I saw this mentioned in the archives on May 4, 2001 in a question posted by Chris Sharpe. He describes very similar exhibition of the symptom, and goes on to speculate association with a FIXME in server.c for 1.11p1. I didn't see any additional mention beyond that in the archives. Here are the particulars:: $cvs co xxxxxuniverse <Enter> .... .... Protocol error: uncounted data discarded. Configuration: Client: NT 4.0 SP5 cvs 1.10.8 running inside bash, version 2.04.0(1)-release-(i586-pc-cygwin) Server: Linux 2.2.14-5.0 cvs 1.10.7 pserver cvspasswd usage (i.e. no user accounts) Repository access is via a local network segment using DHCP. The repo box has an externally visible IP while the clients do not. The repo box can be pinged and ssh'ed. Small (100 or so files) modules do not display the symptom. We're using cvswrappers for substitution mode and a multi-module modules file. The above error occurs only when checking out a 'universe' describing module and is reproducible. FYI - this error did not present itself yesterday. No [known] CVS or repository changes have been made. Is this a pserver error message? Has the 'universe' module simply grown to some statically "too large" value? _____________________________________ John Minnihan mailto:[EMAIL PROTECTED] http://www.freepository.com _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
