tom@ismdqa.intel.com (Tom Soukup ~) (10/13/89)
Hi all,
I have a pretty serious elm related problem that has caused several users
to lose mail. What I believe is happening is that when elm tries to create
the temporary mailbox in /tmp and the filesystem overflows, elm continues
on assuming that the file was created. The user sees a message that the
filesystem is full and their mail is gone. Then when the user exits elm,
whatever is in the /tmp/mbox.username file (usually nothing) gets copied back to
/usr/spool/mail/username and everything is gone forever.
My question is this. Is this possibly what's happening or is there
something else that I need to look for? If this is an elm bug, is it known
and getting fixed? For know I'll increase the disk space available for
/tmp, but I have limits on how much I can use and I have users who just
can't (won't) keep their mail files down to a reasonable size.
Any info/ideas would be greatly apreciated.
Thanks,
Tom Soukup
________________________________________________________________________________
DISCLAMER: Intel doesn't agree with much of anything that I say.
UUCP: {amdcad,decwrl,hplabs,oliveb,pur-ee,qantel}!intelca!mipos3!ismdqa!tom
ARPA: tom%ismdqa.intel.com@relay.cs.net CSNET: tom@ismdqa.intel.comsyd@DSI.COM (Syd Weinstein) (10/14/89)
tom@ismdqa.intel.com (Tom Soukup ~) writes: >I have a pretty serious elm related problem that has caused several users >to lose mail. What I believe is happening is that when elm tries to create >the temporary mailbox in /tmp and the filesystem overflows, elm continues >on assuming that the file was created. ... >My question is this. Is this possibly what's happening or is there >something else that I need to look for? If this is an elm bug, is it known >and getting fixed? It's a known problem, and no, its not getting fixed, as it is a very generic problem, and I will hold out for a complete solution and not a partial one. The problem is EB09 from the known bug list. Those wishing to solve the problem, and not just patch part of it, are welcome to do so. As to why I will not accept patch work solutions, because if the importance of this one gets reduced, it will never get solved at all. As long as people have problems with it, it stands a chance of getting fixed. -- ===================================================================== Sydney S. Weinstein, CDP, CCP Elm Coordinator Datacomp Systems, Inc. Voice: (215) 947-9900 syd@DSI.COM or {bpa,vu-vlsi}!dsinc!syd FAX: (215) 938-0235