hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout: Label: 'fedora_localhost-live' uuid: ceb296ac-852b-420c-af45- 3cb3d5cb1f9d Total devices 1 FS bytes used 156.97GiB devid 1 size 269.47GiB used 200.02GiB path /dev/nvme0n1p6
nvme0n1
├─nvme0n1p1 │ vfat FAT32 A544-D3FB ├─nvme0n1p2 │ ext4 1.0 e42d229f-bb34-460b-80cc-4b50fcdf1754 ├─nvme0n1p3 │ swap 1 3021cbd8-230e-4c61-8f2a-6faeb8fd6b5f [SWAP] ├─nvme0n1p4 │ vfat FAT32 E9A3-DD63 581,4M 3% /boot/efi ├─nvme0n1p5 │ ext4 1.0 92299ff3-bb02-42a8-a734-16b7c0dc16a6 558,7M 36% /boot └─nvme0n1p6 btrfs fedora_localhost-live ceb296ac-852b-420c-af45-3cb3d5cb1f9d /var/lib/docker/btrfs
/home
/
i tried it with gparted, but if i want to shrink, a warning appears, that it is possible,to not boot anymore.
my question: is there a safe way to shrink?
regards karl
Hello, i want to shrink the big partition Nr 6, as small as possible without touching the used space.
The ordering is going to prevent you from shrinking.
/boot is last and you likely cannot get enough space out of that one. I am not sure what would happen if you made p2 smaller or if it would even let you create a p7 that is after p2 but before p3.
And with all shrinks you must shrink the fs first and then make the partition smaller. if you reverse the order it won't work and you will have to revert the partition change to mount the fs and/or find your data after reboot. it may appear to work immediately (still has access) but fail hard on reboot.
And you cannot resize smaller some fses (xfs for sure, btrfs seems to have limits and needs to be carefully done--don't ask me how I run away from anything btrfs).
On Wed, Jul 26, 2023 at 12:22 PM KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout: Label: 'fedora_localhost-live' uuid: ceb296ac-852b-420c-af45- 3cb3d5cb1f9d Total devices 1 FS bytes used 156.97GiB devid 1 size 269.47GiB used 200.02GiB path /dev/nvme0n1p6
nvme0n1
├─nvme0n1p1 │ vfat FAT32 A544-D3FB ├─nvme0n1p2 │ ext4 1.0 e42d229f-bb34-460b-80cc-4b50fcdf1754 ├─nvme0n1p3 │ swap 1 3021cbd8-230e-4c61-8f2a-6faeb8fd6b5f [SWAP] ├─nvme0n1p4 │ vfat FAT32 E9A3-DD63 581,4M 3% /boot/efi ├─nvme0n1p5 │ ext4 1.0 92299ff3-bb02-42a8-a734-16b7c0dc16a6 558,7M 36% /boot └─nvme0n1p6 btrfs fedora_localhost-live ceb296ac-852b-420c-af45-3cb3d5cb1f9d /var/lib/docker/btrfs
/home
/
i tried it with gparted, but if i want to shrink, a warning appears, that it is possible,to not boot anymore.
my question: is there a safe way to shrink?
regards karl _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
Hmmh, that is very sad. I never touched something, only installed with the preferred options from the installer... Fyi: I want to make space for a second linux Installation, not an additional partition.
Does it mean that i have erase the whole disk, because it is btrfs?
I got the partition that was last wrong.
You would have to know the sizes and used of the various mount points and have to know how/if btrfs has enough free and how to re-arrange that free space to be at the end so you can resize.
In general adding a 2nd installation after the fact is very complicated. Even adding the 2nd install when you left free space is not the most straightforward as the installer program has to do everything right, and the installer person has to also not pick the wrong options and damage the 1st install.
The few times I do a 2nd install the 2nd install is on its own boot device (usually a usb disk), and I tend to not run 2 installs for very long. I have done 3 install with the same grub setup (legacy boot) but that setup was designed to have 3 installs and the grub merging of the 3 /boots into one set of boot options had to be done entirely manually as each installer at the time wanted to do its own grub.cfg and not share. That all might be better using UEFI since UEFI can work as the switch between the 3 /boots.
On Thu, Jul 27, 2023 at 1:27 AM Karlderletzte karlderletzte@holnishaus.de wrote:
Hmmh, that is very sad. I never touched something, only installed with the preferred options from the installer... Fyi: I want to make space for a second linux Installation, not an additional partition.
Does it mean that i have erase the whole disk, because it is btrfs? _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
Thanks you for the info. Fyi: my laptop use UEFI.
