| Date Of Receipt: 4-OCT-1996 17:22:18.05
From: FLUME::"[email protected]" "Susan Schueller USG 04-Oct-1996 1718"
To: "Joshua M. Friedman, Digital UNIX, 381-1548" <[email protected]>
CC: [email protected], [email protected], [email protected]
Subj: Re: bsubmit problem [in ptbqa]
Josh,
Thanks for carrying this forward.
Sue
> Brian, here are some items to check:
>
> There is not a command switch -time on bsubmit, so I'm guessing
> you really didn't do this exactly as you said:
> bsubmit -resub -time <time>
>
> Since it's now a new date, the appropriate command for your resub
> would be:
> workon -sb <sandboxname>
> (now cd to sandbox src if you're not already there)
> bsubmit -resub 16:34 -date 10/3/96
>
> There is a problem which sometimes crops up in which a stopped &
> resumed bsubmit will encounter "missing" files because the copy in
> place has been removed and not yet replaced; the phase of the submit
> determines what state they should be in (read-only or read-write),
> however if you can check the full list of 361 files and for any that
> are "absent" re-check them out using
>
> bco -u <file> <file>...
>
> After all the 361 files are in place re-try the bsubmit -resub
> as above.
>
> If you still have problems, it may be necessary for us to examine the
> contents of your 16:34.Brian_Baker and 16:34.bcssteps files; these
> preserve the state information for the bsubmit.
>
> Regarding Jeff Detjen's files in the Thold file; this is normal and
> just shows other submit activity that happened to be pending in the
> same pool; you can ignore this, however your name and file list is also
> in this hold file as follows (it was below Jeff's entry):
>
> : Brian_Baker; Date: 10/3/96; Time: 16:34
> ./test/filesystems/suites/Makefile.ode
> ./test/filesystems/suites/advfs/Makefile.ode
> ....
>
> At any time you can do "sadmin -all | more" to see the current contents
> of what's on hold (contents of ptbqa/logs/bsubmit.hold file).
>
> I've checked the state in rcs of the first and last files in your list
> and it appears that you can actually safely clear the state of the
> bsubmit and begin it again if you like, since you don't have any
> "submit branch locks".
>
> Here's how you can remove the submit (it won't affect your bci'd files;
> they'll still be in the bci'd state ready to bsubmit):
>
> sadmin -rm 16:34 -date 10/3/96
>
> You can also do the previously mentioned sadmin -all before and after
> this sadmin -rm just to confirm to you that your submit was indeed in
> the hold queue and is afterward no longer on hold.
>
> You will still need to check if any of the 361 files are missing and
> replace them with a "bco -u" which will just put back your last bci
> and leave the file read-only.
>
> -josh
>
>
> > ------- Forwarded Message
> >
> > From: Baker Brian USG <[email protected]>
> > Message-Id: <[email protected]>
> > Subject: bsubmit problem
> > To: [email protected]
> > Date: Fri, 4 Oct 1996 10:04:39 -0400 (EDT)
> >
> > Hi,
> >
> > I'm having some bsubmit difficulties with which I need assistance.
> >
> > I've bcreated and bcied about 363 files for ptbqa. Upon bsubmitting
> > (bsubmit -l -all), I received a couple errors. Initially they were
> > somewhat minor (I had files set for write and bsubmit did not like that).
> >
> > At one point, bsubmit said there were errors and it was exiting, but I had
> > the option to preserve my bsubmit state. I selected preserve. On a
> > subsequent bsubmit (bsubmit -resub -time <time>) I resolve the error
> > (missing copyright tag in one file), but I received additional errors that
> > I've not seen before:
> >
> > >> FATAL ERROR in /usr/sde/tools/alpha_osf1/bin/ode/bsubmit:
> > >> open ./test/filesystems/suites/cthon96/tools/Makefile
> >
> > There were a dozen or so of these. I suspected (and hoped) that they
> > were due to the directory above the files not having write access. I
> > set the directory to write access and tried to bsubmit again with
> > bsubmit -resub -time <time>, but received the same set of errors.
> >
> > I noticed some additional files in ./src in my sandbox.
> > 16:34.bcssteps
> > Appears to contain files from .BCSconfig
> > 16:34.Thold
> > This file contains names of files I am NOT checking in. In fact,
> > Jeff Detjen's name is in this file. Don't know if that's good or
> > bad.
> >
> > Any assistance would be appreciated.
> >
> > Brian
> >
> >
> > ------- End of Forwarded Message
> >
--
------------------------------------------------------------
Susan Schueller Digital Equipment Corporation
USG Release Engineering MS ZKO3-3/W20
(603) 881-6348 110 Spit Brook Road
[email protected] Nashua, NH 03062-2698
------------------------------------------------------------
|
| Date Of Receipt: 7-OCT-1996 11:09:32.66
From: FLUME::jmf "Joshua M. Friedman Digital UNIX"
To: bbaker@DEC:.zko.flume
CC: odehelp@DEC:.zko.flume
Subj: Re: bsubmit problem [in ptbqa]
The files in the bcssteps file are the files which succeeded in the
current phase of the submit. So the missing ones were the ones that
had some problem associated with them.
By the way, before you restart a new bsubmit, you can use bsubmit -info ...
which will check things and warn you of merges without actually doing
the submit work.
-josh
-----------
From [email protected] Fri Oct 4 17:34:13 1996
Delivery-Date: Fri, 04 Oct 96 17:34:15 -0400
From: Baker Brian USG <[email protected]>
Subject: Re: bsubmit problem [in ptbqa]
To: [email protected] (Joshua M. Friedman, Digital UNIX, 381-1548)
Date: Fri, 4 Oct 1996 17:34:33 -0400 (EDT)
In-Reply-To: <[email protected]> from "Joshua M. Friedman, Digital UNIX, 381-1548" at Oct 4, 96 05:09:11 pm
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text
Content-Length: 534
Hi,
Thanks for the info. You are correct, I was overzealous in my typing.
The command I used was bsubmit -resub <time> (no -time switch).
I'm going to go the removal route.
Out of curiosity...
The 16:34.bcssteps file contains the majority of my files (297 out of the
361). Are the files in 16:34.bcssteps those that bsubmit got to or those
it didn't get to? The "absent" files you referred to...are those the
files that are _not_ in the 16:34.bcssteps file?
Thanks for your help, I really appreciate it.
Brian
|