Hardware ask?

[castro_2]

I have this error in syslog and dmesg


vmunix: SCSI: Third party detected bus hang -- lbolt: 9 936287, bus: 0
vmunix: lbp->state: 6020
vmunix: lbp->offset: 0
vmunix: lbp->uPhysScript: 300000
vmunix: From most recent interrupt:
vmunix: ISTAT: 22, SIST0: 04, SIST1: 00, DSTAT: 80, DSPS: 0000000a
vmunix: lsp: 0000000000000000
vmunix: lbp->owner: 0000000040e1bd00
vmunix: bp->b_dev: cb000002
vmunix: scb->io_id: 1f
vmunix: scb->cdb: 12 00 00 00 80 00
vmunix: lbolt_at_timeout: 9936187, lbolt_at_start: 9934687
vmunix: lsp->state: 85
vmunix: SCSI: Resetting SCSI -- lbolt: 9936387, bus: 0
vmunix: SCSI: Reset detected -- lbolt: 9936387, bus: 0
vmunix: SCSI: Unexpected Disconnect -- lbolt: 9936887,dev: cb000002, io_id: 1f

Which device is cb000002? and why the reser SCSI?
Thanks

[Cheryl Griffin · Accepted solution]

Third party detected bus hang is referring to a device on the bus. To find the device from the message "dev: cb000002 "
# ll /dev/* |grep 000002

If the error is frequent, you need to have the device checked out (power, cable, termination, etc.) and check for things like the SCSI patch (PHKL_18543 if 11.00).

[A. Clay Stephenson]

The first two hex digits indicate the major device. 0xCB=203dec

Do an lsdev and you will find that 203 is sctl - a scsi pass-thru.

The next two hex digits (00) indicate the controller instance - c0

The next hex digit (0) is the SCSI ID; the next (0) is the LUN. The remaining hex digits are driver specific.

In your case, the problem is with
a device node with a 203 major device and a minor device number of 0x000002.

Look under /dev (maybe using lssf) until you find a node that matches. There is a good chance that this is a device node for a robotic media changer.

SCSI resets can happen if a device on the bus was power-cycled, if a device initiated a reset, or if a cable was removed. They can also be triggered by faulty hardware or if the bus is not properly terminated. A few resets are nothing to worry about. The SCSI protocol is expected to handle these events but if you are seeing more than a few then start looking.

[castro_2]

There is a SCSI patch for 11.i?
Thanks

[S.K. Chan]

Yes .. 2 came in mind ..
1- SCSI IO Subsystem Cumulative (PHKL_23666)
2- SCSI IO Cumulative (PHKL_29039)
I have not check the description of these patches yet. You want to make sure all dependencies are met before installing them. Did you see alot of these reset error in syslog.log or just a few ?

[castro_2]

A few but when I do ioscan -fn write the syslog with the error.
Thanks