----------------------------------------
Von: Roger Heflin rogerheflin@gmail.com An: Community support for Fedora users users@lists.fedoraproject.org Datum: 27.07.2023 12:48:01 Betreff: Re: Need help; safe way to shrink my fedora38 installation disk
I got the partition that was last wrong.
You would have to know the sizes and used of the various mount points and have to know how/if btrfs has enough free and how to re-arrange that free space to be at the end so you can resize.
In general adding a 2nd installation after the fact is very complicated. Even adding the 2nd install when you left free space is not the most straightforward as the installer program has to do everything right, and the installer person has to also not pick the wrong options and damage the 1st install.
The few times I do a 2nd install the 2nd install is on its own boot device (usually a usb disk), and I tend to not run 2 installs for very long. I have done 3 install with the same grub setup (legacy boot) but that setup was designed to have 3 installs and the grub merging of the 3 /boots into one set of boot options had to be done entirely manually as each installer at the time wanted to do its own grub.cfg and not share. That all might be better using UEFI since UEFI can work as the switch between the 3 /boots.
On Thu, Jul 27, 2023 at 1:27 AM Karlderletzte karlderletzte@holnishaus.de wrote:
Hmmh, that is very sad. I never touched something, only installed with the preferred options from the installer... Fyi: I want to make space for a second linux Installation, not an additional partition.
Does it mean that i have erase the whole disk, because it is btrfs? _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
On Wed, Jul 26, 2023 at 2:22 PM KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout: [...]
my question:
is there a safe way to shrink?
The only safe way to shrink is to have a backup. Shrink is I/O intensive
and has a much higher probability of revealing any weaknesses in hardware or software than normal operations where weaknesses may remain undiscovered for a long time. Also note that running a filesystem with too little free space slows down file operations and also increases the probability that weaknesses will be revealed.
See: < https://btrfs.readthedocs.io/en/latest/btrfs-filesystem.html#man-filesystem-...
Adding to what George says.
If doing a 2nd install, have a plan of what to do if it all fails badly and you end up with no working installs (there are posts all of the time for failures, not sure how many dual installs work as people only post when it goes wrong). There are a few places it can go wrong during the install, and if it goes wrong there are a very limited number of experts that can sort out what is salvageable or not. Make sure to have a tested (to boot) livecd (not a installer iso) such that you can boot up and examine the disk.
On Thu, Jul 27, 2023 at 5:28 AM George N. White III gnwiii@gmail.com wrote:
On Wed, Jul 26, 2023 at 2:22 PM KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout: [...]
my question: is there a safe way to shrink?
The only safe way to shrink is to have a backup. Shrink is I/O intensive and has a much higher probability of revealing any weaknesses in hardware or software than normal operations where weaknesses may remain undiscovered for a long time. Also note that running a filesystem with too little free space slows down file operations and also increases the probability that weaknesses will be revealed.
See: https://btrfs.readthedocs.io/en/latest/btrfs-filesystem.html#man-filesystem-resize
-- George N. White III
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
On Wed, 26 Jul 2023 19:21:50 +0200 KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout:
[snip]
i tried it with gparted, but if i want to shrink, a warning appears, that it is possible,to not boot anymore.
my question: is there a safe way to shrink?
I don't know that I would even be able to help, since I use fixed ext4 partitions for the same purpose, but I think the output of df would give a better idea of your system layout and partition sizes and usage in case there is someone who *can* help.
I'm not familiar with btrfs from using it, only from reading the comments of people responding to problems, but from what I understand, this should be *easy* in btrfs for someone who knows what they are doing (definitely not me!).
Am Donnerstag, dem 27.07.2023 um 08:27 -0700 schrieb stan via users:
On Wed, 26 Jul 2023 19:21:50 +0200 KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout:
[snip]
i tried it with gparted, but if i want to shrink, a warning appears, that it is possible,to not boot anymore.
my question: is there a safe way to shrink?
