Tumblewwed cannot be update crahed sytemm form 20220423 to 20220425
Dear OpenSUSE experts, I would like to recover crashed system via update from USB stick. Instalation of new version with no partition to be formated does not finish correctly but raises snapshot version. All listed errors are mkdir based Snapshot update cannot remove snapshot version= repository snapshot. All *repo files were moved from original position to backup directory to try to recover incorrect items in it Is there any solution to remove such notes from sytem? rpm managing files does not work on original system and rpm database Is there any possibility to get information about fault case and way to recover it. Update crashes with no umount od original / possible lsof in not present op the system and top, ps aux does not show me anythink usable for me. I look forward hearing from you Yours fiathfully Peter Fodrek,j.r. P.S. Fault during update is in .rb script in missing method
Dear Mr. Braun, Dňa 28. 4. 2022 o 21:39 Axel Braun napísal(a):
Am Donnerstag, 28. April 2022, 16:53:57 CEST schrieb Peter Fodrek,ml.- súkromne:
I would like to recover crashed system via update from USB stick. Did you try booting into an older snapshot first, and do a 'snapper rollback' from there?
Thnak you for your answer Crahed system is unbootable and in chroot environment there is problem with glibc. Crahed shanpshot was 20210523. after using from USB stck 1. Update crashes due to snapshot item from somewhere unable to remove due to missing method 2. Clean install withoout formating any partition from mount swap from device files /dev/sdb2 and /dev/sdb7 / from device file /dev/sdb3 /home from /dev/sdb4 /boot from /dev/sdb5 /var/log from /dev/sdb6 /boot/efi from /dev/sdb1 and /usr from /dev/sdb8 Maybe /usr on separate partition is able to be cause as of /usr/bin /usr/sbin /usr/lib /usr/local etc. I look forward hearing from you and anybody else Yours faithfully Peter
Hello Peter, reading through the support list, you are facing regular update issues. Not sure what and how you are doing it, but you should consider to change the way you upgrade..... Am Freitag, 29. April 2022, 08:50:13 CEST schrieb Peter Fodrek,ml.-súkromne:
I would like to recover crashed system via update from USB stick.
Did you try booting into an older snapshot first, and do a 'snapper rollback' from there?
Thnak you for your answer
Crahed system is unbootable and in chroot environment there is problem with glibc.
Crahed shanpshot was 20210523. after using from USB stck
So your last working system was snapshot 20210523. For some reason it did not boot anymore (out of the blue sky, or after a failed update?), and you try now to upgrade it from an up-to-date snapshot. Is that right?
1. Update crashes due to snapshot item from somewhere unable to remove due to missing method
snapshots are in /.snapshots ...just in case....
2. Clean install withoout formating any partition from mount
swap from device files /dev/sdb2 and /dev/sdb7
/ from device file /dev/sdb3
/home from /dev/sdb4
/boot from /dev/sdb5
/var/log from /dev/sdb6
/boot/efi from /dev/sdb1
and
/usr from /dev/sdb8
and that worked or not?
Maybe /usr on separate partition is able to be cause as of /usr/bin /usr/sbin /usr/lib /usr/local etc.
I dont understand what you mean here.... Maybe, in case of a new installation, you should consider to tie all your partitions - except /dev/sdb4 - together in a LVM volume and follow the default settings for tumbleweed regarding partitioning? HTH Axel
Dear Axel Dňa 29. 4. 2022 o 11:22 Axel Braun napísal(a):
reading through the support list, you are facing regular update issues. Not sure what and how you are doing it, but you should consider to change the way you upgrade.....
Am Freitag, 29. April 2022, 08:50:13 CEST schrieb Peter Fodrek,ml.-súkromne:
I would like to recover crashed system via update from USB stick. Did you try booting into an older snapshot first, and do a 'snapper rollback' from there? Thnak you for your answer
Crahed system is unbootable and in chroot environment there is problem with glibc.
Crahed shanpshot was 20210523. after using from USB stck So your last working system was snapshot 20210523. For some reason it did not boot anymore (out of the blue sky, or after a failed update?), and you try now to upgrade it from an up-to-date snapshot. Is that right?
Yes, and it was crashed after power lost during running zypper dup.
