I've seen several posts on the newsgroups and even one or 2 in the archives for this list which mention this problem, but I have yet to hear, from any source, an real solution for this. The problem is that it is not possible to dump core from a remote login session (from a telnet or ssh session, for example). Many people report the symptom of this (the error "limit: cannot set coredumpsize" from tcsh). But, this is just proof of the problem. The solution of "comment out the line in the .cshrc file" (the only "solution" anyone ever posts) doesn't fix anything, it simply covers up the problem by making the error messages go away. This is not a tcsh issue. Go into bash, type 'ulimit -c 10000' it will fail with the same message tcsh gives you. I do a lot of remote development. I require the ability to dump core from a remotely run program. Does anyone know why this cannot be set?? Even root cannot dump core remotely! Unlike regular remote users, root appears to be able to set it's coredumpsize. When you actually dump a core file, there appears to be one, but then you notice that it is a 0 length file. Attempt this same thing from a local shell (as a super or non-super user) and you get a big juicy 8meg coredump, all ready for gdb. I was thinknig this was a pamd issue. But, the default /etc/security/limits.conf file is completely commented out. So the line that limits remote coredumps to 0 shouldn't be processed. The limits module also doesn't appear to be used by any of the files in the /etc/pam.d/ directory. I've been trying to figure out what's going on with this for over a week with no responses to any of my emails and newsgroup postinsg, and no luck with any of my configuration tweaking and FAQ and manual readings. Does anyone know anything about this problem? Can pamd be removed from a mandrake system, if it is to blame and is unfixable? Thanks, - Anil.
