About two weeks ago, I posted a question about troubles with an
HP35480A tape drive connected to a DEC 2000 Model 300. This note
contains 1) thanks, 2) the suggestions so far, and 3) more detail on
the problem. Once again, any --additional-- help would be greatly
appreciated.
1) Thanks for all of the quick responses:
"Bill Wang" <wcwang_at_cs.indiana.edu>
"Dr. Tom Blinn, 603-881-0646" <tpb_at_zk3.dec.com>
"Mark F. Rondinaro" <mfr_at_lns598.lns.cornell.edu>
golden_at_falcon.invincible.com
David Giesen <giesen_at_chemsun.chem.umn.edu>
2) Suggestions included: a bad cable, a drive design flaw,
. . . and a request for more data (uerf -o full):
(responses edited)
____________
From: "Bill Wang" <wcwang_at_cs.indiana.edu>
I had the exactly same kind of problem with a different kind of HP tape
drive, it turns out to be a "non-comforming" SCSI cable to the tape drive.
____________
From: "Dr. Tom Blinn, 603-881-0646" <tpb_at_zk3.dec.com>
I bet you got a "medium error". I have seen them on my Digital TLZ06, which
may well be the same drive as the HP 35480A DAT tape drive. I'm not sure if
we sell the same drive or not.
On the TLZ06, the problem is caused when you get a write error (the drives
do read after write verification), and backspace to re-write the record. On
the TLZ06, there is a mechanical design problem that sometimes results in
the tape being slightly out of alignment when it's rewritten. The record is
re-written and read back just fine, so the drive keeps going, but when you
try to read the tape later on, you can't, because the record was in fact bad
(because the tape was skewed in the write path when it was written, this is
the mechanical design flaw). The "fix" (maybe) is new firmware that backs
up the tape several records, spaces forward, then re-writes. This should be
good enough to get the tape aligned again, but in my experiences it leaves a
bit to be desired.
Tapex will not uncover this problem in my experience -- I don't think that
it is likely to make it happen, anyway, as the error that causes the drive
to re-write the record is often intermittent.
____________
From: "Mark F. Rondinaro" <mfr_at_lns598.lns.cornell.edu>
This is a response to the message you posted to the alpha-osf-managers list.
The error log entries included with your message don't look complete. I would
suggest running uerf with the "-o full" flag and then reposting the message
with the complete error log entry. A full error log entry should report the
SCSI sense data from the error; this is often helpful in diagnosing problems.
(others had the same request. . .see below)
____________
____________
3) More information: I have had several backup jobs fail recently.
The backup jobs (locally written C-shell scripts) take tar files
created by several local PCs on the Alpha, gzip them, and then tar
them to the tape drive (I also tried cpio'ing them to the tape drive
with the same result). The backup log files generate the messages:
example a)
BRDC tape backup listing
Sat Jul 8 08:59:14 EDT 1995
<stuff deleted>
Checking if compressed files will fit on a tape (max 2000 MB) . . .
The total size is 1984.57
The tar files will all fit on one tape.
edgeworth_c.tar.z
edgeworth_d.tar.z
edgeworth_e.tar.z
edgeworth_f.tar.z
edgeworth_g.tar.z
edgeworth_h.tar.z
cpio: /dev/tty Unavailable
<stuff deleted>
Sun Jul 9 07:44:31 EDT 1995
example b)
BRDC tape backup listing
Fri Jun 30 16:45:01 EDT 1995
<stuff deleted>
Checking if compressed files will fit on a tape (max 2000 MB) . . .
The total size is 1919.99
The tar files will all fit on one tape.
a edgeworth_c.tar.z 246122 Blocks
a edgeworth_d.tar.z 412166 Blocks
a edgeworth_e.tar.z 558559 Blocks
a edgeworth_f.tar.z 99566 Blocks
a edgeworth_g.tar.z 368243 Blocks
a edgeworth_h.tar.z 459021 Blocks
tar: /dev/tty Unavailable
<stuff deleted>
Sat Jul 1 04:26:51 EDT 1995
Corresponding to these failures in the backup log files are the uerf
events:
******************************** ENTRY 1. *********************************
----- EVENT INFORMATION -----
EVENT CLASS ERROR EVENT
OS EVENT TYPE 199. CAM SCSI
SEQUENCE NUMBER 1.
OPERATING SYSTEM DEC OSF/1
OCCURRED/LOGGED ON Sun Jul 9 07:42:21 1995
OCCURRED ON SYSTEM MARSHALL
SYSTEM ID x00580006 CPU TYPE: DEC 2000
SYSTYPE x00000000
----- UNIT INFORMATION -----
CLASS x0001 TAPE
SUBSYSTEM x0000 DISK
BUS # x0000
x0018 LUN x0
TARGET x3
----- CAM STRING -----
ROUTINE NAME ctape_wfm
----- CAM STRING -----
ERROR TYPE Hard Error Detected
----- CAM STRING -----
DEVICE NAME UNKNOWN
----- CAM STRING -----
Active CCB at time of error
----- CAM STRING -----
CCB request completed with an error
ERROR - os_std, os_type = 11, std_type = 10
----- ENT_CCB_SCSIIO -----
*MY ADDR x855D8728
CCB LENGTH x00C0
FUNC CODE x01
CAM_STATUS x00C4 CAM_REQ_CMP_ERR
SIM QFRZN
AUTOSNS_VALID
PATH ID 0.
