[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
| Title: | Storage Library System |
|
| Moderator: | COOKIE::REUTER |
|
| Created: | Sun Oct 13 1991 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2270 |
| Total number of notes: | 7850 |
2214.0. "tapeposition counter wrong after fatal drive error" by UTRTSC::KNOL (isn't every bug a bit wrong) Mon Mar 10 1997 05:13
title : wrong tapepostion after fatal backup error
effected products sls 2.8 sls2.8a, sls2.9ft1
problem:
If the backup fails with a fatal error, the next backup
on the next volume doesn't start at tapeposition 0!!! but on
tapeposition 3
see below the sequense of events:
started with a backup of two disks,
then during backup powerfailed the tapedrive, backup fails
on that volume and continues on the next volume but on the wrong
tapeposition.
Solution would be to reset the tm to 0 before mounting the next
volume!
$!
$! SYSBAK summary file for KNOL_JANS_SBK
$!
$ DISK :== KITS
$ FILES :== KITS:[*...]*.*;*
$ FIRST :== TEST04
$ HISTORY :==
$ HISTSET :==
$ LAST :== TEST04
$ LISTING :==
$ NODE :== UTURBO
$ PARAMS :== KNOL_JANS_SBK
$ QUALS :== /IGNORE=INTERLOCK
$ SAVESET :== KITS.BAK
$ SCHINDX :==
$ STARTED :== 10-MAR-1997 12:37:52.74
$ STATUS == %X10A38044
$ STOPPED :== 10-MAR-1997 12:40:34.92
$ SUMFN :== SLS$ROOT:[SYSBAK.SUMMARY_FILES]KITS.SUM_199703101237_KNOL_JANS_SB1
$!
$ REEL_1 :== TEST04
$ N_REELS == 1
$!
$!
$ EXIT
$!
$! Messages from backup:
$!
PID 21800CE7
STARTED 10-MAR-1997 12:37:52.74
HISTORY
LISTING
Unable to allocate volume to follow
- %SLS-E-NOVOLS, not enough volumes available for request
Load volume on _UTURBO$MKA300: write enabled
- (first volume in volume set)
- (required tape pool: TESTJE)
- reply when ready, optionally with alternate volume name,
NEW to get a new volume name or QUIT to abort job
%SLS-I-STARTING, starting volume TEST04 at position 0 on drive _UTURBO$MKA300: 2
VOL1KITS 3
VOL1TEST04 3
%BACKUP-I-LBLOVRWRITE, volume label TEST04 overwritten, new label is TEST
HDR1KITS.BAK KITS 00010001000100 97069 97069 000000DECVMSBACKUP
HDR1KITS.BAK KITS 00010001000100 97069 97069 000000DECVMSBACKUP
HDR2F0819208192 M 00
HDR2F0819208192 M 00
%BACKUP-I-READYWRITE, mount volume 1 on _UTURBO$MKA300: for writing
Enter "YES" when ready:
YES
%BACKUP-I-NEXTVOL, volume 1 mounted - operation continues
******************************
power failed the tapedrive...................
******************************
%BACKUP-E-FATALERR, fatal error on _UTURBO$MKA300:[]KITS.BAK;
-SYSTEM-F-TIMEOUT, device timeout
%BACKUP-I-SPECIFY, specify option (QUIT or CONTINUE)
BACKUP>
QUIT
%SLS-E-BACKUPFAIL, backup directive failed
%BACKUP-F-ABORT, operator requested abort on fatal error
%SLS-I-FINISHED, finished volume TEST04 on drive _UTURBO$MKA300: at 10-MAR-19977
INIT_FIRST 1
DRIVE 1 _UTURBO$MKA300:
FIRST_REEL TEST04
LAST_REEL TEST04
END_TM 3 <----------------- our last tape position
STOPPED 10-MAR-1997 12:40:34.92
STATUS %X10A38044
PID 218010EA
STARTED 10-MAR-1997 12:41:25.28
HISTORY
LISTING
Unable to allocate volume to follow
- %SLS-E-NOVOLS, not enough volumes available for request
Load volume on _UTURBO$MKA300: write enabled
- (first volume in volume set)
- (required tape pool: TESTJE)
- reply when ready, optionally with alternate volume name,
NEW to get a new volume name or QUIT to abort job
Unable to allocate volume TEST04
- %SLS-F-NOTFREE, requested volume is not free
%SLS-E-VOLINQ, error inquiring about volume
-SLS-E-NOVOLAL, no such volume or not allocated to requestor
Unable to allocate volume to follow
- %SLS-E-NOVOLS, not enough volumes available for request
Load volume on _UTURBO$MKA300: write enabled
- (first volume in volume set)
- (required tape pool: TESTJE)
- reply when ready, optionally with alternate volume name,
NEW to get a new volume name or QUIT to abort job
%SLS-I-STARTING, starting volume TEST05 at position 3 on drive _UTURBO$MKA300: 5
^
^
see here the wrong position
for the new volume....
VOL1SPEC_U 3
VOL1TEST05 3
%BACKUP-I-LBLOVRWRITE, volume label TEST05 overwritten,
new label is TEST
HDR1SPEC_USER.BAK SPEC_U00010001000100 97069 97069 000000DECVMSBACKUP
HDR1SPEC_USER.BAK SPEC_U00010001000100 97069 97069 000000DECVMSBACKUP
HDR2F0819208192 M 00
HDR2F0819208192 M 00
%BACKUP-I-READYWRITE, mount volume 1 on _UTURBO$MKA300: for writing
Enter "YES" when ready:
YES
%BACKUP-I-NEXTVOL, volume 1 mounted - operation continues
regards,
jan
| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 2214.1 | | COOKIE::MCCLELLAND | Marty, SLS/MDMS Engineering | Fri Mar 28 1997 13:36 | 5 |
|
Engineering has received an ipmt case for this problem. Case
number is CFS.49835. Under investigation.
-Marty.
|
| 2214.2 | solution.. | COOKIE::MCCLELLAND | Marty, SLS/MDMS Engineering | Fri May 16 1997 13:50 | 55 |
|
Prepared by: Marty McClelland and Jay Rubenstein
SMS_CXO (SLS/MDMS) Engineering
Date: May 7,1997
Problem Statement:
During a multi-disk backup if job "A" fails with a fatal error.
Job B starts on a new tape at the end tape position of job "B".
This is incorrect behavour, job "B" should start with a new tape at
tape position zero.
Solution:
sls$system:sysbak.com needs to have one line add to
the failure code path. Add the correction by following these steps:
1) Copy sls$system:sysbak.com to
sls$custom:
2) edit sls$custom:sysbak.com
2a) search for the label "FATAL_NO_TRASH:"
$FATAL_NO_TRASH:
$ SET ON
$ IF REEL .NES. "" THEN NEXT_REEL :=
$ FIRST_REEL_'FINISHED' := ! forget about bad reels
$ LAST_REEL_'FINISHED' := ! ie, don't get next
from last
$ ! or try to
split set again
2b) Five lines down from the label "FATAL_NO_TRASH:"
add the following line
$ END_TM_'FINISHED' := 0 ! Start at begining
the end result follows
$FATAL_NO_TRASH:
$ SET ON
$ IF REEL .NES. "" THEN NEXT_REEL :=
$ FIRST_REEL_'FINISHED' := ! forget about bad reels
$ LAST_REEL_'FINISHED' := ! ie, don't get next
from last
$ ! or try to
split set again
$ END_TM_'FINISHED' := 0 ! start the next tape
at pos 0
2c) exit the editor
|