[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference smurf::buildhelp

Title:USG buildhelp questions/answers
Moderator:SMURF::FILTER
Created:Mon Apr 26 1993
Last Modified:Mon Jan 20 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2763
Total number of notes:5802

1263.0. "Is this an ODE bug ?" by AOSG::FILTER (Automatic Posting Software - mail to flume::puck) Tue Feb 14 1995 19:48

Date Of Receipt: 	13-FEB-1995 10:15:13.14
From: 	SMURF::FLUME::"katz@decvax.dec.com" "13-Feb-1995 1004"
To: 	odehelp@flume.zk3.dec.com
CC: 	
Subj: 	Is this an ODE bug ?

I've entered this in a notes conference for the support group. It
saved me from annoying delays, but I would like you look into it
as it may be wrong to have to do this, or might cause some other
problems, thanks:

If you've accidentally checked out an unnecessary file you may
need to remember this.
If I remember correctly the bsubmit complains because if finds
a problem with the file that hasn't changed at all, and it
aborts.
So the first remedial step I tried was 'bcs -unlock foo.c'
This reported apparent success:
	rm: removing ./.BCSlock
However, a bstat -all still shows the file (and the bsubmit will
still try to include it too).

I have found that there is also a .BCSset-yourname_and_patch file
in the src directory. I edited this file and deleted the reference
to the unnecessary file. And the bsubmit went through.

Should this be necessary ? I've asked the ODE folks to look into
it. But in the meantime, you can complete your bsubmit if you
follow my workaround.

T.RTitleUserPersonal
Name
DateLines
1263.1Re: Is this an ODE bug ?AOSG::FILTERAutomatic Posting Software - mail to flume::puckTue Feb 14 1995 19:5659
Date Of Receipt: 	13-FEB-1995 11:10:14.88
From: 	SMURF::FLAMBE::"vandyck@zk3.dec.com" "Grant Van Dyck  13-Feb-1995 1108"
To: 	katz@decvax.dec.com
CC: 	odehelp@decvax.dec.com
Subj: 	Re: Is this an ODE bug ?

No, it's not a bug - Just pilot error.

When you check out a file, it gets you a numbered copy of the file by creating 
a private branch for you in RCS with you own number series and then locks the 
file for that branch. All files that you checkout in that sandbox have a 
branch setname consisting of your name and your sandbox name.

If you then try and bsubmit -all and haven't touched one of the files in the 
set, it'll complain because it isn't in the appropriate state to be submitted.

You need to either submit all the files EXCEPT that one so it doesn't complain 
about it, or remove it from your set. One way to remove it is to edit the 
setfile as you did, however this has the side effect of leaving the private 
branch you created for that file lingering in RCS. 

If you have no intention of using that file the correct thing to do is unlock 
the file AND OUTDATE YOUR BRANCH. This can be done in one step:

bcs -u -o file 

It will remove the RCS branch, the file copy in your sandbox and the reference 
to it in your setfile.


		-Grant



| I've entered this in a notes conference for the support group. It
| saved me from annoying delays, but I would like you look into it
| as it may be wrong to have to do this, or might cause some other
| problems, thanks:
| 
| If you've accidentally checked out an unnecessary file you may
| need to remember this.
| If I remember correctly the bsubmit complains because if finds
| a problem with the file that hasn't changed at all, and it
| aborts.
| So the first remedial step I tried was 'bcs -unlock foo.c'
| This reported apparent success:
| 	rm: removing ./.BCSlock
| However, a bstat -all still shows the file (and the bsubmit will
| still try to include it too).
| 
| I have found that there is also a .BCSset-yourname_and_patch file
| in the src directory. I edited this file and deleted the reference
| to the unnecessary file. And the bsubmit went through.
| 
| Should this be necessary ? I've asked the ODE folks to look into
| it. But in the meantime, you can complete your bsubmit if you
| follow my workaround.