TARGET ID 3.
TARGET LUN 0.
CAM FLAGS x000000C0
CAM_DIR_NONE
*PDRV_PTR x855D8428
*NEXT_CCB x00000000
*REQ_MAP x00000000
VOID (*CAM_CBFCNP)() x004ADF00
*DATA_PTR x00000000
DXFER_LEN x00000000
*SENSE_PTR x855D8450
SENSE_LEN x40
CDB_LEN x06
SGLIST_CNT x0000
CAM_SCSI_STATUS x0002 SCSI_STAT_CHECK_CONDITION
SENSE_RESID x2D
RESID x00000000
CAM_CDB_IO x000000000000000200000010
CAM_TIMEOUT x0000012E
MSGB_LEN x0000
VU_FLAGS x0000
TAG_ACTION x00
----- CAM STRING -----
Error, exception, or abnormal
_condition
----- CAM STRING -----
NO SENSE - No error or no sense
_information
----- ENT_SENSE_DATA -----
ERROR CODE x0070 CODE x70
SEGMENT x00
SENSE KEY x0000 NO SENSE
INFO BYTE 3 x00
INFO BYTE 2 x00
INFO BYTE 1 x00
INFO BYTE 0 x00
ADDITION LEN x0B
CMD SPECIFIC 3 x00
CMD SPECIFIC 2 x00
CMD SPECIFIC 1 x00
CMD SPECIFIC 0 x00
ASC x00
ASQ x02
FRU x00
SENSE SPECIFIC x000000
ADDITIONAL SENSE
0000: 00000000 00000000 00000000 00000000 *................*
0010: 00000000 00000000 00000000 00000000 *................*
0020: 00000000 00000000 00000000 00000000 *................*
0030: 7E250000 00005E3C 00000000 00000000 *..%~<^..........*
******************************** ENTRY 2. *********************************
----- EVENT INFORMATION -----
EVENT CLASS ERROR EVENT
OS EVENT TYPE 199. CAM SCSI
SEQUENCE NUMBER 17.
OPERATING SYSTEM DEC OSF/1
OCCURRED/LOGGED ON Sat Jul 1 04:24:46 1995
OCCURRED ON SYSTEM MARSHALL
SYSTEM ID x00580006 CPU TYPE: DEC 2000
SYSTYPE x00000000
----- UNIT INFORMATION -----
CLASS x0001 TAPE
SUBSYSTEM x0000 DISK
BUS # x0000
x0018 LUN x0
TARGET x3
----- CAM STRING -----
ROUTINE NAME ctape_wfm
----- CAM STRING -----
ERROR TYPE Hard Error Detected
----- CAM STRING -----
DEVICE NAME UNKNOWN
----- CAM STRING -----
Active CCB at time of error
----- CAM STRING -----
CCB request completed with an error
ERROR - os_std, os_type = 11, std_type = 10
----- ENT_CCB_SCSIIO -----
*MY ADDR x854E1B28
CCB LENGTH x00C0
FUNC CODE x01
CAM_STATUS x00C4 CAM_REQ_CMP_ERR
SIM QFRZN
AUTOSNS_VALID
PATH ID 0.
TARGET ID 3.
TARGET LUN 0.
CAM FLAGS x000000C0
CAM_DIR_NONE
*PDRV_PTR x854E1828
*NEXT_CCB x00000000
*REQ_MAP x00000000
VOID (*CAM_CBFCNP)() x004AD6D0
*DATA_PTR x00000000
DXFER_LEN x00000000
*SENSE_PTR x854E1850
SENSE_LEN x40
CDB_LEN x06
SGLIST_CNT x0000
CAM_SCSI_STATUS x0002 SCSI_STAT_CHECK_CONDITION
SENSE_RESID x2D
RESID x00000000
CAM_CDB_IO x000000000000000200000010
CAM_TIMEOUT x0000012E
MSGB_LEN x0000
VU_FLAGS x0000
TAG_ACTION x00
----- CAM STRING -----
Error, exception, or abnormal
_condition
----- CAM STRING -----
NO SENSE - No error or no sense
_information
----- ENT_SENSE_DATA -----
ERROR CODE x0070 CODE x70
SEGMENT x00
SENSE KEY x0000 NO SENSE
INFO BYTE 3 x00
INFO BYTE 2 x00
INFO BYTE 1 x00
INFO BYTE 0 x00
ADDITION LEN x0B
CMD SPECIFIC 3 x00
CMD SPECIFIC 2 x00
CMD SPECIFIC 1 x00
CMD SPECIFIC 0 x00
ASC x00
ASQ x02
FRU x00
SENSE SPECIFIC x000000
ADDITIONAL SENSE
0000: 00000000 00000000 00000000 00000000 *................*
0010: 00000000 00000000 00000000 00000000 *................*
0020: 00000000 00000000 00000000 00000000 *................*
0030: 7E250000 00005E3C 00000000 00000000 *..%~<^..........*
Thanks,
Russ Murphy
Boston Research Data Center
Center for Economic Studies
US Bureau of the Census
2 Copley Place
Boston, Massachusetts
(617) 424-0550
rmurphy_at_census.gov
rdmurphy_at_acs.bu.edu
Received on Wed Jul 12 1995 - 00:29:14 NZST