Bug#543375: initscripts: last superblock write time in future
Tomasz Grzelak pisze: 2009/10/16 Mader, Alexander auma...@gmx.net mailto:auma...@gmx.net Tomasz Grzelak schrieb: what is the status of the case - am I the only one experiencing the problem still? Is there anyone affected by this even with the new version of util-linux 2.16.1-4? As mentioned in an earlier e-mail everything works fine since I edited the hwclock-udev-rule as decribed in another e-mail in this bug report. Hello, I am sorry, but I'm still affected by this problem. I have added the lines to the file, so now it contains: m...@daemon:~$ cat /lib/udev/rules.d/85-hwclock.rules # Reset the System Clock to UTC if the hardware clock from which it was # copied by the kernel was in localtime. KERNEL==rtc0, RUN+=/lib/udev/hwclock-set $root/$name KERNEL==rtc0, RUN+=/bin/touch /dev/rtc0-appeared KERNEL==rtc0, RUN+=logger.agent but it does not change anything for me. Immediatelly after I had removed [options] buggy_init_scripts = 1 from e2fsck.conf file, and rebooted the machine, I got the error about last superblock write time in future (exactly the same as on the screenshots I attached previously). In my opinion the problem has not been solved. How about others? B. Regards, Tomasz Grzelak P.S. Before having made the test I updated all packages, so my system was fresh. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future
Hi, what is the status of the case - am I the only one experiencing the problem still? Is there anyone affected by this even with the new version of util-linux 2.16.1-4? B. Regards, Tomasz Grzelak
Bug#543375: initscripts: last superblock write time in future
Tomasz Grzelak schrieb: what is the status of the case - am I the only one experiencing the problem still? Is there anyone affected by this even with the new version of util-linux 2.16.1-4? As mentioned in an earlier e-mail everything works fine since I edited the hwclock-udev-rule as decribed in another e-mail in this bug report. Best regards, -- Alexander Mader -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future
2009/10/16 Mader, Alexander auma...@gmx.net Tomasz Grzelak schrieb: what is the status of the case - am I the only one experiencing the problem still? Is there anyone affected by this even with the new version of util-linux 2.16.1-4? As mentioned in an earlier e-mail everything works fine since I edited the hwclock-udev-rule as decribed in another e-mail in this bug report. thank you, I'll check that during the weekend. Somehow I thought the change would have been done by a regular update, and not editing files manually. I'll let you know if that works for me. B. Regards, Tomasz Grzelak
Bug#543375: initscripts: last superblock write time in future, occurring sometimes
Hello, I'm still having the same issue. I made an update, and rebooted the machine which casued the error appeared again. I have a dual boot machine, but I have not run vista today yet. And there's a 'screenshot' what happend after the restart right after having updated all the packages: https://hellfire666.no-ip.org/~myst/deb_error_11.10.2009.jpg And my version of util-linux is 2.16.1-4 as depicted below. You had mentioned the bug should have been solved in this version of util-linux, but for me it does not work. Or maybe it's related to some other package/settings? Please help! :) B. Regards, Tomasz Grzelak --- m...@daemon:~$ dpkg -p util-linux Package: util-linux Essential: yes Priority: required Section: utils Installed-Size: 1968 Maintainer: LaMont Jones lam...@debian.org Architecture: amd64 Version: 2.16.1-4 Replaces: e2fsprogs, fdisk, linux32, miscutils, schedutils, setterm, sparc-utils Provides: linux32, schedutils Depends: lsb-base (= 3.0-6), tzdata (= 2006c-2), initscripts, dpkg (= 1.15.4) | install-info Pre-Depends: libblkid1 (= 2.16), libc6 (= 2.4), libncurses5 (= 5.6+20071006-3), libselinux1 (= 1.32), libslang2 (= 2.0.7-1), libuuid1 (= 2.16), zlib1g (= 1:1.1.4) Suggests: util-linux-locales, kbd | console-tools, dosfstools Conflicts: console-tools ( 1:0.2.3-21), fdisk, kbd ( 1.05-3), linux32, schedutils, setterm Size: 623322 Description: Miscellaneous system utilities This package contains a number of important utilities, most of which are oriented towards maintenance of your system. Some of the more important utilities included in this package allow you to partition your hard disk, view kernel messages, and create new filesystems. --- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future, occurring sometimes
