Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Raid shuts down uncleanly?
0 views
Skip to first unread message
Land Haj
Mar 5, 2007, 4:30:14 PM3/5/07
to
Hi!
I'm still trying to get used to having raid-1 on my new etch installation. I have a very simple question:
When I reboot or turn off my computer, the system always has trouble umounting the raid devices. It says they are busy (in red, warning letters). But the shutdown completes anyway.
Then, when I boot into the system again, it says that the journal is being recovered.
Have I somehow misconfigured my system so that it shuts down uncleanly, and if so, how do I fix it?
The raid configuration was done at installation with the rc1 debian installer for etch.
Please help!
/landhaj
I'm still trying to get used to having raid-1 on my new etch installation. I have a very simple question:
When I reboot or turn off my computer, the system always has trouble umounting the raid devices. It says they are busy (in red, warning letters). But the shutdown completes anyway.
Then, when I boot into the system again, it says that the journal is being recovered.
Have I somehow misconfigured my system so that it shuts down uncleanly, and if so, how do I fix it?
The raid configuration was done at installation with the rc1 debian installer for etch.
Please help!
/landhaj
Never miss an email again!
Yahoo! Toolbar alerts you the instant new Mail arrives. Check it out.
Land Haj
Mar 6, 2007, 3:20:06 PM3/6/07
to
Thank you for your reply!
Today, I've dpkg-reconfigured mdadm. It had all md-devices listed as necessary for root, and so I changed it so that only md0 (where root is) is listed as necessary. This for some reason got rid of the message saying that swap (md1) was busy. But md0 and md2 are still not unmounted.
I've been trying to find the shutdown log, but I've failed. Is it supposed to be in wtmp? If so, how do I read it -- it's binary.
The end of the shutdown messages are displayed so fast and then blanked out so that I have no time to see them all. I finally managed to catch a few of them with my camera (!), and this is what is said, among other things:
* It can't create the /etc/mtab log file
* It can't create the /lib/init/rw/.mdadm directory
Both are because the file system is read-only, but I do not know if they are normal messages or related to my issue somehow.
The rest is only a repetition of "md0 still in use", etc.
If someone can instruct me in how to log the shutdown properly I'll post the log here to see if anyone can help.
Thanks for helping!
/landhaj
Today, I've dpkg-reconfigured mdadm. It had all md-devices listed as necessary for root, and so I changed it so that only md0 (where root is) is listed as necessary. This for some reason got rid of the message saying that swap (md1) was busy. But md0 and md2 are still not unmounted.
I've been trying to find the shutdown log, but I've failed. Is it supposed to be in wtmp? If so, how do I read it -- it's binary.
The end of the shutdown messages are displayed so fast and then blanked out so that I have no time to see them all. I finally managed to catch a few of them with my camera (!), and this is what is said, among other things:
* It can't create the /etc/mtab log file
* It can't create the /lib/init/rw/.mdadm directory
Both are because the file system is read-only, but I do not know if they are normal messages or related to my issue somehow.
The rest is only a repetition of "md0 still in use", etc.
If someone can instruct me in how to log the shutdown properly I'll post the log here to see if anyone can help.
Thanks for helping!
/landhaj
----- Original Message ----
From: Bob McGowan <bob_m...@symantec.com>
To: debia...@lists.debian.org
Sent: Monday, March 5, 2007 10:44:07 PM
Subject: Re: Raid shuts down uncleanly?
From: Bob McGowan <bob_m...@symantec.com>
To: debia...@lists.debian.org
Sent: Monday, March 5, 2007 10:44:07 PM
Subject: Re: Raid shuts down uncleanly?
Land Haj wrote:
> Hi!
>
> I'm still trying to get used to having raid-1 on my new etch
> installation. I have a very simple question:
>
> When I reboot or turn off my computer, the system always has trouble
> umounting the raid devices. It says they are busy (in red, warning
> letters). But the shutdown completes anyway.
>
> Then, when I boot into the system again, it says that the journal is
> being recovered.
>
> Have I somehow misconfigured my system so that it shuts down uncleanly,
> and if so, how do I fix it?
>
> The raid configuration was done at installation with the rc1 debian
> installer for etch.
>
> Please help!