1. Update crashes due to snapshot item from somewhere unable to remove due to missing method
snapshots are in /.snapshots ...just in case....
Thank you for your information. I will try.
2. Clean install withoout formating any partition from mount
swap from device files /dev/sdb2 and /dev/sdb7
/ from device file /dev/sdb3
/home from /dev/sdb4
/boot from /dev/sdb5
/var/log from /dev/sdb6
/boot/efi from /dev/sdb1
and
/usr from /dev/sdb8
and that worked or not?
Instalation was not complete due to fialed mkdir about 100 times in 4000 packages. But No nformation about which directory is missing - mostly but no exclusively -docs and locale packages.
Maybe /usr on separate partition is able to be cause as of /usr/bin /usr/sbin /usr/lib /usr/local etc. I dont understand what you mean here....
I mentioned that /usr on separate patrition may cause the problem. As well as cross partition links
Maybe, in case of a new installation, you should consider to tie all your partitions - except /dev/sdb4 - together in a LVM volume and follow the default settings for tumbleweed regarding partitioning?
Even if they were separate originally? So best was is to make format for all "non home" partitions. It was last chance for me due to lost of third party SW. I was to do so only when every other attempt will fail. And It seems that non only me has no idea what to do, yet. Thank you for your help Kind regards Peter
Am Freitag, 29. April 2022, 11:40:05 CEST schrieb Peter Fodrek,ml.-súkromne:
/usr from /dev/sdb8
and that worked or not?
Instalation was not complete due to fialed mkdir about 100 times in 4000 packages. But No nformation about which directory is missing - mostly but no exclusively -docs and locale packages.
Try to use a live system and make sure the file systems are not corrupt
Dňa 29. 4. 2022 o 11:57 Axel Braun napísal(a):
Am Freitag, 29. April 2022, 11:40:05 CEST schrieb Peter Fodrek,ml.-súkromne:
/usr from /dev/sdb8 and that worked or not? Instalation was not complete due to fialed mkdir about 100 times in 4000 packages. But No nformation about which directory is missing - mostly but no exclusively -docs and locale packages. Try to use a live system and make sure the file systems are not corrupt
Every time I start instalation I use terminal based chceck via fsck /dev/sdb*;fsck -pvcf /dev/sdb*; fsck -pvf /dev/sdb*; fsck /dev/sdb* that lasts about 110 minutes for all(20) swap,ext4, vfat and ntfs partitions on /dev/sdb It may be enough to repair partitions Peter
On 2022-04-29 11:22, Axel Braun wrote:
Hello Peter,
reading through the support list, you are facing regular update issues. Not sure what and how you are doing it, but you should consider to change the way you upgrade.....
Am Freitag, 29. April 2022, 08:50:13 CEST schrieb Peter Fodrek,ml.-súkromne:
I would like to recover crashed system via update from USB stick.
Did you try booting into an older snapshot first, and do a 'snapper rollback' from there?
Thnak you for your answer
Crahed system is unbootable and in chroot environment there is problem with glibc.
Crahed shanpshot was 20210523. after using from USB stck
So your last working system was snapshot 20210523. For some reason it did not boot anymore (out of the blue sky, or after a failed update?), and you try now to upgrade it from an up-to-date snapshot. Is that right?
1. Update crashes due to snapshot item from somewhere unable to remove due to missing method
snapshots are in /.snapshots ...just in case....
2. Clean install withoout formating any partition from mount
swap from device files /dev/sdb2 and /dev/sdb7
/ from device file /dev/sdb3
/home from /dev/sdb4
/boot from /dev/sdb5
/var/log from /dev/sdb6
/boot/efi from /dev/sdb1
and
/usr from /dev/sdb8
and that worked or not?
Notice that with that layout it is impossible to use snapshots and rollback feature. Feature should be disabled to avoid confusion. -- Cheers / Saludos, Carlos E. R. (from 15.3 x86_64 at Telcontar)
Dear Mr. Carlos, Dňa 29. 4. 2022 o 13:00 Carlos E. R. napísal(a):
On 2022-04-29 11:22, Axel Braun wrote:
Hello Peter,
reading through the support list, you are facing regular update issues. Not sure what and how you are doing it, but you should consider to change the way you upgrade.....
