Hi
do you use tumbleweed by any chance ? i had a problem : no more ext4 fs known, wrong kernel ( .37) when i was using .39 and also a bad eror on /bin/sh not recognized … is this regular or expected with tumbleweed ?
the only fact i can notice is that i turned off my external hdd while the pc was being shut down , can this disturb so much ? (??) also modprobe was not able to work anymore …
Do you regard tumbleweed as experimental ? is this supposed to be so far ? i thought it was not .
thanks in advance
so i reboot my pc and i had a black screen with qwerty only (a pain when your keyboard is azerty) :
how do i get back azerty ?
Then what are my best chances to correct these errors ? (modprobe, ext4, /bin/sh not working)
Do you regard tumbleweed as experimental ? is this supposed to be so far ? i thought it was not .
I didn’t use it before 11.4 but it has behaved very stably since the advent of 11.4. The only Tumbleweed-specific major problems I have seen are dependency issues when users don’t understand the interactions of the packages in the various repos. I had a few problems like that until I learned to pay close attention to the dialogue with zypper.
It looks like you’ve had a full or part regression from Tumbleweed distro to openSUSE distro (kernel 37 is the original openSUSE kernel). So you need to upgrade properly if you can arrange it. You can upgrade from the CLI. What repos are showing when you run “zypper lr -u” ?
in fact i was sutck with a qwerty console and and azerty keyboard and no time available to solve this or research a cause, so i erased it all and reinstalled 11.4 . I’m thus trying to get what happened.
How come i t came back to the original kernels ? plus ext4 and /bin/sh and modprobe ko; this seems a lot for a stable version :’(
I’m not the only one who had this kernel being jeopardized. Also yast update factory seems to mess up kernels (not my machine but another had this problem when it added 2 different kernels).
Giving the Tumbleweed repos higher priority than the others can help avoid the kernel issue
Not clear if you have this from your post, but I wouldn’t think using both factory and tumbleweed repos would be a good idea (but as always I could be wrong)
i had a default 11.4 and this is what i used to upgrade 11.4 to tumbleweed : opensuse:tumbleweed [Linuxpédia]
it’s not in english but the cli commands are understandable.
Using tumbleweed from shortly after openSUSE-11.4 was released, this is stable. What I am unsure about:
Do we have automatic tests if a dist-upgrade to tumbleweed is possible without errors now at a more advanced state of tumbleweed (udev!)?
I did a quick skim thru that and it had a whole bunch of extra repositories I would never add: ie tumbleweed-testing !! (DO NOT ADD THAT), banshee, virtualisation, games, games-unknown, games-unknown-horizon.
But the others after tumbleweed-testing are a mute point. Stick with tumbleweed but not tumbleweed testing.
Plus, with the basic tumbleweed, its not clear to me that article provides good guidance for the priority of the tumbleweed repos which is important.
Tumbleweed from the beginning on 11.4 here too. Very stable. The only issues I have/had concern the conflicting networkmanager versions of KDE and GNOME 3, which I both have installed. But that’s a known issue and being worked upon.
i only had the 11.4 default repos + drivers for webcams and also packman for tumbleweed plus the tumbleweed main repo .
In fact i just used the part of the tutorial that described how to upgrade to tumbleweed ; and then i used # zypper ref && zypper up (from init 3) or yast to upgrade packages.
Never added all the rest of these repos because i wanted to see 1st of all how tumbleweed is reacting for a while. I was expecting it to be ok untill 12.1 though.
Btw the tutorial clearly mentions that Tumbleweed testing is risky.
Is turning off an external hdd by myself (manually with the off button) while system is being shut down wrong ? And can it thus impede the system to modprobe something (?) (1st error was modprobe)
The link gives bad advice. You should not use it if you upgrade again. The CLI commands there will cause dependency errors. And it will be worse if you have the wrong repos as well as the wrong commands. Almost certainly that’s what dragged your kernel back to zero (openSUSE) and broke your system. Try the link in my sig. But (as oldcpu suggests) first publish your repositories if you try again.
Is turning off an external hdd by myself (manually with the off button) while system is being shut down wrong ? And can it thus impede the system to modprobe something (?) (1st error was modprobe)
It’s too hard to separate this process from the “zypper up” process. But – generally it’s a bad idea to unmount a USB drive by cutting the power.
On balance, I thing the “zypper up” command probably did the damage, but there should have been clear warning of the downgrade before it proceeded, warning in the dialogue in the CLI. But we can’t be very wise in diagnosis after the drive was formatted. Good luck in the future.
If a user has the right repos, there shouldn’t be a problem like this one, there shouldn’t be a need to prioritise. Tumbleweed is designed and tested on having the right repos and with uniform priorities. Of course, if there are left-field repos added in, that takes the process over the edge of stability and into the experimental zone (where priorities might well be needed).
You’re not wrong. But advanced users would be able to do that, understanding the dangers. (and likely there would be a need for repos to have differing priorities in that case).
Its very difficult to know what you did when you stated you followed the guide, then when I point out the problem there, you come back and state, no, you only followed part of the guide. I know you must be frustrated and that you want to save time and solve this quicker by simply showing a reference and NOT stating what you did, but it just slows things down for everyone. Stating you followed the reference, when you only followed part of the reference, only serves to confuse those who were not leaning over your shoulder when you did this. … I know, I know … I should be used to this by now as many users do what you just did, but it is tiring to have one’s efforts at help non-time efficient.
When you set the priority for tumbleweed, what number did you assign it ?
If it is not yet unmounted - Yes, its wrong.
IMHO the best way to do this is to 1st unmount the drive, then unplug the drives cable (usb or firewire) and then switch OFF the drive, and then shut down the PC.
Don’t know if you remember or not swerdna but you and I were posting on a thread about a similar issue recently that giving the tumbleweed repos higher priority fixed
It also occurs to me that maybe manchette_fr isn’t aware of swernda’s page on how to switch over to tumbleweed, if that’s the case manchette_fr you may have more succes using swerdna’s instructions, you can find them here:
I want to try the new kdePim-4.6 packages, where i have highest priority. And I want to use all fitting media codecs, which is also highest!
When there is something missing of kde-4.6 I have a lower prioritiesed Kde-4.6 Release repo in it. Also google-chrome gets installed from google repo even it is lowest priority because it is the only repo having the package google-chrome!
This way you can ever and only:
zypper ref
zypper dup -d
zypper -vv dup
At first refreshing the repos, I don’t like automatics
At second only downloading, in case your internet breaks away
At last look carefully what you are doing before install of packages start
