[p4] cvs2p4: troubleshooting [signed]

Thierry Lam Thierry.Lam at autodesk.com
Mon Nov 19 10:03:04 PST 2007 

I'm trying to run the test which came with cvs2p4-3.0b6 along with the patched rlog and I'm getting the following errors, any idea what I'm missing:

runtest> rm -rf /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir
runtest> mkdir /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir
runtest> cp /home/perforce/tools/cvs2p4-3.0b6/test/config.test /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir/config
runtest> cp /home/perforce/tools/cvs2p4-3.0b6/test/brmap.pl /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir
runtest> cp /home/perforce/tools/cvs2p4-3.0b6/test/exclude_branches /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir
runtest> cp /home/perforce/tools/cvs2p4-3.0b6/test/exclude_tags /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir
runtest: create bad filename "\002"
runtest: create bad filename "x\001y"
runtest: create bad filename "bad\177\377\324\003\004\005dab"
runtest > chdir /home/perforce/tools/cvs2p4-3.0b6
runtest> bin/genmetadata test_conv_dir
genmetadata: loaded 0 tag exclusions from test_conv_dir/exclude_tags.
genmetadata: loaded 0 branch tag exclusions from test_conv_dir/exclude_branches.
eval'ed test_conv_dir/brmap.pl ok.
genmetadata> /bin/rm -rf test_conv_dir/logmsgs test_conv_dir/logmsgs.db test_conv_dir/logmsgs.pag test_conv_dir/logmsgs.dir test_conv_dir/p4root test_conv_dir/changes test_conv_dir/p4 test_conv                                             _dir/tags.txt test_conv_dir/tags test_conv_dir/tags.db test_conv_dir/tags.pag test_conv_dir/tags.dir test_conv_dir/tagfiles.txt test_conv_dir/tagfiles test_conv_dir/tagfiles.db test_conv_dir/ta                                             gfiles.pag test_conv_dir/tagfiles.dir test_conv_dir/brtags.txt test_conv_dir/brtags.db test_conv_dir/brtags.pag test_conv_dir/brtags.dir test_conv_dir/revmap test_conv_dir/revmap.db test_conv_d                                             ir/revmap.pag test_conv_dir/revmap.dir test_conv_dir/labels
========== test/file,vrlog: output error: Invalid argument
rlog aborted
 ok
genmetadata: RCS filename with non-printable characters (skipped): bad\177\377\324\003\004\005dab,v
========== test/at at file,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/percent_%file,vrlog: output error: Invalid argument
rlog aborted
 ok
genmetadata: RCS filename with non-printable characters (skipped): x\001y,v
genmetadata: RCS filename with non-printable characters (skipped): \002,v
========== test/space file,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/datefile_readd,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/dollar$file,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/phone.gif,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/pound#file,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/Attic/datefile,vrlog: output error: Invalid argument
rlog aborted
 ok
========== test/star*file,vrlog: output error: Invalid argument
rlog aborted
 ok
Wrote test_conv_dir/tags.txt
genmetadata> rm -f test_conv_dir/tags test_conv_dir/tags.db test_conv_dir/tags.pag test_conv_dir/tags.dir
genmetadata> rm -f test_conv_dir/tagfiles test_conv_dir/tagfiles.db test_conv_dir/tagfiles.pag test_conv_dir/tagfiles.dir
Wrote test_conv_dir/brtags.txt
genmetadata> rm -f test_conv_dir/brtags test_conv_dir/brtags.db test_conv_dir/brtags.pag test_conv_dir/brtags.dir
===== Lines referenced:
cabesa
chupa
curly
curly2
curly3
ha
larry
vtag
xxx
yyy
runtest> diff /home/perforce/tools/cvs2p4-3.0b6/test/lines.good /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir/lines
9d8
< shemp
runtest: *** "diff /home/perforce/tools/cvs2p4-3.0b6/test/lines.good /home/perforce/tools/cvs2p4-3.0b6/test_conv_dir/lines" exited with signal 0 status 1
runtest: *** genmetadata bad results
runtest> rm -f /home/perforce/tools/cvs2p4-3.0b6/test/,v
runtest> rm -f /home/perforce/tools/cvs2p4-3.0b6/test/xy,v
runtest> rm -f /home/perforce/tools/cvs2p4-3.0b6/test/baÿÔdab,v

-----Original Message-----
From: perforce-user-bounces at perforce.com [mailto:perforce-user-bounces at perforce.com] On Behalf Of Gabor Maghera
Sent: Friday, November 16, 2007 5:10 PM
To: Mamadou Cisse [c]
Cc: perforce-user at perforce.com
Subject: Re: [p4] cvs2p4: troubleshooting [signed]

It may be due to the metadata associated with the Power Point files, but I'm
not sure.  Some of the scripts have a verbosity flag you can set, it may
help you zero in on the issue.  You could also try stepping through the
script in a Perl debugger.

