Syslog message

[Matthew Pegge_1]

Getting the following message in syslog, we've had HP hardware engineer look at this and there is no fault. However the error persists. Any ideas?

Apr 14 16:40:02 cnshau01 vmunix:
Apr 14 16:40:02 cnshau01 Unix: SCSI: Unexpected Disconnect -- lbolt: 1895788, dev: cb03f002, io_id: 300010e

[Ken Hubnik_2]

I would get a second opinion from another engineer. Have you tried a new terminator??

[Christian Gebhardt]

Hi

there are several postings in this forum about this error (Search: Unexpected Disconnect)

First try this:
http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0xde338f960573d611abdb0090277a778c,00.html

Chris

[Paula J Frazer-Campbell]

HI

Check the following

Make sure that the SCSI cabling and termination is OK.

Try increasing the timeout using pvchange.

If still the problem is there,then replace the disk.

Paula

[Matthew Pegge_1]

Unfortunately ths box is in shanghai and I am based in UK. So it is hard to know exactly what is wrong. All I know is that they replaced all the SCSI terminals, and there isn't even anything on the path that it is complaining about. If I check the . The engineer suggested patching but I find this hard to believe as it is no different to other systems we have??!

[Michael Steele_2]

Looks like device c3t15d2. Do you have a disk array?

Review the timeout of the disk with pvdisplay and refer to the 'IO Timeout' value and to increase it up to 180. The default is 30 so start with something in between.

pvchange -t 120 /dev/dsk/c3t15d2.

I also encourgage you to review LOGTOOL and check for failing devices:

STM > TOOLS > UTILITY > RUN > LOGTOOL > FILE > VIEW > RAW SUMMARY.

Note the first and last dates of transactions and calculate the difference. If the difference is short, like 4 hours, then this is important to note. Now read down the report of hardware addresses and observe the integer numbers in parenthesis. Anything over 150 in this example 4 hour period should be called into HP for replacement.

[Cheryl Griffin]

The lbolt is nothing more than system ticks since the system has been up.

The device in question is dev: cb03f002. To determine what device this is quickly
# ll /dev/* |grep 3f002

Run ioscan. Look for No_hw, unknown, -l, not claimed, etc.

Check termination, cables, etc. including making sure they are the ones suggested/provided by the manufacturer depending on what equipment you have connected to the machine.

Open another case with HW Support and have them check the firmware of the equipment as well.

[Angus Crome]

If you have the Online Diagnostics installed (EMS/Predictive/etc), try upgrading to at least Sept 2002 (HP-UX 11.XX). There is also an included patch which should be installed.

The problem this solves is with specific types of hardware, and the output from SCSI-queries being mis-interpreted. This could also occur for devices that are off or improperly removed from active service.