop these messages?

Thank you.

Regards,
Ronny

[Dave Bechtold]

Hi Ronny,

This may be additional information being provided by the SCSI CAM sub-system based on a sysconfig attribute being set to report more information than usual.

Check the sysconfig cam subsystem attributes, especially cam_printdb_flag, cam_bootmsgs, cam_ccfg_debug - these are all set to 0 by default. Maybe cam_ccfg_debug is set to a non-zero value?

Beyond that I'd check with the OEM for the Jukebox to make sure the ddr entry and any configuration settings are properly setup.

Hope that helps,
Dave Bechtold

[Ronny_7]

Hi Dave,

Thank you for the reply but I do not quite understand your suggestion.

Are you saying cam_ccfg_debug should be set to zero?
Is there any implication to the system by changing this parameter?
Is there anywhere I can find the definitions for the cam attributes?

You mentioned also to check the ddr entry, are you saying the Judebox vendor should review this file and edit accordingly?

Below is the output from the system, I do not see any parameters you mentioned.
#sysconfig -q cam
cam:
cam_initialized_flag = 1
cam_printdb_flag = 0
cam_param_1 = 1
number_of_SCSI_buses = 0
num_peripheral_drivers = 2

Regards,
Ronny

[Mohamed K Ahmed]

The parameters Dave told you about are not in the Tru64 UNIX V4.0D, they started from V5.x
Anyway, V4.0D is obsolete now, I really recommend to upgrade to the latest version, or at least to V 4.0G where sustained prior version support will end in 2005.

Mohamed

[Ralf Puchner]

vmunix: [2/1/0] scmn_ReportError: RZxx: test unit ready failed, sense key = 0x2

"RZxx" means a disk not a tape!
"test unit ready failed" is a follow on to check the unit, the sense key 0x2 has the following subcodes:

04 01 Logical unit is in process of becoming ready
04 03 Logical unit not ready, manual intervention required
30 00 Incompatible medium installed
30 03 Cleaning cartridge installed
30 04 Diagnostic cartridge installed
3A 00 Cartridge or magazine not present
53 00 Cartridge load/eject failure
5A 91 Eject button pushed

so please call

# uerf -R -o full

for further investigation, if the tape is the cause of the message there is no medium in the drive, but if a disk is the target of the message there seems to be a problem.

[Ronny_7]

Hi Ralf,

Thank you for the reply.

On this SCSI bus 2 are two HP judeboxes, daisy-chained and no other devices.
Device: C1160A Bus: 2, Target: 0, Lun: 0, Type: Medium Changer
Device: C1716T Bus: 2, Target: 1, Lun: 0, Type: Optical Memory
Device: C1716T Bus: 2, Target: 2, Lun: 0, Type: Optical Memory
Device: C1160A Bus: 2, Target: 3, Lun: 0, Type: Medium Changer
Device: C1716T Bus: 2, Target: 4, Lun: 0, Type: Optical Memory
Device: C1716T Bus: 2, Target: 5, Lun: 0, Type: Optical Memory

There is no errors in the binary.errlog.
In messages, the scmn_reporterror errors are logging on target 0 to 5 on SCSI bus 2.

Do you think these could be what you suggest as no medium in the judebox or could be missing entry in ddr.dbase as what Dave had suspected?

Regards,
Ronny

[Joris Denayer]

Ronny,

fyi.

Optical Storage is no longer supported in the V5 release stream. :-{

Joris

[Ralf Puchner · Accepted solution]

Joris,

good point, but the problem here is 4.0d and not the V5.x codestream.

So the problem here is the check of the medium within the drives. In this case you can ignore the message.

[Joris Denayer]

Ralf,

That is correct. I just want to mention it, because one of my customers lost 2 days because he did not read the Release Notes.

[Ronny_7]

Hi Ralf,

Thank you for the explanation and clarification.

Hi Joris,

Thank you for highlighting that to me.

Regards,
Ronny

[Ronny_7]

As per replies.