I have found what is causing the backup to complete with errors.
One of the backup items is a list of type *SPL that is supposed to save all files in a specific OUTQ. If this outq is empty the backup will complete "with errors", even though the backup item generates no messages at all in the logs. If I put a spooled file in the OUTQ and run the backup again it will complete normally.
This also explains why the backup completes normally on our old system where the outq contains multiple files. I noticed that on this machine there are also no messages about a succesful save of the spooled files.
Sounds to me like a bug in BRMS? An empty outq should not cause the backup to end with errors, or at the very least should include a message in the log.
Erwin
-----Oorspronkelijk bericht-----
Van: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> Namens Steinmetz, Paul via MIDRANGE-L
Verzonden: dinsdag 6 augustus 2019 16:12
Aan: 'Midrange Systems Technical Discussion' <midrange-l@xxxxxxxxxxxxxxxxxx>
CC: Steinmetz, Paul <PSteinmetz@xxxxxxxxxx>
Onderwerp: RE: BRMS backup completes with untraceable errors
Erwin,
I get the same BRM10A1 message, Control group SAVFULL05 type *BKU completed with errors.
I also submit the control group from AJS.
I do SWA on *ALLUSR, get a CPF3751 Some libraries not saved.
I also save all spoolfiles.
The reason for the two libraries not saved is due to open spoolfiles, QUSRSYS and QGPL partially saved.
SWA doesn't work for spoolfiles.
CPD33A8 Diagnostic 20 08/05/19 23:11:44.779270 QSPSAVQ QSYS 0976 QSRSVPRE QSYS 0E58
Message . . . . : Spooled file QPRINT number 2 in job
153814/CPAAWATS/EVUE_JOB not available for save.
Cause . . . . . : Spooled file QPRINT number 2 in job
153814/CPAAWATS/EVUE_JOB created on system PENCOR05 on 08/02/19 14:15:22 is
currently open and cannot be saved at this time. Recovery . . . : Try the
request again when the spooled file becomes available. The spooled file was
not saved from output queue CM_NIGHTLY in library QUSRSYS.
Cause . . . . . : 318 libraries were saved, 2 libraries were partially
saved, and 0 libraries were not saved. The save operation ended on volume
001344.
Because the spoolfiles are not saved, BRM16A1
Message . . . . : Backup *ALLUSR, type *FULL completed with errors.
For the *LINK save,
I get CPD37C4 Message . . . . : 57 objects not saved.
Cause . . . . . : 52 objects were not saved due to the ALWSAV attribute set
to *NO. A CPD37C3 message was sent for each of those objects. 5 objects
were not saved because the system has indicated they should not be saved.
Also get a CPF3837
Message . . . . : 1831831 objects saved. 28 not saved.
Cause . . . . . : 1831831 objects were saved on volumes 001344 sequence
number 301605 at 08/06/19 00:05:00. 28 objects were not saved. The save
operation ended on volume 001344.
As Rob stated, SWA is worthless for IFS.
My IFS objects not saved are mostly /www objects in use, mostly log files.
Paul
De disclaimer van toepassing op e-mail van de gemeente Den Haag vindt u op:
http://www.denhaag.nl/disclaimer
midrange.com
