Some time ago I discussed this issue with tobias
about opencvs. It seems that the servers put a limit
on the upload max data (128 MB I think) and cvs
only uploads data until it needs to download
any file from the server, so if the src tree it's bigger than
the server limit, it makes "weird" things.
-Jesus
Daniel Ouellet escribis:
Chris wrote:
I have had to interrupt (^c) cvs -d$CVSROOT checkout -P src command
about three times. I was wondering whether checking out src three
times would overwrite the old files or ignore what's already on the
disk and update files that are not there or do anything else?
cvs will never redo what's already there.
I know I am supposed to run cvs -d$CVSROOT checkout -P src once and
once I have a tree, I can update it at a later time with cvs
-d$CVSROOT up -Pd
Here, if you really want to be nice and save time to you and everyone
else that use the CVS, you could be much nicer and start by getting
the two files specially done for this as explain in the FAQ.
http://openbsd.org/anoncvs.html#starting
Get the src.tar.gz, and sys.tar.gz.
# cd /usr/src
# tar xzf ../sys.tar.gz
# tar xzf ../src.tar.gz
And because you did this, be careful to the note about it:
"NOTE: If you are updating a source tree that you initially fetched
from a different server, or from a CD, you must add the -d
anon...@anoncvs.ca.openbsd.org:/cvs options to cvs."
So, that you don't download it all over again.
This would be much faster for you and much nicer to the CVS server
itself instead of starting fresh with a plain checkout. Obviously
later we are in the process before the release, more changes are
present and will needs to be updated, but obviously not all files in
the tree change at every release, so you still save time and resources.
Just a small reminder along to way if you want to play nice. (;>
Best,
Daniel
