RE: BRMS recovery report - recovery includes objects which were not saved - preparing for a P7 to P9 mibration

Rob,

I have a similar doc, BRMS - IFS folders set not to be saved.
It contains:
NWS
HTTP log files
Tmp
QTCPTMM
etc

Basically any IFS object that is locked during a normal save.
We use Save While Active, but as we all know, this does not work/apply for IFS objects.
Thus why they get skipped.

Do you know, is there a simple way of finding any IFS object marked with Can be saved . . . . . . . . . . . . . : No?

Ah, you mentioned Omit lists in BRMS, specially QLNKOMT list.
I found a serious problem with QLNKOMT back in 2012,
Long PMR with IBM, not sure if the issue every got resolved.
I think it required a DCR back then, now an RFE.

When using QLNKOMIT list, and if ignoring them on a system save control group
Omits . . . . . . . . . . . . . > *IGNORE

On normal saves they were *PROCESSED.
Omits . . . . . . . . . . . . . *PROCESS

When doing a BRMS recovery, BRMS recovery failed to use these items.
One of the reasons my BRMS recoveries were failing.

Paul






-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Rob Berendt
Sent: Wednesday, November 07, 2018 11:02 AM
To: Midrange Systems Technical Discussion
Subject: Re: BRMS recovery report - recovery includes objects which were not saved - preparing for a P7 to P9 mibration

<snip>
How do others in the group handle IFS items "not saved" when preparing for
a P7 to P9 migration?
</snip>

I read the report. As you did, I would have noticed the
CPD37C3 - Cannot save /QFPNWSSTG.
And then I would have realized that I just saved my hosting lpar. And I
generally do not restore all my guests when I restore my hosting lpar so
this is all right.
If, however, you really wanted to restore the guests also then before the
save I would have ran
CHGATR OBJ('/QFPNWSSTG') ATR(*ALWSAV) VALUE(*YES) SUBTREE(*ALL)
I have done this, on RARE occasions.
Actually I have a
CHGATR OBJ('/QFPNWSSTG') ATR(*ALWSAV) VALUE(*NO) SUBTREE(*ALL)
right before my save in my documentation. This reminds me to change it to
*NO in case I ever changed it in the past. It also reminds me that if
this is one of those rare occasions that I really want to save it then I
should probably change this.
I also do a
CHGATR OBJ('/fixes') ATR(*ALWSAV) VALUE(*NO) SUBTREE(*ALL)
because /fixes is where my PTF and OS image catalogs reside. Again, there
are some rare exceptions where I want to save those also.

This is totally different than using Omit lists in BRMS. Omit lists are
easy to flag to ignore by simply using a parameter
STRBKUBRM CTLGRP(...) OMITS(*IGNORE)...
But this still will not save any directory with *ALWSAV turned off.


Rob Berendt