If I commit only one or two files, I have no problem. However, the latency is exacerbated if I commit a lot of files. By a lot I mean say at least 15, as when I merge a branch onto a trunk.
-chris >-----Original Message----- >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] >] On Behalf Of Fouts Christopher (IFNA MP DC) >Sent: Friday, August 13, 2004 9:59 AM >To: [EMAIL PROTECTED]; [EMAIL PROTECTED] >Subject: RE: CVS slooowwww over SSH??? > > >No binary files at all, in fact. I have about 50 or so ASCII >files, but yes with lots of history data. I'm trying to see if >this is purely SSH-related (not SSH itself but maybe the way >we've installed it) or CVS/SSH related. I'll try the -z option. Thanks! > >-chris > >>-----Original Message----- >>From: [EMAIL PROTECTED] >>[mailto:[EMAIL PROTECTED] >>] On Behalf Of Izo >>Sent: Friday, August 13, 2004 4:02 AM >>To: [EMAIL PROTECTED] >>Subject: Re: CVS slooowwww over SSH??? >> >> >>Do you have large files with long history (lot of versions on >>repository) ? Are they binary or ASCII ? Namely, CVS first >>does the diff >>of the whole bunch of files for each one and during the >>process you see >>nothing - yet the full file transfer is being performed at >>that time. It >>can last a little bit longer if you have large binary files in >>comparission to the small ASCII files. Also, try use the -z option >>(suggested compression value is 3, thus -z3). As for my experience, >>there is no need to use ssh compression if you use -z . >> >>Izo >> >>_______________________________________________ >>Info-cvs mailing list >>[EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/info-cvs >> > > >_______________________________________________ >Info-cvs mailing list >[EMAIL PROTECTED] >http://lists.gnu.org/mailman/listinfo/info-cvs > _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/info-cvs