Hello, like I was afraid, having used older versions of e2fslibs and progs did not change the situation, today (Sat 10.10.2009) I got the error again. Next I updated all packages, and while rebooting I got the same error again :( Please have a look at the picture I took with my camera: https://hellfire666.no-ip.org/~myst/deb_error.jpg This bug has not been fixed. How about others? Are you still affected by this problem? B. Regards, -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Let's try some simple and stupid debugging. Somebody who can reliably reproduce the problem please add something like this to the *end* of /lib/udev/rules.d/85-hwclock.rules : KERNEL==rtc0, RUN+=/bin/touch /dev/rtc0-appeared KERNEL==rtc0, RUN+=logger.agent Hello, I did try it and it was successful :-) No falsely motivated fsck any more. Something additional: I am running a Debian kernel not a self built one, and /lib/udev/rules.d/85-hwclock.rules was not there in the first place so I had to create it. Best regards and many thanks, -- Alexander Mader alexander.ma...@niles.de smime.p7s Description: S/MIME Cryptographic Signature
Bug#543375: initscripts: last superblock write time in future occurring sometimes
[Marco d'Itri]
Petter determined that this works. I think that the real problem is
that the RUN rules in 85-hwclock.rules are only executed if
ENV{BADYEAR} exists.
Could be.
Importing /etc/default/rcS as if were a list of variables looks like
a bad idea to me anyway, so please replace the file with a single
RUN rule which runs a shell script which properly sources
/etc/default/rcS and then runs hwclock if needed.
If this is the case, this patch should solve it. It is tested and
worked for me.
diff -urN ../util-linux-2.16.1/debian/hwclock-set
../util-linux-2.16.1-pere/debian/hwclock-set
--- ../util-linux-2.16.1/debian/hwclock-set 1970-01-01 01:00:00.0
+0100
+++ ../util-linux-2.16.1-pere/debian/hwclock-set2009-10-05
11:06:14.54511 +0200
@@ -0,0 +1,17 @@
+#!/bin/sh
+# Reset the System Clock to UTC if the hardware clock from which it
+# was copied by the kernel was in localtime.
+
+dev=$1
+
+if [ -f /etc/default/rcS ] ; then
+. /etc/default/rcS
+fi
+
+if [ yes != $UTC ] ; then
+if [ yes = $BADYEAR ] ; then
+ /sbin/hwclock --rtc=$dev --systz --localtime --noadjfile --badyear
+else
+ /sbin/hwclock --rtc=$dev --systz --localtime --noadjfile
+fi
+fi
diff -urN ../util-linux-2.16.1/debian/hwclock.rules
../util-linux-2.16.1-pere/debian/hwclock.rules
--- ../util-linux-2.16.1/debian/hwclock.rules 2009-10-05 11:02:20.0
+0200
+++ ../util-linux-2.16.1-pere/debian/hwclock.rules 2009-10-05
11:08:07.970856000 +0200
@@ -1,8 +1,4 @@
# Reset the System Clock to UTC if the hardware clock from which it was
# copied by the kernel was in localtime.
-KERNEL==rtc0, IMPORT{file}=/etc/default/rcS
-KERNEL==rtc0, ENV{UTC}!=yes, ENV{BADYEAR}!=yes, \
- RUN+=/sbin/hwclock --rtc=$root/$name --systz --localtime --noadjfile
-KERNEL==rtc0, ENV{UTC}!=yes, ENV{BADYEAR}==yes, \
- RUN+=/sbin/hwclock --rtc=$root/$name --systz --localtime --noadjfile
--badyear
+KERNEL==rtc0, RUN+=/lib/udev/hwclock-set $root/$name
diff -urN ../util-linux-2.16.1/debian/rules
../util-linux-2.16.1-pere/debian/rules
--- ../util-linux-2.16.1/debian/rules 2009-10-05 11:02:20.0 +0200
+++ ../util-linux-2.16.1-pere/debian/rules 2009-10-05 11:04:16.430037000
+0200
@@ -117,6 +117,7 @@
fi
if [ -f debian/util-linux/sbin/hwclock ] ; then \
install -m 644 debian/hwclock.rules
debian/util-linux/lib/udev/rules.d/85-hwclock.rules; \
+ install -m 755 debian/hwclock-set
debian/util-linux/lib/udev/hwclock-set; \
fi
# copy blkid library and symlink into udeb
ln debian/libblkid1/lib/libblkid.so.1.* debian/libblkid1-udeb/lib/
On IRC (#pkg-sysvinit), Scott James Remnant mentioned that a better
fix is probably to assume the hwclock is set to UTC in the early boot,
and just adjust it with the time zone when the zone info is available
without consulting the hardware clock. It would avoid all these
issues completely, but I got the impression it require a new upstream
version of util-linux.