I don't know that I would even be able to help, since I use fixed ext4 partitions for the same purpose, but I think the output of df would give a better idea of your system layout and partition sizes and usage in case there is someone who *can* help.
here is the output of df and btrfs filesystem. i resized the filesystem to a desired value. my ssd has 500GB capacity. it is the nvme0n1..6
sudo btrfs filesystem sync / sudo btrfs filesystem df / Data, single: total=195.01GiB, used=154.75GiB System, single: total=4.00MiB, used=48.00KiB Metadata, single: total=5.01GiB, used=2.21GiB GlobalReserve, single: total=301.27MiB, used=0.00B
df Dateisystem 1K-Blöcke Benutzt Verfügbar Verw% Eingehängt auf devtmpfs 4096 0 4096 0% /dev tmpfs 7850952 0 7850952 0% /dev/shm tmpfs 3140384 1964 3138420 1% /run /dev/nvme0n1p6 282561536 164889684 115040620 59% / /dev/nvme0n1p6 282561536 164889684 115040620 59% /home tmpfs 7850952 12 7850940 1% /tmp /dev/nvme0n1p5 996780 355908 572060 39% /boot /dev/nvme0n1p4 613184 17796 595388 3% /boot/efi tmpfs 1570188 192 1569996 1% /run/user/1000 /dev/sda1 495312 5920 412352 2% /run/media/held/40f020fb-33eb-406b-a3b0-7f7ba297557b
I'm not familiar with btrfs from using it, only from reading the comments of people responding to problems, but from what I understand, this should be *easy* in btrfs for someone who knows what they are doing (definitely not me!). _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
On Thu, 27 Jul 2023 18:47:50 +0200 KarlderLetzte karlderletzte@holnishaus.de wrote:
Am Donnerstag, dem 27.07.2023 um 08:27 -0700 schrieb stan via users:
On Wed, 26 Jul 2023 19:21:50 +0200 KarlderLetzte karlderletzte@holnishaus.de wrote:
hello everybody,
i need some space on disk for a second installation. therefore i want to safely shrink my fedora installation.
here is the disk layout:
[snip]
i tried it with gparted, but if i want to shrink, a warning appears, that it is possible,to not boot anymore.
my question: is there a safe way to shrink?
I don't know that I would even be able to help, since I use fixed ext4 partitions for the same purpose, but I think the output of df would give a better idea of your system layout and partition sizes and usage in case there is someone who *can* help.
here is the output of df and btrfs filesystem. i resized the filesystem to a desired value. my ssd has 500GB capacity. it is the nvme0n1..6
sudo btrfs filesystem sync / sudo btrfs filesystem df / Data, single: total=195.01GiB, used=154.75GiB System, single: total=4.00MiB, used=48.00KiB Metadata, single: total=5.01GiB, used=2.21GiB GlobalReserve, single: total=301.27MiB, used=0.00B
df Dateisystem 1K-Blöcke Benutzt Verfügbar Verw% Eingehängt auf devtmpfs 4096 0 4096 0% /dev tmpfs 7850952 0 7850952 0% /dev/shm tmpfs 3140384 1964 3138420 1% /run /dev/nvme0n1p6 282561536 164889684 115040620 59% / /dev/nvme0n1p6 282561536 164889684 115040620 59% /home tmpfs 7850952 12 7850940 1% /tmp /dev/nvme0n1p5 996780 355908 572060 39% /boot /dev/nvme0n1p4 613184 17796 595388 3% /boot/efi tmpfs 1570188 192 1569996 1% /run/user/1000 /dev/sda1 495312 5920 412352 2% /run/media/held/40f020fb-33eb-406b-a3b0-7f7ba297557b
No one is stepping forward, but this looks fairly straightforward, since there is unused space; I think the commands in man btrfs-filesystem and man btrfs-subvolume are what you want, but I am not qualified to tell you how to use them. It seems there aren't any btrfs experts on this list. Maybe try discourse, the web forum? Or a web search might turn up a recipe, since this has to be a common request.
Hello everybody, thank you all for the information. Unfortunately i do not fully understand, how to manage the shrinking. For now, i will install the new linux on a external disc, boot and then copy content from the f38 installation. the old way, long and work intensive (finding the correct partition size, choose the filesystem typ and so on.). i think, it is annoying, that btrfs with f38 is so hard to use for this task. my new filesystem will not be btfrs.
Regards Karl
Am Freitag, dem 28.07.2023 um 06:40 -0700 schrieb stan via users:
On Thu, 27 Jul 2023 18:4 KarlderLetzte karlderletzte@holnishaus.de wrote:
No one is stepping forward, but this looks fairly straightforward, since there is unused space; I think the commands in man btrfs-filesystem and man btrfs-subvolume are what you want, but I am not qualified to tell you how to use them. It seems there aren't any btrfs experts on this list. Maybe try discourse, the web forum? Or a web search might turn up a recipe, since this has to be a common request.
_______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
On Sat, Jul 29, 2023 at 3:04 PM KarlderLetzte karlderletzte@holnishaus.de wrote:
Hello everybody, thank you all for the information. Unfortunately i do not fully understand, how to manage the shrinking. For now, i will install the new linux on a external disc, boot and then copy content from the f38 installation.