Am Freitag, 29. April 2022, 08:50:13 CEST schrieb Peter Fodrek,ml.-súkromne:
I would like to recover crashed system via update from USB stick.
Did you try booting into an older snapshot first, and do a 'snapper rollback' from there?
Thnak you for your answer
Crahed system is unbootable and in chroot environment there is problem with glibc.
Crahed shanpshot was 20210523. after using from USB stck
So your last working system was snapshot 20210523. For some reason it did not boot anymore (out of the blue sky, or after a failed update?), and you try now to upgrade it from an up-to-date snapshot. Is that right?
1. Update crashes due to snapshot item from somewhere unable to remove due to missing method
snapshots are in /.snapshots ...just in case....
2. Clean install withoout formating any partition from mount
swap from device files /dev/sdb2 and /dev/sdb7
/ from device file /dev/sdb3
/home from /dev/sdb4
/boot from /dev/sdb5
/var/log from /dev/sdb6
/boot/efi from /dev/sdb1
and
/usr from /dev/sdb8
and that worked or not?
Notice that with that layout it is impossible to use snapshots and rollback feature.
I do not metion filesystem snapshot but Tumbleweed snapshot as of Subject:New Tumbleweed snapshot 20220427 released! Date: Fri, 29 Apr 2022 02:00:46 +0000 To: factory@lists.opensuse.org or New Tumbleweed snapshot 20220426 released! Wed, 27 Apr 2022 15:00:45 +0000 To: factory@lists.opensuse.org etc. Kind regards Peter
On 2022-04-29 16:54, Peter Fodrek,ml.-súkromne wrote:
Dear Mr. Carlos,
Dňa 29. 4. 2022 o 13:00 Carlos E. R. napísal(a):
On 2022-04-29 11:22, Axel Braun wrote:
Hello Peter,
reading through the support list, you are facing regular update issues. Not sure what and how you are doing it, but you should consider to change the way you upgrade.....
Am Freitag, 29. April 2022, 08:50:13 CEST schrieb Peter Fodrek,ml.-súkromne:
I would like to recover crashed system via update from USB stick.
Did you try booting into an older snapshot first, and do a 'snapper rollback' from there?
Thnak you for your answer
Crahed system is unbootable and in chroot environment there is problem with glibc.
Crahed shanpshot was 20210523. after using from USB stck
So your last working system was snapshot 20210523. For some reason it did not boot anymore (out of the blue sky, or after a failed update?), and you try now to upgrade it from an up-to-date snapshot. Is that right?
1. Update crashes due to snapshot item from somewhere unable to remove due to missing method
snapshots are in /.snapshots ...just in case....
2. Clean install withoout formating any partition from mount
swap from device files /dev/sdb2 and /dev/sdb7
/ from device file /dev/sdb3
/home from /dev/sdb4
/boot from /dev/sdb5
/var/log from /dev/sdb6
/boot/efi from /dev/sdb1
and
/usr from /dev/sdb8
and that worked or not?
Notice that with that layout it is impossible to use snapshots and rollback feature.
I do not metion filesystem snapshot but Tumbleweed snapshot as of
Subject:New Tumbleweed snapshot 20220427 released!
Date: Fri, 29 Apr 2022 02:00:46 +0000
To: factory@lists.opensuse.org
or
New Tumbleweed snapshot 20220426 released!
Wed, 27 Apr 2022 15:00:45 +0000
To: factory@lists.opensuse.org
You do not understand, I see. Axel is asking you to use btrfs snapshots feature, when he said:
snapshots are in /.snapshots ...just in case....
and when he said: ]> Did you try booting into an older snapshot first, and do a 'snapper rollback' ]> from there? That refers to btrfs snapshot feature, installed by default on Tumbleweed, not to snapshot as in "Subject:New Tumbleweed snapshot 20220427 released!". But you can not do that, because you have a different partition for /usr and /boot and /var/log/. The "modern" repair methods will not work for you. -- Cheers / Saludos, Carlos E. R. (from 15.3 x86_64 at Telcontar)
participants (3)
-
Axel Braun -
Carlos E. R. -
Peter Fodrek,ml.-súkromne