[PLUG] Debian auto logout?

Sat Sep 7 02:10:15 UTC 2002

On Fri, 6 Sep 2002, Richard Steffens wrote:
> I went to a console, logged in and checked uptime:
>
> 16:52:26 up 19:59, 1 user, load average: 0.00, 0.00, 0.00

I think Richard was suggesting to compare your xserver uptime
to your system uptime... the output above looks like system uptime, which
might not be useful by itself since it wouldn't be affected by restarting
X.

Unfortunately I don't know the canonical method for checking the uptime of
the X server... I suppose you could hunt around for X processes and check
their execution length but some processes would remain running and
some would be respawned during both a crash and a regular logout. Perhaps
the xserver or xfree86 manuals have info on where X keeps activity logs.

HTH,

Tyler

On Fri, 6 Sep 2002, Richard Steffens wrote:

> Richard Langis wrote:
>
> > You might leave the system on for a while, note the time it was turned
> > on and log in.  Leave it running, see if when you get back it's at the
> > xdm login prompt.  If so, drop to console mode and check the uptime.  If
> > it's LESS than what you'd calculate it to be from your poweron time,
> > then it's rebooting, not simply restarting X.
>
> While I was reading e-mail on this machine, I noticed a blink on the
> screen of the Debian machine. I watched as it changed a few times and
> finally settled at an X login. I went to a console, logged in and
> checked uptime:
>
> 16:52:26 up 19:59, 1 user, load average: 0.00, 0.00, 0.00
>
> The uptime sounds correct. I had the machine off to install the new USB
> board yesterday, plus I booted to Win98SE to install the HP 656c driver
> software. So, about 20 hours ago, I booted into Linux.
>
> > You might also check /var/log with 'ls -lt | head' to get the top 10
> > most recently modified files, and start tailing them to see what's going on.
>
> ls -lt | head shows: auth.log, syslog, messages, faillog, lastlog, wtmp,
> XFree86.0.log, daemon.log and the directory gdm. some of these files are
> not human readable. Here are some of the other ones:
>
> Here's tail syslog:
>
> Sep 6 15:53:01 Avatar /USR/SBIN/CRON[2629]: (mail) CMD (  if [ -x
> /usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
> Sep 6 16:08:01 Avatar /USR/SBIN/CRON[2682]: (mail) CMD (  if [ -x
> /usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
> Sep 6 16:23:01 Avatar /USR/SBIN/CRON[2697]: (mail) CMD (  if [ -x
> /usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
> Sep 6 16:26:28 Avatar gnome-name-server[2609]: input condition is: 0x10,
> exiting
> Sep 6 16:26:29 Avatar gdm[2535]: gdm_slave_xioerror_handler: Fatal X
> error - Restarting :0
> Sep 6 16:26:31 Avatar modprobe: modprobe: Can't locate module
> char-major-13
> Sep 6 16:38:01 Avatar /USR/SBIN/CRON[2727]: (mail) CMD (  if [ -x
> /usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
> Sep 6 16:46:33 Avatar gdm[2717]: gdm_slave_xioerror_handler: Fatal X
> error - Restarting :0
> Sep 6 16:46:35 Avatar modprobe: modprobe: Can't locate module
> char-major-13
> Sep 6 16:53:01 Avatar /USR/SBIN/CRON[2764]: (mail) CMD (  if [ -x
> /usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
>
> --------------------------------
>
> tail messages produces 10 lines that, with the exception of the time,
> all say:
>
> Sep 6 13:33:38 Avatar -- MARK --
> ... (every 20 minutes)
> Sep 6 17:13:38 Avatar -- MARK --
>
> But, the lines just before that are:
>
> Sep 6 15:00:38 Avatar kernel: hda: status timeout: status-0xd0 { Busy }
> Sep 6 15:00:38 Avatar kernel: ide0: reset: success
> Sep 6 15:13:13 Avatar -- MARK --
> Sep 6 15:24:19 Avatar kernel: hda: status timeout: status-0xd0 { Busy }
> Sep 6 15:24:19 Avatar kernel: ide0: reset: success
> Sep 6 13:13:02 Avatar kernel: hda: status timeout: status-0xd0 { Busy }
> Sep 6 13:13:02 Avatar kernel: ide0: reset: success
> Sep 6 13:33:38 Avatar -- MARK --
>
> The time seems a little odd, but I recall looking at the clock on the
> Gnome screen after lunch and seeing a time that was morning. At that
> time I ran ntpdate and the clock reset. I'm not sure what went wrong
> with the time. When I log in to Windows, I have a similar utility set
> the clock using utcnist.colorado.edu, or one of several others the
> utility went out and found, so the clock should have been close to
> correct. Anyway, that may account for some events occurring later in
> time, but earlier in the log. Although, those aren't the correct "wrong"
> times. Wierd.
>
> ---------------------------------
>
> daemon.log also has some of the same messages from gdm and
> gnome-name-server plus a couple of others:
>
> Sep 6 15:50:22 Avatar gnome-name-server[2611]: starting
> Sep 6 15:50:22 Avatar gnome-name-server[2611]: name server was running
> on display, exiting
> Sep 6 15:50:25 Avatar gnome-name-server[2620]: server_is_alive:
> cnx[IDL:GNOME/Panel2:1.0] = 0x8055160
> Sep 6 15:50:25 Avatar gnome-name-server[2623]: server_is_alive:
> cnx[IDL:GNOME/Panel2:1.0] = 0x8055538
> Sep 6 15:50:25 Avatar gnome-name-server[2624]: server_is_alive:
> cnx[IDL:GNOME/Panel2:1.0] = 0x8055748
> Sep 6 15:50:25 Avatar gnome-name-server[2609]: input condition is: 0x10,
> exiting
> Sep 6 16:26:28 Avatar gdm[2535]: gdm_slave_xioerr_handler: Fatal X error
> - Restarting :0
> Sep 6 16:26:31 Avatar modprobe: modprobe: Can't locate module
> char-major-13
> Sep 6 16:46:33 Avatar gdm[2717]: gdm_slave_xioerr_handler: Fatal X error
> - Restarting :0
> Sep 6 16:46:35 Avatar modprobe: modprobe: Can't locate module
> char-major-13
>
> So, it looks like X is restarting, and that's why I get the login
> screen.
> Looks like gnome, gdm and modprobe don't like something.
>
> --
> Regards,
>
> Dick Steffens
> "Quando Omni Flunkus Moritati"
> http://rsteff.home.attbi.com/
>
> _______________________________________________
> PLUG mailing list
> PLUG at lists.pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
>