Happy hacking,
--
Petter Reinholdtsen
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
On Mon, Oct 05, 2009 at 08:04:23AM +0200, Mader, Alexander (N-MSR) wrote: Something additional: I am running a Debian kernel not a self built one, and /lib/udev/rules.d/85-hwclock.rules was not there in the first place so I had to create it. That file is delivered by util-linux what version of util-linux do you have installed?? lamont -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
LaMont Jones schrieb: On Mon, Oct 05, 2009 at 08:04:23AM +0200, Mader, Alexander (N-MSR) wrote: Something additional: I am running a Debian kernel not a self built one, and /lib/udev/rules.d/85-hwclock.rules was not there in the first place so I had to create it. That file is delivered by util-linux what version of util-linux do you have installed?? Hello, since Sep 28th it is 2.16.1-3. My tests ran yesterday and today. Best regards, Alexander. 8 --- 8 :~# aptitude show util-linux Paket: util-linux Essentiell: ja Zustand: Installiert Automatisch installiert: nein Version: 2.16.1-3 8 --- 8 Aptitude 0.4.11.11: Protokoll Mo, Sep 28 2009 09:01:00 +0200 ... [AKTUALISIERUNG] util-linux 2.16-3 - 2.16.1-3 8 --- 8 -- Alexander Mader alexander.ma...@niles.de GPG public key: http://mader.seppi.de/pubkey-dienst.asc smime.p7s Description: S/MIME Cryptographic Signature
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Hi! The two lines in /lib/udev/rules.d/85-hwclock.rules didn't change a thing for me. Btw, I'm using the 2.6.30er kernel from here: http://kernel-archive.buildserver.net/debian-kernel sid main Since the last Debian update, however, the fsck problem doesn't seem to appear anymore. My guess is it's due to the new sysv-rc with the conversion to dependency based boot, but here's the whole update list: Aptitude 0.4.11.11: log report Mon, Oct 5 2009 15:28:54 +0200 IMPORTANT: this log only lists intended actions; actions which fail due to dpkg problems may not be completed. Will install 137 packages, and remove 12 packages. 44.4MB of disk space will be used === [REMOVE, NOT USED] claws-mail-smime-plugin [REMOVE, NOT USED] gstreamer0.10-ffmpeg [REMOVE, NOT USED] gstreamer0.10-fluendo-mp3 [REMOVE, NOT USED] gstreamer0.10-gnomevfs [REMOVE, NOT USED] libboost-date-time1.34.1 [REMOVE, NOT USED] libboost-filesystem1.39.0 [REMOVE, NOT USED] libboost-python1.39.0 [REMOVE, NOT USED] libboost-system1.39.0 [REMOVE, NOT USED] libboost-thread1.34.1 [REMOVE, NOT USED] libgail-common [REMOVE, NOT USED] shorewall-common [REMOVE, NOT USED] shorewall-shell [INSTALL, DEPENDENCIES] acroread-dictionary-en [INSTALL, DEPENDENCIES] bsfilter [INSTALL, DEPENDENCIES] claws-mail-bsfilter-plugin [INSTALL, DEPENDENCIES] claws-mail-fancy-plugin [INSTALL, DEPENDENCIES] csound [INSTALL, DEPENDENCIES] csound-gui [INSTALL, DEPENDENCIES] csound-manpages [INSTALL, DEPENDENCIES] csound-utils [INSTALL, DEPENDENCIES] insserv [INSTALL, DEPENDENCIES] libasync-interrupt-perl [INSTALL, DEPENDENCIES] libboost-date-time1.39.0 [INSTALL, DEPENDENCIES] libboost-filesystem1.40.0 [INSTALL, DEPENDENCIES] libboost-program-options1.39.0 [INSTALL, DEPENDENCIES] libboost-python1.40.0 [INSTALL, DEPENDENCIES] libboost-system1.40.0 [INSTALL, DEPENDENCIES] libboost-thread1.40.0 [INSTALL, DEPENDENCIES] libcommon-sense-perl [INSTALL, DEPENDENCIES] libcsound64-5.2 [INSTALL, DEPENDENCIES] libdb4.8 [INSTALL, DEPENDENCIES] libfltk1.1 [INSTALL, DEPENDENCIES] libfluidsynth1 [INSTALL, DEPENDENCIES] libgdbm-ruby1.8 [INSTALL, DEPENDENCIES] libgirepository1.0-0 [INSTALL, DEPENDENCIES] liblash2 [INSTALL, DEPENDENCIES] liblo0ldbl [INSTALL, DEPENDENCIES] libportmidi0 [INSTALL, DEPENDENCIES] libruby1.8 [INSTALL, DEPENDENCIES] ruby1.8 [UPGRADE] acroread-debian-files 0.0.39 - 0.0.40 [UPGRADE] adduser 3.110 - 3.111 [UPGRADE] alsa-utils 1.0.20-3 - 1.0.21-1 [UPGRADE] aspell-de 20071211-1 - 20071211-2 [UPGRADE] bind9-host 1:9.6.1.dfsg.P1-1 - 1:9.6.1.dfsg.P1-3 [UPGRADE] binutils 2.19.91.20090910-1 - 2.19.91.20090927-1 [UPGRADE] blender 2.49+dfsg-1 - 2.49+dfsg-2 [UPGRADE] bpython 0.9.4-1 - 0.9.5.1-1 [UPGRADE] claws-mail 3.5.0-2.1 - 3.7.2-2 [UPGRADE] claws-mail-acpi-notifier 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-archiver-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-attach-remover 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-attach-warner 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-bogofilter 3.5.0-2.1 - 3.7.2-2 [UPGRADE] claws-mail-cache-saver 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-extra-plugins 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-feeds-reader 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-fetchinfo-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-html2-viewer 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-i18n 3.5.0-2.1 - 3.7.2-2 [UPGRADE] claws-mail-mailmbox-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-multi-notifier 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-newmail-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-perl-filter 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-pgpinline 3.5.0-2.1 - 3.7.2-2 [UPGRADE] claws-mail-pgpmime 3.5.0-2.1 - 3.7.2-2 [UPGRADE] claws-mail-spam-report 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-synce-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-tnef-parser 3.5.0-3 - 3.7.2-1 [UPGRADE] claws-mail-vcalendar-plugin 3.5.0-3 - 3.7.2-1 [UPGRADE] dnsutils 1:9.6.1.dfsg.P1-1 - 1:9.6.1.dfsg.P1-3 [UPGRADE] freedink 1.08.20090120-1 - 1.08.20090918-1 [UPGRADE] freedink-dfarc 3.2.2-1 - 3.4-1 [UPGRADE] freedink-engine 1.08.20090120-1 - 1.08.20090918-1 [UPGRADE] fvwm 1:2.5.27.ds-12 - 1:2.5.28.ds-1 [UPGRADE] gcc-4.4-base 4.4.1-1 - 4.4.1-4 [UPGRADE] gnash 0.8.4-2 - 0.8.6-1 [UPGRADE] gnash-common 0.8.4-2 - 0.8.6-1 [UPGRADE] groff-base 1.20.1-5 - 1.20.1-6 [UPGRADE] info 4.13a.dfsg.1-4 - 4.13a.dfsg.1-5 [UPGRADE] initscripts 2.86.ds1-65 - 2.87dsf-6 [UPGRADE] install-info 4.13a.dfsg.1-4 - 4.13a.dfsg.1-5 [UPGRADE] lib32gcc1 1:4.4.1-1 - 1:4.4.1-4 [UPGRADE] lib32stdc++6 4.4.1-1 - 4.4.1-4 [UPGRADE] libanyevent-perl 5.112-1 - 5.200-1 [UPGRADE] libaprutil1 1.3.9+dfsg-1+b1 - 1.3.9+dfsg-2 [UPGRADE] libbind9-50 1:9.6.1.dfsg.P1-1 - 1:9.6.1.dfsg.P1-3 [UPGRADE] libclass-accessor-perl 0.33-1 - 0.34-1 [UPGRADE] libdatetime-locale-perl 1:0.43-1 - 1:0.44-1 [UPGRADE] libdb4.7 4.7.25-7 - 4.7.25-8 [UPGRADE] libdevmapper1.02.1 2:1.02.37-1 - 2:1.02.38-1 [UPGRADE] libdns50 1:9.6.1.dfsg.P1-1 - 1:9.6.1.dfsg.P1-3 [UPGRADE] libept0 0.5.28 - 0.5.29 [UPGRADE] libffi5
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Hello, this is my first post here, and I'd like to report the bug is also affecting me. I have been using Debian unstable (AMD64) for about two years (with regular updates every several days), and the issue started to occur around the beginning of August (never had this problem before). My first thought was it was my hardware malfunction, but fortunately I've found this topic, so hopefully it's not my machine :) My hardware is Intel Core Duo, Gigabyte GA-EP45T-UD3R; I have tried different kernels (distribution provided, and my own), but no difference. The problems has statistically occured once per a few reboots. But near a week ago I downgraded e2fslibs and e2fsprogs to the following versions: e2fslibs_1.41.3-1_amd64.deb e2fsprogs_1.41.3-1_amd64.deb and have not had this issue since then. I know it may be only a coincidence and a false positive, but maybe it could be worth a try... And maybe a clue (hopefully not a confusion). (of course I'll let you all know if this happens to me again using those downgraded libs). I hope this issue will be fixed soon. Good luck and keep the great work going on! B. Regards, Tomasz Grzelak
Bug#543375: initscripts: last superblock write time in future occurring sometimes
On Sep 25, Marco d'Itri m...@linux.it wrote:
KERNEL==rtc0, RUN+=logger.agent
Petter determined that this works. I think that the real problem is
that the RUN rules in 85-hwclock.rules are only executed if
ENV{BADYEAR} exists.
