(Illustration by Gaich Muramatsu)
According to Jan Harkes: > >On Wed, May 16, 2001 at 03:10:39PM -0500, Ryan M. Lefever wrote: >> I'm using glibc-2.1.3-15, and gcc version egcs-2.91.66 19990314/Linux >> (egcs-1.1.2 release). These are the ones that come on Redhat 6.2 > >Puzzled look on my face. > >That's pretty much what is used here at CMU as well. >I can't see what is going wrong here. > Hmmm perhaps an unrelated observation may help here then :-) I happened to be debugging a malloc'ing problem in some of my own code on a NetBSD machine so I had set MALLOC_OPTIONS to tell malloc to always allocate a new memory area on realloc and fill the unused portion (and the old area) with the value '0xD0'. Just by chance I attempted to use clog with this environment and I get: [blymn_at_rover] clog username: blymn Password: Invalid login (RPC2_NOBINDING (F)). I _know_ my password was typed correctly and clog does work from another window that does not have the MALLOC_OPTIONS set so all is right with my server/venus connection. To me, this sounds like someone is doing something naughty with a realloc'ed bit of memory which could lead to a lot of weirdness. This is with coda 5.3.5 so these problems may have been fixed already... -- =============================================================================== Brett Lymn, Computer Systems Administrator, BAE SYSTEMS ===============================================================================Received on 2001-05-17 08:47:48