From: "Larry Jones" <[EMAIL PROTECTED]>
> The last part of the lock file name is the PID of the process that
> created the lock -- since there doesn't appear to be any current CVS
> process with those PIDs, I'd say that you have a problem with CVS
> crashing.
Hmm, I figured that would be the case, but I can't seem to figure out
what is the problem. But, just to put my mind at ease, this is not normal,
right? Can an error on the client side screw up the server? That is,
what happens if I kill (or crash) the client at random points? Will the
server
behave properly and clean up after itself?
> You're not running with PreservePermissions enabled, are you?
No. Here is my CVSROOT/config file:
# Set this to "no" if pserver shouldn't check system users/passwords
SystemAuth=no
# Set `PreservePermissions' to `yes' to save file status information
# in the repository.
# Do NOT set this to yes -- support for it is quite buggy
PreservePermissions=no
# Set `TopLevelAdmin' to `yes' to create a CVS directory at the top
# level of the new working directory when using the `cvs checkout'
# command.
#TopLevelAdmin=no
> That's the quickest way I know to crash CVS. If not, you're going to
> have to do some investigating to find out what's causing the crashes.
;(
> Have you had complaints from users?
Well, quite a few, as you can imagine -- we get these locks laying around
all the time, so a lot of people are running into them. Apart from this
problem, the cvs server dies on a few files with this error:
fileattr.c:52: fileattr_stardir: Assertion 'unrecog_head = ((void *) 0)'
failed
I've asked about this problem on the list some time ago (before I've left
for vacation),
but I can't seem to find any replies now.
> Do you have any core dumps lying around?
I'll look.
TIA,
Dimi.
>
> -Larry Jones
>
> Do you think God lets you plea bargain? -- Calvin