Importing /etc/default/rcS as if were a list of variables looks like a
bad idea to me anyway, so please replace the file with a single RUN
rule which runs a shell script which properly sources /etc/default/rcS
and then runs hwclock if needed.
--
ciao,
Marco
signature.asc
Description: Digital signature
Bug#543375: initscripts: last superblock write time in future occurring sometimes
[LaMont Jones] I completely agree on the severity, and would love to hear from the people who convinced me this would work. again. I guess that was me and Scott. :) The idea was to allow the init.d scripts detect when udev would take care of setting the system clock from the hardware clock when the rtc device became available, and make sure the scripts became no-ops in these cases to save 1-2 seconds during boot. I was told that udev would run the udev rule to set the system clock if it was enabled, and thus believed it would be sufficient to look for the existence of /dev/.udev to decode when to disable the init.d scripts. Obviously this assumtion is not quite right. I have asked Marco d'Itri, the udev maintainer, about this and he believed such udev rule would be called during boot even if no kernel module is loaded. Anyway, a workaround if the udev rule only work when rtc is a kernel module, would be a patch like this, to only disable the init.d scripts if rtc is a loaded module and udev is enabled. diff -ur util-linux-2.16.1/debian/hwclock.sh util-linux-2.16.1-pere/debian/hwclock.sh --- util-linux-2.16.1/debian/hwclock.sh 2009-09-25 10:49:27.0 +0200 +++ util-linux-2.16.1-pere/debian/hwclock.sh2009-09-25 10:54:22.0 +0200 @@ -67,7 +67,9 @@ case $1 in start) - if [ -d /dev/.udev ]; then + # No need to set the system clock if udev take care of it. + # This happen when rtc was loaded as a module by udev. + if [ -d /dev/.udev ] grep -qw ^rtc /proc/modules; then return 0 fi It should solve the described problems until we figure out why the udev rule is not called for home built kernels. Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
On Sep 25, Petter Reinholdtsen p...@hungry.com wrote: I was told that udev would run the udev rule to set the system clock if it was enabled, and thus believed it would be sufficient to look for the existence of /dev/.udev to decode when to disable the init.d scripts. Obviously this assumtion is not quite right. I have asked There is no doubt that a RUN rule matching on KERNEL==rtc0 is triggered even if the driver is built-in. Let's try some simple and stupid debugging. Somebody who can reliably reproduce the problem please add something like this to the *end* of /lib/udev/rules.d/85-hwclock.rules : KERNEL==rtc0, RUN+=/bin/touch /dev/rtc0-appeared KERNEL==rtc0, RUN+=logger.agent Anyway, a workaround if the udev rule only work when rtc is a kernel module, would be a patch like this, to only disable the init.d scripts This would only cure the symptoms. Please Cc me on updates. -- ciao, Marco signature.asc Description: Digital signature
Bug#543375: initscripts: last superblock write time in future occurring sometimes
severity 543375 serious thanks No response from the maintainers, no action, raising the priority. that is happening again and again, although it never happened in all the many years I am using Debian before. Now it is getting a pain, esp when you are watching the system fs[u]ck-ing away precious battery time on a lon distance flight. On Di, 01 Sep 2009, preining wrote: Dear maintainers of util-linux, since it happened again, I dare to ask if there is any comment from your side on that matter? It cannot be that normal booting forces me to fsck the whole partition. On Mo, 24 Aug 2009, Norbert Preining wrote: Hi Petter, sorry for my ignorance ... On Mon, 24 Aug 2009, Petter Reinholdtsen wrote: no module, but self compiled kernel Oh. No idea if the udev hooks are called when there is no kernel module loading. Me neither, but I guess that a init script checks for *modules* would be quite, well, innovative approach (or, more direct: stupid), there is the /sys hierarchy, and /proc/devices. Modules are the most unreliable way ;-) Anyway. What should then happen? Then the system clock should be set from the hardware clock in hwclockfirst.sh which is called before checkroot.sh, independently from any udev calls. Which means it wont happen. Ok, but since I don't have a guaranteed way to reproduce this problem I cannot check it either. Anyway, I wait for comments of util-linux maintainers, maybe they have better suggestions. Best wishes Norbert --- Dr. Norbert Preining prein...