All of lore.kernel.org
 help / color / mirror / Atom feed
* unmount immediately followed by remount
@ 2007-06-08 19:29 Jeff Weber
  2007-06-08 20:10 ` Jeff Moyer
  0 siblings, 1 reply; 2+ messages in thread
From: Jeff Weber @ 2007-06-08 19:29 UTC (permalink / raw)
  To: autofs

I have 2 WinXP servers, I am automounting using smbs.  For one server, the 
automounter unmounts the smbfs after inactivity as expected.  For the other 
server, supermad2, the automounter expires and unmounts the 4 shares after 
approx the 60sec timeout, as expected.  However, the shares are then 
immediately remounted.  Thus, the shares from the server are effectively 
continuously mounted.  Here's some sample automount debug output:

Jun  8 14:09:36 quetzal automount[31118]: handle_packet: type = 2
Jun  8 14:09:36 quetzal automount[31118]: handle_packet_expire_multi: token 
297,
 name supermad2
Jun  8 14:09:36 quetzal automount[31651]: expiring path /smb/supermad2
Jun  8 14:09:36 quetzal automount[31651]: umount_multi: path=/smb/supermad2 
incl
=1
Jun  8 14:09:36 quetzal automount[31651]: umount_multi: unmounting 
dir=/smb/supermad2/TRANSFER
Jun  8 14:09:36 quetzal automount[31651]: umount_multi: unmounting 
dir=/smb/supermad2/MDUsers
Jun  8 14:09:36 quetzal automount[31651]: umount_multi: unmounting 
dir=/smb/supermad2/CAD
Jun  8 14:09:36 quetzal automount[31651]: umount_multi: unmounting 
dir=/smb/supermad2/ENG
Jun  8 14:09:36 quetzal automount[31651]: rm_unwanted: /smb/supermad2/TRANSFER
Jun  8 14:09:36 quetzal automount[31651]: rm_unwanted: /smb/supermad2/MDUsers
Jun  8 14:09:36 quetzal automount[31651]: rm_unwanted: /smb/supermad2/ENG
Jun  8 14:09:36 quetzal automount[31651]: rm_unwanted: /smb/supermad2/CAD
Jun  8 14:09:36 quetzal automount[31651]: expired /smb/supermad2
Jun  8 14:09:36 quetzal automount[31118]: handle_child: got pid 31651, sig 0 
(0)
, stat 0
Jun  8 14:09:36 quetzal automount[31118]: sig_child: found pending iop pid 
31651
: signalled 0 (sig 0), exit status 0
Jun  8 14:09:36 quetzal automount[31118]: send_ready: token=297
Jun  8 14:09:36 quetzal automount[31118]: handle_packet: type = 0
Jun  8 14:09:36 quetzal automount[31118]: handle_packet_missing: token 298, 
name
 supermad2
Jun  8 14:09:36 quetzal automount[31118]: attempting to mount 
entry /smb/supermad2

lsof and fuser show the mounted filesystems are not busy.  I would think if 
the filesystems were busy, they would not be unmounted.  I can't imagine why 
the automounter would cycle with mount, unmount, mount, ...
with effectively zero unmounted time.

my config:
SuSE 10.1
Linux kernel 2.6.16.27-0.9-default
Linux automount version 4.1.4

	thanks,
	Jeff
-- 
Jeff Weber
American Superconductor Corp.

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: unmount immediately followed by remount
  2007-06-08 19:29 unmount immediately followed by remount Jeff Weber
@ 2007-06-08 20:10 ` Jeff Moyer
  0 siblings, 0 replies; 2+ messages in thread
From: Jeff Moyer @ 2007-06-08 20:10 UTC (permalink / raw)
  To: Jeff Weber; +Cc: autofs

==> On Fri, 8 Jun 2007 14:29:02 -0500, Jeff Weber <jweber@amsuper.com> said:

Jeff> I have 2 WinXP servers, I am automounting using smbs.  For one
Jeff> server, the automounter unmounts the smbfs after inactivity as
Jeff> expected.  For the other server, supermad2, the automounter
Jeff> expires and unmounts the 4 shares after approx the 60sec
Jeff> timeout, as expected.  However, the shares are then immediately
Jeff> remounted.  Thus, the shares from the server are effectively
Jeff> continuously mounted.  Here's some sample automount debug
Jeff> output:

This used to happen with older versions of hal:

     https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=192923

-Jeff

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2007-06-08 20:10 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-06-08 19:29 unmount immediately followed by remount Jeff Weber
2007-06-08 20:10 ` Jeff Moyer

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.