This is the mail archive of the cygwin@sources.redhat.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

sshd and stack dump problem


I've updated both client and server systems to
current cygwin.  If my host system didn't
have a /home/user_name directory I ended up
with a stack trace during every attempted
execution (i.e. initial bash setup, ls, 
whatever).  This correlated with a memory
access error dialog box popping up on the host.
If you clear the memory error dialog box
I would end up on the host via ssh, but every
attempted command execution would throw the
stack trace-memory access error. 

The stack error signature ... STATUS_ACCESS_VIOLATION

Once I added /home/user_name directory that
problem went away.  I can now ssh to the
host Win2K box running sshd as a service
(system userid and password auth).  HOWEVER

 * ls doesn't work, hangs, though set, echo
   cd and pwd all work; I didn't try others

 * when I kill the service (sc stop service_name)
   the service doesn't die.  I'm guessing SRVANY
   is stopped but that sshd has spawned itself
   out of SRVANY's control and lives on.

Anyone have suggestions/corrections on these two
things (ls hanging, sshd.exe never dies thereby
blocking cygwin update w/o rebooting box)?  We're
trying to tighten up an NT box that's living in
a DMZ and most other services (RCMD) are being
axed on it, so having a SSH mechanism to get into
it would be nice.

TIA,
Heitzso



--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]