Am I correct in thinking you were able to shrink the btrfs filesystem, but are unable to find software to adjust the partition size to match?
You could use the new disk for a backup, then shrink the oversized partition. See: https://feldspaten.org/2018/06/01/resizing-btrfs-max/ -- shrink the btrfs filesystem a bit smaller than you want, use fstab to resize the partition, then expand btrfs to use the full partition.
https://itsfoss.com/btrfs/ lists the advantages of btrfs
the old way, long and work intensive (finding the correct partition size, choose the filesystem typ and so on.). i think, it is annoying, that btrfs with f38 is so hard to use for this task. my new filesystem will not be btfrs.
All filesystems need backups. It seems wasteful to do an extra install.
Am Sonntag, dem 30.07.2023 um 11:46 -0300 schrieb George N. White III:
On Sat, Jul 29, 2023 at 3:04 PM KarlderLetzte karlderletzte@holnishaus.de wrote:
Hello everybody, thank you all for the information. Unfortunately i do not fully understand, how to manage the shrinking. For now, i will install the new linux on a external disc, boot and then copy content from the f38 installation.
Am I correct in thinking you were able to shrink the btrfs filesystem, but are unable to find software to adjust the partition size to match?
You could use the new disk for a backup, then shrink the oversized partition. See: https://feldspaten.org/2018/06/01/resizing-btrfs-max/ -- shrink the btrfs filesystem a bit smaller than you want, use fstab to resize the partition, then expand btrfs to use the full partition.
https://itsfoss.com/btrfs/ lists the advantages of btrfs
Yes you are right. I could shrink the filesystem. My problem starts, because gparted complains a warning, that it is possible not able to boot. therefore i cancel the operation and read a bit deeper.
the old way, long and work intensive (finding the correct partition size, choose the filesystem typ and so on.). i think, it is annoying, that btrfs with f38 is so hard to use for this task. my new filesystem will not be btfrs.
All filesystems need backups. It seems wasteful to do an extra install.
-- George N. White III
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
I had a "dnf update" interrupted.
Running "dnf clean all" and then "dnf update" again results in:
Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction test error: file /usr/share/doc/glib2/NEWS from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/en_GB/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/pt_BR/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 ....... 160 lines of "... conflicts with file from package ..."
Running "dnf check" gives: NetworkManager-1:1.42.6-1.fc38.x86_64 is a duplicate with NetworkManager-1:1.42.8-1.fc38.x86_64 ... systemd-253.5-1.fc38.x86_64 is a duplicate with systemd-253.7-1.fc38.x86_64 ... systemd-udev-253.5-1.fc38.x86_64 is a duplicate with systemd-udev-253.7-1.fc38.x86_64 ... xxhash-libs-0.8.1-4.fc38.x86_64 is a duplicate with xxhash-libs-0.8.2-1.fc38.x86_64 Error: Check discovered 430 problem(s)
What is the problem (aside from the fact that I interrupted the update) and How can I recover?
Thanks.
On Mon, Jul 31, 2023 at 5:12 PM richard emberson emberson.rich@gmail.com wrote:
I had a "dnf update" interrupted.
Running "dnf clean all" and then "dnf update" again results in:
Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction test error: file /usr/share/doc/glib2/NEWS from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/en_GB/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/pt_BR/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 ....... 160 lines of "... conflicts with file from package ..."
Running "dnf check" gives: NetworkManager-1:1.42.6-1.fc38.x86_64 is a duplicate with NetworkManager-1:1.42.8-1.fc38.x86_64 ... systemd-253.5-1.fc38.x86_64 is a duplicate with systemd-253.7-1.fc38.x86_64 ... systemd-udev-253.5-1.fc38.x86_64 is a duplicate with systemd-udev-253.7-1.fc38.x86_64 ... xxhash-libs-0.8.1-4.fc38.x86_64 is a duplicate with xxhash-libs-0.8.2-1.fc38.x86_64 Error: Check discovered 430 problem(s)
What is the problem (aside from the fact that I interrupted the update) and How can I recover?
An equivalent for yum-complete-transaction was requested but never implemented. https://bugzilla.redhat.com/show_bug.cgi?id=1091702
There's probably a way to use `dnf repoquery` based commands to list the problem packages and them remove them.
Have you tried doing an update with --allowerasing?
Thanks, Richard
Running "dnf update --allowerasing results in:
Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction test error: file /usr/share/doc/glib2/NEWS from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/en_GB/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 ....... same 160 lines of "... conflicts with file from package ..."