Gabor

On Nov 16, 2007 8:00 AM, Mamadou Cisse [c] <mcisse at acquisinc.com> wrote:

> Gabor,
> I've manage to get the version cvs2p4-2.5 somehow working except that
> it's throwing an rcstok(): "Internal Error" on some MS Point Point
> Presentation files. I've included the ppt extension in the config file
> but that doesn't seem to be helping.
> Any ideas?
>
> Regards,
> MCisse.
>
> Gabor Maghera wrote:
> > Running Perforce without a commercial license should have no effect.  By
> > default you get a 2 user/5 client license for free, which is more then
> > sufficient for the conversion.  You could run into a problem if you have
> two
> > users defined already and now you're trying to run the conversion with a
> > third.  The config file has a setting for this, if it is not set to a
> > specific user, then it will try to use the UNIX/domain account you are
> > running the cvs2p4 script with.  To aire on the side of caution, I would
> > define the p4 user ahead of running the script, then set the config file
> to
> > use that account.
> >
> > On the other hand, if you haven't set up any users yet, chances are you
> are
> > below the two user limit, so this may be a moot point.
> >
> > Good luck,
> > Gabor
> >
> > On Nov 15, 2007 9:18 AM, Mamadou Cisse [c] <mcisse at acquisinc.com> wrote:
> >
> >> Yeah.
> >> I have written a tiny PERL script which has gotten rid of the cvsNT
> >> extra luggage (filename, deltatype, commentid, ...).
> >> But still, I can't even get the test script to run correctly.
> >> It's always finishing with a diff exited with signal 0 status 1 error.
> >>
> >> Since this is finding fact endeavor I'm running the perforce without a
> >> license and I'm wondering if that could influence the outcome.
> >> Any ideas?
> >>
> >> I'm using the p4d server release 05.2 on ubuntu dapper machine.
> >>
> >> Regars,
> >> MCisse.
> >>
> >> Smith, Jeff wrote:
> >>> As I recall, CVSNT uses a slightly different RCS format than standard
> >>> CVS which confuses cvs2p4.  I can't remember the exact differences but
> >>> when I made a similar conversion a few years ago, I had to preprocess
> >>> all the rcs files first.
> >>>
> >>> Jeff Smith
> >>> Development Tools Group
> >>> Managing Your Development Environment
> >>>
> >>>
> >>> -----Original Message-----
> >>> From: perforce-user-bounces at perforce.com
> >>> [mailto:perforce-user-bounces at perforce.com] On Behalf Of Gabor Maghera
> >>> Sent: Wednesday, November 14, 2007 12:58 PM
> >>> To: Mamadou Cisse [c]
> >>> Cc: perforce-user at perforce.com
> >>> Subject: Re: [p4] cvs2p4: troubleshooting [signed]
> >>>
> >>> The error log indicates that it may be running the Perforce server
> (p4d)
> >>> which is failing, possibly due to bad metadata generated with
> >>> genmetadata.
> >>>
> >>> This line:
> >>> */usr/bin/p4d -r . -jr dbmeta
> >>> *attempts to start p4d and restore metadata from checkpoint and/or
> >>> journal.
> >>>
> >>> I would test the install of p4d separately from running the cvs2p4
> >>> script, to make sure it is OK.  The fact that the test conversion
> >>> provided with the script is failing is an indication of a
> >>> configuration/environment issue.
> >>>
> >>> Cheers,
> >>> Gabor
> >>> *
> >>> *On Nov 14, 2007 2:25 AM, Mamadou Cisse [c] <mcisse at acquisinc.com>
> >>> wrote:
> >>>
> >>>> Yes, I did!
> >>>>
> >>>> Gabor Maghera wrote:
