-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
The Thursday 2007-09-27 at 09:27 +0200, Lukas Ocilka wrote:
> Carlos E. R. wrote:
> > Hi,
> >
> > I told Yast to update all updatables. It downloaded over a gigabit, it took
> > the
> > whole afternoon. It is a nuisance when it screeches over a network failure
> > and
> > doesn't retry till I come by the computer and tell it to retry.
>
> Good point, YaST should either try to download automatically (after 60
> seconds or so) or should give the opportunity to
> [ Retry Automatically ].
>
> Anyway, this would need to be configurable:
> * Try automatically
> * Max tries (for not to end up in a loop)
> * Time wait
Another thing is that as it downloads one file, installs it, downloads
another, installs it... if it stops midway the system remains in an
undefined state: for instance, not RC1, not RC2. There might be
inconsistencies, too. It would be better to download everything, then
install everything.
>
> Please, file an Enhancement request.
Ok, will do.
[...]
Huh... er... how do I do that? I can't find a bugzilla option or button
for that. Enter new bug, perhaps? Then what? Choose "enhancement" in the
"severity" drop list?
Sorry, I have never done that, and the FAQ doesn't explain it.
> > In the end, I told it to reboot, or perhaps halt; in any case, it halted. I
> > booted again, and I saw the system running a long filesystem check because
> > the
> > root partition had not been cleanly umounted.
> >
> > I thought this was a known bug with the RC1 DVD, but not with the installed
> > system.
>
> Yes, it was a bug in RC1 when rebooting from First Stage to Second Stage
> of the installation. Sometimes, there were two processes running under
> the /mnt path (/mnt == Installation chroot). These processes were:
> 'ntpd' and 'dhcpcd'.
I remember that. But this was not the install system, but the already
installed system.
>
> How exactly did you rebooted the system?
- From inside kde user session, choose halt or reboot. My intention was to
select reboot, but maybe I miss-clicked and choose halt. In any case, the
system halted. Then I powered up again, and saw the fsck messages.
Notice that after a large YOU update, like from RC1 to RC2, many libraries
and programs have a different version on disk and in memory. Many things
could fail.
> I'm not sure, but I have a feeling that in a minimal installation I saw
> something like:
> reboot: fuser: command not found
>
> And 'fuser' is a command that can list or even kill processes running
> depending on the directory that they have open. Please, write mote
> details, you can additionally inspect the end of YaST logs:
> /var/log/YaST2/y2log
But Yast was not running when I rebooted. The YOU session finished without
saying anything, but I knew a reboot was in order, so I did. I don't
think there will be anything in the Yast log, nor in any other log, as it
will have been a problem with the rc or halt scripts: as the system is
not cleanly umounted and flushed, the last entries in the logs, if any,
will be in the syslog memory or kernel buffers and lost.
What I see in the /var/log/messages is this:
Sep 27 01:11:09 minas-morgul smartd[24826]: Device: /dev/hdd, SMART Usage
Attribute: 195 Hardware_ECC_Recovered changed from 59 to 60
Sep 27 01:27:56 minas-morgul sudo: cer : pam_authenticate: Conversation
error ; TTY=pts/8 ; PWD=/home/cer ; USER=root ;
COMMAND=/opt/kde3/bin/kdesu_stub -
Sep 27 01:27:58 minas-morgul gconfd (cer-4475): Received signal 15, shutting
down cleanly
Sep 27 01:27:58 minas-morgul gconfd (cer-4475): Exiting
Sep 27 01:28:07 minas-morgul console-kit-daemon[2388]: GLib-CRITICAL:
g_queue_push_head: assertion `queue != NULL' failed
Sep 27 01:28:08 minas-morgul shutdown[4628]: shutting down for system halt
Sep 27 01:28:09 minas-morgul init: Switching to runlevel: 0
Sep 27 01:28:11 minas-morgul ntpd[22031]: ntpd exiting on signal 15
Sep 27 01:28:11 minas-morgul smartd[24826]: smartd received signal 15:
Terminated
Sep 27 01:28:12 minas-morgul smartd[24826]: smartd is exiting (exit status 0)
Sep 27 01:28:12 minas-morgul auditd[17825]: The audit daemon is exiting.
Sep 27 01:28:12 minas-morgul audispd[17830]: input read: EOF
Sep 27 01:28:13 minas-morgul sshd[21181]: Received signal 15; terminating.
Sep 27 01:28:13 minas-morgul syslog-ng[25437]: syslog-ng version 1.6.12 going
down
Sep 27 01:30:55 minas-morgul syslog-ng[3132]: syslog-ng version 1.6.12 starting
Nothing there. The kernel log (verbosity=7) contains:
Sep 27 01:28:11 minas-morgul kernel: bootsplash: status on console 0 changed to
on
Sep 27 01:28:12 minas-morgul kernel: audit(1190849291.979:471): audit_pid=0
old=17825 by auid=1000
Sep 27 01:28:12 minas-morgul kernel: audit(1190849291.979:472): auid=1000 uid=0
gid=0 pid=17830 comm="audispd" sig=6
Sep 27 01:28:13 minas-morgul kernel: pnp: Device 00:0e disabled.
Sep 27 01:28:13 minas-morgul kernel: ACPI: PCI interrupt for device
0000:00:1f.5 disabled
Sep 27 01:28:13 minas-morgul kernel: Kernel logging (proc) stopped.
Sep 27 01:28:13 minas-morgul kernel: Kernel log daemon terminating.
Sep 27 01:31:00 minas-morgul kernel: klogd 1.4.1, log source = /proc/kmsg
started.
The only way to see something would be a "halt" log, but it could not be
saved. Perhaps if the halt process stops at the last second so that we can
have a peep at the screen... :-?
- --
Cheers,
Carlos E. R.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Made with pgp4pine 1.76
iD8DBQFG/PAOtTMYHG2NR9URAgiGAJ4k8bnLMspHoREGlOA198nvdyZG+gCbBXWs
COfljd65Y4rhiriaR0275/A=
=7LqD
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