@logic.atVienna University of Technology Debian Developer prein...@debian.org Debian TeX Group gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 --- FORDSix pints of bitter. And quickly please, the world's about to end. BARMAN Oh yes, sir? Nice weather for it. --- Douglas Adams, The Hitchhikers Guide to the Galaxy -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
On Thu, Sep 24, 2009 at 02:54:31PM +0200, Norbert Preining wrote: No response from the maintainers, no action, raising the priority. that is happening again and again, although it never happened in all the many years I am using Debian before. Now it is getting a pain, esp when you are watching the system fs[u]ck-ing away precious battery time on a lon distance flight. I expect that I'll just revert the change and put hwclockfirst.sh back into play even on udev systems, until they can convince me that they've actually addressed things completely for both halves of the world. I completely agree on the severity, and would love to hear from the people who convinced me this would work. again. lamont -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Dear maintainers of util-linux, since it happened again, I dare to ask if there is any comment from your side on that matter? It cannot be that normal booting forces me to fsck the whole partition. On Mo, 24 Aug 2009, Norbert Preining wrote: Hi Petter, sorry for my ignorance ... On Mon, 24 Aug 2009, Petter Reinholdtsen wrote: no module, but self compiled kernel Oh. No idea if the udev hooks are called when there is no kernel module loading. Me neither, but I guess that a init script checks for *modules* would be quite, well, innovative approach (or, more direct: stupid), there is the /sys hierarchy, and /proc/devices. Modules are the most unreliable way ;-) Anyway. What should then happen? Then the system clock should be set from the hardware clock in hwclockfirst.sh which is called before checkroot.sh, independently from any udev calls. Which means it wont happen. Ok, but since I don't have a guaranteed way to reproduce this problem I cannot check it either. Anyway, I wait for comments of util-linux maintainers, maybe they have better suggestions. Best wishes Norbert --- Dr. Norbert Preining prein...@logic.atVienna University of Technology Debian Developer prein...@debian.org Debian TeX Group gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 --- LUBCROY (n.) The telltale little lump in the top of your swimming trunks which tells you you are going to have to spend half an hour with a safety pin trying to pull the drawstring out again. --- Douglas Adams, The Meaning of Liff -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Package: initscripts Version: 2.87dsf-2 Severity: important Dear maintainers, I am not sure whom this bug belongs, please reassign accordingly. Since some time (probably update of initscripts that pulled in insserv, but hard to say) it happens sometimes that at booting time I see: ... checking filesystems ... sda3 ... Last superblock write time in future ... Inconsistencies found, please check manually (copied from by memory, but I guess you know what it is). The point is that it seems that the last superblock write time is in local time which in my case is CEST (GMT+2), and the check time of fsck is in GMT, that gives 2 hours difference. Now, I don't remember ever having seen this problem, so I assume that it surfaced with the new uploads. This is quite a