> >>>>> Have you recompiled RCS 5.7 to create the patched/modified version
> >>>>> of
> >>>> rlog?
> >>>>> Gabor
> >>>>>
> >>>>> On Nov 13, 2007 9:02 AM, Mamadou Cisse [c] <mcisse at acquisinc.com>
> >>> wrote:
> >>>>>> I'm trying to use cvs2p4 to migrate a cvsNT repository to Perforce
> >>>>>> on a Ubuntu Dapper box.
> >>>>>> I couldn't get the test/runtest script to run properly. It always
> >>>>>> came back with a "bad results" from the bin/genmetadata script.
> >>>>>> Being unable to find out why, I've decided to try my luck with the
> >>>>>> bin/cvs2p4 script. This one is throwing the error:
> >>>>>> [[
> >>>>>> Journal file 'dbmeta' replay failed at line <num>!
> >>>>>> Word too big fo a number!
> >>>>>> ]]
> >>>>>> I've been scouring the web for answers for the last week or so and
> >>>>>> couldn't find any clear understanding of the causes or the
> >>> remedies.
> >>>>>> Can anybody help with some sample?
> >>>>>>
> >>>>>>
> >>>>>> ========== change group 1
> >>>>>> dochanges> cd /tmp/cvs2p4-3.0b6/p4root && /usr/bin/p4d -r . -jr
> >>>>>> dochanges> dbmeta
> >>>>>> Perforce db files in '.' will be created if missing...
> >>>>>> Recovering from dbmeta...
> >>>>>> dochanges: *** "cd /tmp/cvs2p4-3.0b6/p4root && /usr/bin/p4d -r .
> >>>>>> -jr dbmeta" exited with signal 0 status 255
> >>>>>> dochanges: "/usr/bin/p4d -r . -jr dbmeta" failed.
> >>>>>> *** dochanges  FAILED
> >>>>>>
> >>>>>>
> >>>>>> Best Regards,
> >>>>>> mcisse.
> >>>>>>
> >>>>>>
> >>>>> _______________________________________________
> >>>>> perforce-user mailing list  -  perforce-user at perforce.com
> >>>>> http://maillist.perforce.com/mailman/listinfo/perforce-user
> >>>>>
> >>>>>
> >>>>>
> >>> _______________________________________________
> >>> perforce-user mailing list  -  perforce-user at perforce.com
> >>> http://maillist.perforce.com/mailman/listinfo/perforce-user
> >>>
> >>> _______________________________________________
> >>> perforce-user mailing list  -  perforce-user at perforce.com
> >>> http://maillist.perforce.com/mailman/listinfo/perforce-user
> >>>
> >>>
> >>>
> > _______________________________________________
> > perforce-user mailing list  -  perforce-user at perforce.com
> > http://maillist.perforce.com/mailman/listinfo/perforce-user
> >
> >
> >
>
>
> --
> --------------------- [ SECURITY NOTICE ] ---------------------
> To: perforce-user at perforce.com.
> For your security, mcisse at acquisinc.com
> digitally signed this message on 16 November 2007 at 16:00:28 UTC.
> Verify this digital signature at http://www.ciphire.com/verify.
> ---------------- [ CIPHIRE DIGITAL SIGNATURE ] ----------------
> Q2lwaGlyZSBTaWcuAjhwZXJmb3JjZS11c2VyQHBlcmZvcmNlLmNvbQBtY2lzc2VA
> YWNxdWlzaW5jLmNvbQBlbWFpbCBib2R5AC4RAAB8AHwAAAABAAAAnL49Ry4RAACa
> AgACAAIAAgAgcT1B7ulrTbhgQFnPbu4QFcLOLXgm/ESnUgjBntK1lDcBAB96H8yw
> x1/YImyCYuRnwVi/WOzx5APTVBtika+ZsH9PN/Eivfd1ZO5LX8uLu7zkYbYeqMp9
> U6J8Dgl0o4ryEhzRU2lnRW5k
> ------------------ [ END DIGITAL SIGNATURE ] ------------------
> _______________________________________________
> perforce-user mailing list  -  perforce-user at perforce.com
> http://maillist.perforce.com/mailman/listinfo/perforce-user
>
_______________________________________________
perforce-user mailing list  -  perforce-user at perforce.com
http://maillist.perforce.com/mailman/listinfo/perforce-user

More information about the perforce-user mailing list