The redhat https://bugzilla.redhat.com/show_bug.cgi?id=1091702 does not seem to provide enough information for solution plan; I'd really don't want to have to do a complete re-install; while I back a lot of stuff up on other machines, doing a new installation from a CD is my last resort.
On 7/31/23 15:41, Richard Shaw wrote:
On Mon, Jul 31, 2023 at 5:12 PM richard emberson <emberson.rich@gmail.com mailto:emberson.rich@gmail.com> wrote:
I had a "dnf update" interrupted. Running "dnf clean all" and then "dnf update" again results in: Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction test error: file /usr/share/doc/glib2/NEWS from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/en_GB/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/pt_BR/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 ....... 160 lines of "... conflicts with file from package ..." Running "dnf check" gives: NetworkManager-1:1.42.6-1.fc38.x86_64 is a duplicate with NetworkManager-1:1.42.8-1.fc38.x86_64 ... systemd-253.5-1.fc38.x86_64 is a duplicate with systemd-253.7-1.fc38.x86_64 ... systemd-udev-253.5-1.fc38.x86_64 is a duplicate with systemd-udev-253.7-1.fc38.x86_64 ... xxhash-libs-0.8.1-4.fc38.x86_64 is a duplicate with xxhash-libs-0.8.2-1.fc38.x86_64 Error: Check discovered 430 problem(s) What is the problem (aside from the fact that I interrupted the update) and How can I recover?An equivalent for yum-complete-transaction was requested but never implemented. https://bugzilla.redhat.com/show_bug.cgi?id=1091702 https://bugzilla.redhat.com/show_bug.cgi?id=1091702
There's probably a way to use `dnf repoquery` based commands to list the problem packages and them remove them.
Have you tried doing an update with --allowerasing?
Thanks, Richard
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
A new issue is best started in a separate thread instead of hijacking someone else's thread, as you did. It keeps threads neat and tidy too.
On Mon, 31 Jul 2023 15:11:06 -0700 richard emberson emberson.rich@gmail.com wrote:
I had a "dnf update" interrupted.
At what stage was it interrupted? If it was during update it isn't as problematic as if it is during cleanup, I think, when a package could be left in a partially finished state.
Running "dnf clean all" and then "dnf update" again results in:
Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction test error: file /usr/share/doc/glib2/NEWS from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/en_GB/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 file /usr/share/locale/pt_BR/LC_MESSAGES/glib20.mo from install of glib2-2.76.4-3.fc38.i686 conflicts with file from package glib2-2.76.3-1.fc38.x86_64 ....... 160 lines of "... conflicts with file from package ..."
Running "dnf check" gives: NetworkManager-1:1.42.6-1.fc38.x86_64 is a duplicate with NetworkManager-1:1.42.8-1.fc38.x86_64 ... systemd-253.5-1.fc38.x86_64 is a duplicate with systemd-253.7-1.fc38.x86_64 ... systemd-udev-253.5-1.fc38.x86_64 is a duplicate with systemd-udev-253.7-1.fc38.x86_64 ... xxhash-libs-0.8.1-4.fc38.x86_64 is a duplicate with xxhash-libs-0.8.2-1.fc38.x86_64 Error: Check discovered 430 problem(s)
What is the problem (aside from the fact that I interrupted the update) and How can I recover?
I think it was better not to run dnf clean all. You should have just run a repeat of dnf update because the system would have been in a sane state. By running clean all, you wiped all metadata.
I think the errors are because there is some cruft left over from the update. Try using dnf distrosync If that doesn't work, you can try dnf -x glib2 update to see if it will fix all the other problems and ignore glib2 for now. Usually, the problem that glib2 is reporting is a packaging error, but I suspect it isn't in this case because of the update interrupt. Once everything else is fixed, you might have to manually remove the glib2 packages in conflict because they are cruft left over from the interrupted update.
Hi
On Mon, 31 Jul 2023 15:11:06 -0700 richard emberson wrote:
What is the problem (aside from the fact that I interrupted the update) and How can I recover?
Try first to clean duplicates with: dnf removes --duplicates
I'll personally call it like that to see what would be done, answer no to the question, then call it with nohup and -y. Example:
nohup dnf -y removes --duplicates >& /var/tmp/rm-dups &
I do the same for "dnf update".
-
Francis.Montagnac＠inria.fr -
George N. White III -
Joe Zeff -
Karlderletzte -
KarlderLetzte
-
richard emberson -
Richard Shaw -
Roger Heflin -
stan