>
> /landhaj
>
> Hi!
>
> I'm still trying to get used to having raid-1 on my new etch
> installation. I have a very simple question:
>
> When I reboot or turn off my computer, the system always has trouble
> umounting the raid devices. It says they are busy (in red, warning
> letters). But the shutdown completes anyway.
>
> Then, when I boot into the system again, it says that the journal is
> being recovered.
>
> Have I somehow misconfigured my system so that it shuts down uncleanly,
> and if so, how do I fix it?
>
> The raid configuration was done at installation with the rc1 debian
> installer for etch.
>
> Please help!
>
> /landhaj
>
> ------------------------------------------------------------------------
> Never miss an email again!
> Yahoo! Toolbar
> alerts you the instant new Mail arrives. Check it out. <
> http://us.rd.yahoo.com/evt=49937/*http://tools.search.yahoo.com/toolbar/features/mail/>
It's normal for the root device to not unmount during shutdown, but
there should also be a message that it was re-mounted read-only.
My system, with a raid-1 for the boot device and similar messages on
shutdown, doesn't have problems with the reboot.
The root device is considered 'in use' because the system and system
shutdown scripts run from it (in other words, there are processes with
open files). In order to complete shutdown cleanly, the mount is
changed to read-only, which sets whatever bit(s) are needed to tell the
boot sequence scripts that the filesystem is clean.
So, for some reason, you're not getting a clean remount. More
information from the shutdown sequence is needed before I could
speculate any further.
Bob
It's normal for the root device to not unmount during shutdown, but
there should also be a message that it was re-mounted read-only.
My system, with a raid-1 for the boot device and similar messages on
shutdown, doesn't have problems with the reboot.
The root device is considered 'in use' because the system and system
shutdown scripts run from it (in other words, there are processes with
open files). In order to complete shutdown cleanly, the mount is
changed to read-only, which sets whatever bit(s) are needed to tell the
boot sequence scripts that the filesystem is clean.
So, for some reason, you're not getting a clean remount. More
information from the shutdown sequence is needed before I could
speculate any further.
Bob
Expecting? Get great news right away with email Auto-Check.
Try the Yahoo! Mail Beta.
Land Haj
Mar 6, 2007, 3:40:15 PM3/6/07
to
I did this:
cat messages | grep shutdown
but it reveals nothing useful in regards to the problem, nor does the rest of the file. The only messages listed at shutdown time are:
Mar 6 20:15:13 greengoblin shutdown[3244]: shutting down for system reboot
Mar 6 20:15:21 greengoblin kernel: Kernel logging (proc) stopped.
Mar 6 20:15:21 greengoblin kernel: Kernel log daemon terminating.
Mar 6 20:15:22 greengoblin exiting on signal 15
Can the shutdown not be logged in some other, more detailed way?
Gratefully,
landhaj
cat messages | grep shutdown
but it reveals nothing useful in regards to the problem, nor does the rest of the file. The only messages listed at shutdown time are:
Mar 6 20:15:13 greengoblin shutdown[3244]: shutting down for system reboot
Mar 6 20:15:21 greengoblin kernel: Kernel logging (proc) stopped.
Mar 6 20:15:21 greengoblin kernel: Kernel log daemon terminating.
Mar 6 20:15:22 greengoblin exiting on signal 15
Can the shutdown not be logged in some other, more detailed way?
Gratefully,
landhaj
<--snipped-->
Check the file /var/log/messages, searching for the word 'shutdown' and
see if there's anything there that's helpful.
Otherwise, I have no other suggestions for you.
Bob
Check the file /var/log/messages, searching for the word 'shutdown' and
see if there's anything there that's helpful.
Otherwise, I have no other suggestions for you.
Bob
Get your own web address.
Have a HUGE year through Yahoo! Small Business.
Land Haj
Mar 6, 2007, 4:20:12 PM3/6/07
to
Ok, I did as you instructed using emacs, but the log messages near the shutdown messages still seem totally unrelated, as far as I can tell. There are a few that are preceded by gconf thingies, which I gather are from times when I've shutdown from inside GNOME, and one or two that have no comments near (timewise) the shutdown. The comments about gconf are in swedish, so I'm not listing them -- the same problem occurs in shutdowns where the gconf comments are not present anyway.