pain since everytime it happens a full check has to be done, which takes ages. Thanks for any insightfull remarks and fixes Norbert -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.31-rc7 (SMP w/2 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages initscripts depends on: ii debianutils 3.2.1 Miscellaneous utilities specific t ii libc6 2.9-25 GNU C Library: Shared libraries ii lsb-base 3.2-23 Linux Standard Base 3.2 init scrip ii mount 2.16-3 Tools for mounting and manipulatin ii sysvinit-utils2.87dsf-2 System-V-like utilities Versions of packages initscripts recommends: ii e2fsprogs 1.41.9-1 ext2/ext3/ext4 file system utiliti ii psmisc22.8-1 utilities that use the proc file s initscripts suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
[Norbert Preining] Dear maintainers, I am not sure whom this bug belongs, please reassign accordingly. I suspect a combination of things, and udev, util-linux, tzdata or initscripts are probably the source of your problem. Can you provide the output from these commands: ls -ld /etc/localtime /dev/.udev grep ^UTC /etc/default/rcS ls /etc/rcS.d df -hP The problem will happen when HW clock is not on UTC and /usr/ is a separate partition from /. Now, I don't remember ever having seen this problem, so I assume that it surfaced with the new uploads. As far as I know, this problem have existed since forever, so I am more surprised you see it just now. :) It is believed to be fixed recently, when /etc/localtime was converted was converted from a symlink into /usr/ to a copy of the file in /usr/.. :) Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Hi Petter, On Mon, 24 Aug 2009, Petter Reinholdtsen wrote: Can you provide the output from these commands: Sure: ls -ld /etc/localtime /dev/.udev drwxr-xr-x 6 root root 140 2009-08-24 19:53 /dev/.udev -rw-r--r-- 1 root root 2211 2009-08-21 16:53 /etc/localtime grep ^UTC /etc/default/rcS UTC=no ls /etc/rcS.d READMES26cryptdisks-early S40networking S02hostname.shS26cryptmount-early S40pcmciautils S02mountkernfs.sh S28cryptdisks S45mountnfs.sh S02readahead S30checkfs.sh S46mountnfs-bootclean.sh S03udev S30etc-setserial S46setserial S04mountdevsubfs.sh S30procps S47lm-sensors S05bootlogd S31hibernate S48console-screen.sh S05keymap.sh S34fuse S49console-setup S06keyboard-setup S35mountall.shS50alsa-utils S07hdparm S36mountall-bootclean.sh S55bootmisc.sh S08hwclockfirst.shS36udev-mtab S55urandom S10checkroot.sh S37mountoverflowtmp S70x11-common S11hwclock.sh S38pppd-dns S75policykit S12mtab.shS38resolvconf S75sudo S18ifupdown-clean S39ifupdown S80microcode.ctl S20module-init-tools S39readahead-desktop S99stop-bootlogd-single df -hP FilesystemSize Used Avail Use% Mounted on /dev/sda3 196G 173G 14G 93% / tmpfs 2.0G 8.0K 2.0G 1% /lib/init/rw udev 10M 168K 9.9M 2% /dev tmpfs 2.0G 0 2.0G 0% /dev/shm /dev/sda2 81G 72G 8.2G 90% /xp The problem will happen when HW clock is not on UTC and /usr/ is a separate partition from /. No separate partition here. As far as I know, this problem have existed since forever, so I am more surprised you see it just now. :) It is believed to be fixed Ha, maybe I didn't have any crashed for such a long time ... haha, just joking, two defective hard discs in 6 months in my brand new sony laptop. Hope the above output helps, if not let me know what else I can do. Best wishes Norbert --- Dr. Norbert Preining prein...@logic.atVienna University of Technology Debian Developer prein...@debian.org Debian TeX Group gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 --- GLUTT LODGE (n.) The place where food can be stored after having a tooth extracted. Some Arabs can go without sustenance for up to six weeks on a full glutt lodge, hence the expression 'the shit of the dessert'. --- Douglas Adams, The Meaning of Liff -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
