CVS code import failing

This looks like something that should be added to https://help.launchpad.net/VcsImportRequests known issues.
rpm5.org is not loading for me right now, I think this site is dead.

Not dead:

    $ ping rpm5.org
    PING rpm5.org (194.97.152.138) 56(84) bytes of data.
    64 bytes from rpm5.org (194.97.152.138): icmp_req=13 ttl=47 time=211 ms
    64 bytes from rpm5.org (194.97.152.138): icmp_req=15 ttl=47 time=205 ms

And cvs checkouts are being done every 2h under buildbots here
    http://harwich.jbj.org:8010
The buildbot's indicate a failure this morning between 6am and 8am.

So a statement of "dead" is perhaps premature. ;-)

Back to the original topic (while diagnosing the cvs server outage):

I can hand you a full backup of the @rpm5.org cvs repository if
that helps jump start the bzr import.

The repository is perhaps 300Mb (from memory) as a compressed
traball.

Would having a complete back-up of the repository help debug the import?

Hi Jeff,

On Wed, Feb 15, 2012 at 03:33:37PM -0000, Jeff Johnson wrote:
> Back to the original topic (while diagnosing the cvs server outage):
>
> I can hand you a full backup of the @rpm5.org cvs repository if
> that helps jump start the bzr import.
>
> The repository is perhaps 300Mb (from memory) as a compressed
> traball.
>
> Would having a complete back-up of the repository help debug the import?
There seem to be two issues here. One is a fairly simple regression
(as can be seen in the import from the 9th of February), the other
is deep down in cscvs (our import tool) somewhere.

cscvs should just work with a remote CVS repository. I don't think
an offline copy would help with debugging (though I haven't actually
looked at what might be going wrong). Thanks for the offer
though.

Cheers,

Jelmer

OK.

If there is anything I can do to assist, please feel free to ask. Launchpad rocks!

BTW, rpm5.org is now back on line and cvs is functional (at least from here):

[jbj@wareham wdj54]$ cvs up
? o
? other.bad
? t
? tmongo
? tmqa
? xx
? perl/MYMETA.yml
? rpmdb/hdrfmt.c-BUGGY
? rpmdb/hdrfmt.c-ICONV
? rpmdb/hdrfmt.c-V0
? rpmdb/hdrfmt.c-V1
? rpmdb/hdrfmt.c-V2
? tests/CM15
? tests/IDMS
? tests/IDMS-SAVE
? tests/MAG2011
? tests/MDV2011
? tests/gen
? tests/t
? tests/tsmart
M devtool.conf
M rpmdb/db3.c
M rpmdb/hdrfmt.c
M rpmio/librpmio.vers
M scripts/mgo
M scripts/wnh_filelists_mongo
M scripts/wnh_other_mongo
M scripts/wnh_primary_mongo
M tests/Makefile.am
M tests/ref/showrc
M tests/ref/version

This isn't a code regression - this repository now happens to trigger a cscvs import bug.