/var/log/messages seems to be a dead end unless I can somehow crank up the detailness (?) so that more messages are logged.
/landhaj
/var/log/messages seems to be a dead end unless I can somehow crank up the detailness (?) so that more messages are logged.
/landhaj
That's because the file has multiple lines, not all of which have the
word shutdown in them, but which may be related. So, you need to use an
editor or pager (vi, gedit, less, more, ...), search for 'shutdown' and
then look at the lines following and/or just before it, for possible clues.
Bob
word shutdown in them, but which may be related. So, you need to use an
editor or pager (vi, gedit, less, more, ...), search for 'shutdown' and
then look at the lines following and/or just before it, for possible clues.
Bob
Finding fabulous fares is fun.
Let Yahoo! FareChase search your favorite travel sites to find flight and hotel bargains.
martin f krafft
Mar 7, 2007, 12:50:17 PM3/7/07
to
also sprach Land Haj <lan...@yahoo.com> [2007.03.06.2054 +0100]:
> If someone can instruct me in how to log the shutdown properly
> I'll post the log here to see if anyone can help.
> If someone can instruct me in how to log the shutdown properly
> I'll post the log here to see if anyone can help.
http://svn.debian.org/wsvn/pkg-mdadm/mdadm/trunk/debian/FAQ?op=file&rev=0&sc=0
Question 8
--
.''`. martin f. krafft <mad...@debian.org>
: :' : proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck - http://debiansystem.info
`- Debian - when you have better things to do than fixing systems
"no woman should ever be quite accurate about her age.
it looks so calculating."
-- oscar wilde
Land Haj
Mar 8, 2007, 4:20:14 AM3/8/07
to
Thank you for this link!
Although I can't seem to find what is keeping the /home array busy, I take it I shouldn't worry too much about this. Do you know if the kernel only stops the root array "just before halting", like your link says, or if it also stops the /home array properly? If the former is the case, I could maybe dpkg-reconfigure mdadm to consider both /root and /home arrays to be necessary for root.
Gratefully,
landhaj
Although I can't seem to find what is keeping the /home array busy, I take it I shouldn't worry too much about this. Do you know if the kernel only stops the root array "just before halting", like your link says, or if it also stops the /home array properly? If the former is the case, I could maybe dpkg-reconfigure mdadm to consider both /root and /home arrays to be necessary for root.
Gratefully,
landhaj
----- Original Message ----
From: martin f krafft <mad...@debian.org>
To: debian users <debia...@lists.debian.org>
Sent: Wednesday, March 7, 2007 6:39:22 PM
Subject: Re: Raid shuts down uncleanly?
From: martin f krafft <mad...@debian.org>
To: debian users <debia...@lists.debian.org>
Sent: Wednesday, March 7, 2007 6:39:22 PM
Subject: Re: Raid shuts down uncleanly?
We won't tell. Get more on shows you hate to love
(and love to hate): Yahoo! TV's Guilty Pleasures list.
martin f krafft
Mar 8, 2007, 6:20:18 AM3/8/07
to
also sprach Land Haj <lan...@yahoo.com> [2007.03.08.1018 +0100]:
> Although I can't seem to find what is keeping the /home array
> busy, I take it I shouldn't worry too much about this. Do you know
> if the kernel only stops the root array "just before halting",
> like your link says, or if it also stops the /home array properly?
> If the former is the case, I could maybe dpkg-reconfigure mdadm to
> consider both /root and /home arrays to be necessary for root.
> Although I can't seem to find what is keeping the /home array
> busy, I take it I shouldn't worry too much about this. Do you know
> if the kernel only stops the root array "just before halting",
> like your link says, or if it also stops the /home array properly?
> If the former is the case, I could maybe dpkg-reconfigure mdadm to
> consider both /root and /home arrays to be necessary for root.
It will stop all arrays, if it can.
--
Please do not send copies of list mail to me; I read the list!
.''`. martin f. krafft <mad...@debian.org>
: :' : proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck - http://debiansystem.info
`- Debian - when you have better things to do than fixing systems
NP: Disturbed / The Sickness
0 new messages