[Norbert Preining] The problem will happen when HW clock is not on UTC and /usr/ is a separate partition from /. No separate partition here. No hardware clock on UTC either. One last thing to check. What is the output from 'ls -l /dev/rtc; lsmod|grep rtc'? I suspect your machine boot so fast that the system clock is not yet set from the hardware clock (which is done using a udev hook in util-linux when udev is enabled) when the checkroot.sh script is running. You can test this hypotesis by disabling the hwclock udev rule and comment out the test for /dev/.udev in hwclockfirst.sh. If this is the case, the problem probably need to be fixed in util-linux, perhaps by changing the hwclockfirst script to wait for the hwclock to be set before allowing checkroot.sh to run. Hope the above output helps, if not let me know what else I can do. It definitely help to narrow down the problem. :) Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
On Mon, 24 Aug 2009, Petter Reinholdtsen wrote: No hardware clock on UTC either. One last thing to check. What is the output from 'ls -l /dev/rtc; lsmod|grep rtc'? lrwxrwxrwx 1 root root 4 2009-08-24 21:51 /dev/rtc - rtc0 $ ls -l /dev/rtc0 crw-rw 1 root root 254, 0 2009-08-24 21:51 /dev/rtc0 no module, but self compiled kernel $ grep RTC /boot/config-2.6.31-rc7 | grep ^CONFIG CONFIG_HPET_EMULATE_RTC=y CONFIG_RTC_LIB=y CONFIG_RTC_CLASS=y CONFIG_RTC_HCTOSYS=y CONFIG_RTC_HCTOSYS_DEVICE=rtc0 CONFIG_RTC_INTF_SYSFS=y CONFIG_RTC_INTF_PROC=y CONFIG_RTC_INTF_DEV=y CONFIG_RTC_DRV_RS5C372=m CONFIG_RTC_DRV_CMOS=y $ I suspect your machine boot so fast that the system clock is not yet set from the hardware clock (which is done using a udev hook in util-linux when udev is enabled) when the checkroot.sh script is running. You can test this hypotesis by disabling the hwclock udev rule and comment out the test for /dev/.udev in hwclockfirst.sh. If What should then happen? Best wishes Norbert --- Dr. Norbert Preining prein...@logic.atVienna University of Technology Debian Developer prein...@debian.org Debian TeX Group gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 --- SKIBBEREEN (n.) The noise made by a sunburned thighs leaving plastic chair. --- Douglas Adams, The Meaning of Liff -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
reassign 543375 util-linux thanks [Norbert Preining] no module, but self compiled kernel Oh. No idea if the udev hooks are called when there is no kernel module loading. What should then happen? Then the system clock should be set from the hardware clock in hwclockfirst.sh which is called before checkroot.sh, independently from any udev calls. Anyway, I believe this must be an issue with the new util-linux package, and reassign it there. Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#543375: initscripts: last superblock write time in future occurring sometimes
Hi Petter, sorry for my ignorance ... On Mon, 24 Aug 2009, Petter Reinholdtsen wrote: no module, but self compiled kernel Oh. No idea if the udev hooks are called when there is no kernel module loading. Me neither, but I guess that a init script checks for *modules* would be quite, well, innovative approach (or, more direct: stupid), there is the /sys hierarchy, and /proc/devices. Modules are the most unreliable way ;-) Anyway. What should then happen? Then the system clock should be set from the hardware clock in hwclockfirst.sh which is called before checkroot.sh, independently from any udev calls. Which means it wont happen. Ok, but since I don't have a guaranteed way to reproduce this problem I cannot check it either. Anyway, I wait for comments of util-linux maintainers, maybe they have better suggestions. Best wishes and thanks a lot Norbert --- Dr. Norbert Preining prein...@logic.atVienna University of Technology Debian Developer prein...@debian.org Debian TeX Group gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 --- KURDISTAN (n.) Hard stare given by a husband to his wife when he notices a sharp increase in the number of times he answers the phone to be told, 'Sorry, wrong number.' --- Douglas Adams, The Meaning of Liff -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
