Intermittent syslog errors- SCSI: isrEscape Controller at 0/1/1/0

[JDR45]

The server is an RP3440 running HP-UX 11.11.  Off an on for a month or so we've been seeing errors in the syslog on the SCSI isrEscape Controller at 0/1/1/0.  When it happens the systems hangs for 2 - 4 minutes but it never kicks the users off or causes other problems.  The server reboots via cron twice a week and the reboots always go fine.

The SCSI card at 0/1/1/0 has two 300GB hard drives on it- 0/1/1/0.0.0 and 0/1/1/0.1.0.  These two are a mirrored pair.  The mirror has always stayed in sync when I check with vgdisplay -v | more.  Both hard drives have always stayed CLAIMED on an ioscan -fnC disk.  Sometimes the system will generate a EMS message for both those two drives, but as far as I can tell both drives seem to be fine.

The server has a third 300GB hard drive at 0/1/1/1.2.0.  This drive is always fine and I never see any errors related to the 0/1/1/1 bus.

0/1/1/0 looks like this- driver c8xx SCSI c1010 Ultra160 Wide LVD

It's December now, but back in August we saw some very similar, if not identical error messages.  Back then we replaced the SCSI card that the three hard drives plug into.  This gave us about three months of error free operation.  I wonder if maybe the replacement card was used, old, or reconditioned.  Maybe the same card just needs to be replaced again?

Here is a look at the messages fromthe syslog.  Is there anything more I can do to narrow the problem down?  Thanks in advance.

---

Dec 13 10:56:04 vmunix: SCSI: isrEscape Controller at 0/1/1/0.
Dec 13 10:56:04 vmunix:
Dec 13 10:56:04 vmunix: SCSI: Parity error -- lbolt: 2523535, dev: 1f0200
00
Dec 13 10:56:04 vmunix: lbp->state: 30008
Dec 13 10:56:04 vmunix: lbp->offset: ffffffff
Dec 13 10:56:04 vmunix: lbp->nominalOffset: 270
Dec 13 10:56:04 vmunix: lbp->Cmdindex: 6
Dec 13 10:56:04 vmunix: lbp->last_nexus_index: 53
Dec 13 10:56:04 vmunix: lbp->nexus_index: 54
Dec 13 10:56:04 vmunix: uCmdSent: 600d980 uNexus_offset: 6d750
Dec 13 10:56:04 vmunix: last lbp->puStatus [0000000041e85730]:
Dec 13 10:56:04 vmunix: 0003007c 00030073 00030079 00030077
Dec 13 10:56:04 vmunix: next lbp->puStatus [0000000041e85740]:
Dec 13 10:56:04 vmunix: 0003007a 00030074 0003007a 00030074
Dec 13 10:56:04 vmunix: From most recent interrupt:
Dec 13 10:56:04 vmunix: ISTAT: 0a, SIST0: 91, SIST1: 00, DSTAT:
80, DSPS: 00000000
Dec 13 10:56:04 vmunix: lsp: 0x00000000406de600
Dec 13 10:56:04 vmunix: bp->b_dev: 1f020000
Dec 13 10:56:04 vmunix: scb->io_id: 2216027
Dec 13 10:56:04 vmunix: scb->cdb: 28 00 13 cc f3 b0 00 00 10 00
Dec 13 10:56:04 vmunix: lbolt_at_timeout: 0, lbolt_at_start: 0
Dec 13 10:56:04 vmunix: lsp->state: 4005
Dec 13 10:56:04 vmunix: Jump Table entry [ffffffff90012cf0]: ff010074 00
069000
Dec 13 10:56:04 vmunix: lsp->puScript [0000000041e81000]:
Dec 13 10:56:04 vmunix: 09001000 1bfd2000 00494002 09001000
Dec 13 10:56:04 vmunix: 0c335000 00490002 78370000 00000000
Dec 13 10:56:04 vmunix: DSAtbl->host_iocb_index: 6
Dec 13 10:56:04 vmunix: DSAtbl->host_iocb_addr: 6d980
Dec 13 10:56:04 vmunix: stored scratcha: 0xff030074
Dec 13 10:56:04 vmunix: scratch_lsp: 0x00000000406de600
Dec 13 10:56:04 vmunix: c8xx_iocb [ffffffff90012b00]:
Dec 13 10:56:04 vmunix: 0600d980 ff000074 00069000 9f001f80
Dec 13 10:56:04 vmunix: 00000003 0006d960 0000000a 0006d968
Dec 13 10:56:04 vmunix: Pre-DSP script dump [0000000041e81000]: 09001000
Dec 13 10:56:04 vmunix: Script dump [0000000041e81004]:
Dec 13 10:56:04 vmunix: 1bfd2000 00494002 09001000 0c335000
Dec 13 10:56:04 vmunix: 00490002 78370000 00000000 80080000
Dec 13 10:56:04 vmunix: NCR chip register dump for: 0x400200a
Dec 13 10:56:04 vmunix: 00: SCNTL3: 9f SCNTL2: 80 SCNTL1:
10 SCNTL0: da
Dec 13 10:56:04 vmunix: 04: GPREG: 0e SDID: 00 SXFER:
1f SCID: 47
Dec 13 10:56:04 vmunix: 08: SBCL: ae SSID: 80 SOCL:
0d SFBR: 01
Dec 13 10:56:04 vmunix: 0c: SSTAT2: 09 SSTAT1: 0e SSTAT0:
01 DSTAT: 80
Dec 13 10:56:04 vmunix: 10: DSA: 90012b00
Dec 13 10:56:04 vmunix: 14: MBOX1: 00 MBOX0: 00 ISTAT1:
00 ISTAT: 08
Dec 13 10:56:04 vmunix: 1c: TEMP: 00069000
Dec 13 10:56:04 vmunix: 24: DCMDDBC: 0d000600
Dec 13 10:56:04 vmunix: 28: DNAD: 1bfd2a00
Dec 13 10:56:04 vmunix: 2c: DSP: 0006900c
Dec 13 10:56:04 vmunix: 30: DSPS: 1bfd2000
Dec 13 10:56:04 vmunix: 34: SCRATCHA: ff030074
Dec 13 10:56:04 vmunix: 38: DCNTL: a1 DWT: 00 DIEN:
7f DMODE: 4c
Dec 13 10:56:04 vmunix: 3c: ADDER: 0003b00c
Dec 13 10:56:04 vmunix: 40: SIST1: 00 SIST0: 00 SIEN1: 

[JDR45]

Well our luck ran out.  The occasional 3 minute hang-ups that we would see related to these syslog messages finally turned into a 30 minute monster lock up.  Had to use the console cart to reboot the server.  Still investigating.  

[JDR45 · Accepted solution]

Would it be possible to have this locked or closed?

The server that was putting up all the error messages is bascially dead.  We took the hard drives out and moved them into a new chassis.  All is well, and we're just going to write off the old server as beyond repair.

Thanks!

[Dennis Handly]

> Would it be possible to have this locked or closed?

You can mark your reply as a "solution", using the Options menu on the right.