Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to sales@march-hare.com.
> What would be the point of running that command though? We understood the 'cvs history' documentation that way, that we can use it to get a list of added, removed or changed files since a certain date relatively fast. SmartCVS just uses this information to launch an rlog command for the effected directories, because running an rlog command on the whole repository (or the whole project) often is extremely slow (expensive operation on the server side). > History *will* lose records on a busy server, so you can't rely on it. This is the first time I read that - is this a CVSNT "feature" or also valid for GNU CVS? Taking a look at the quasi-standard documentation Cederqvist (http://ximbiot.com/cvs/manual/cvs-1.11.20/cvs_16.html#SEC138) I can't find information, that one cannot rely on the output of the 'cvs history' command (I know, that one can run the server without keeping the history, but this is a different story). -- Best regards, Thomas Singer _____________ smartcvs.com smartsvn.com Tony Hoyle schrieb: > Thomas Singer wrote: > >> Hi Tony, >> >> The same crash happens when invoked without the -f: >> >>> C:\temp\c>cvs -d :pserver:tom at localhost:f:/repository history -a -x >>> AMR smartcvs-tori > > > What would be the point of running that command though? > > History is not an accurate way of getting log information... use log for > that. History *will* lose records on a busy server, so you can't rely > on it. > > Tony