Forum Replies Created
-
AuthorPosts
-
einar.hjortdal
Participanthmm what I pasted above comes from that “Console output” box. To me it seems like it lacks standard error, but maybe I am wrong.
I mean, this doesn’t tell me what the error is, so I can’t fix that.
LD [M] virt/lib/irqbypass.ko make: *** [Makefile:224: __sub-make] Error 2 error: Bad exit status from /var/tmp/autodist/rpm-tmp.mcuZwR (%build)I looked in
/var/webbuild/tmpbut it doesn’t contain any relevant information.
I’m currently building the kernel without any change. Likely there is a missing depeinar.hjortdal
ParticipantLD [M] virt/lib/irqbypass.ko make: *** [Makefile:224: __sub-make] Error 2 error: Bad exit status from /var/tmp/autodist/rpm-tmp.mcuZwR (%build) RPM build errors: Bad exit status from /var/tmp/autodist/rpm-tmp.mcuZwR (%build) ERROR: rpmbuild exited with error code: '1' # WEBBUILD_EXITCODE=1I am getting some errors in the console output, but they don’t contain much information. Where do I find the logs?
einar.hjortdal
ParticipantThe issue is not a real issue, it was fixed by installing flex manually in the container.
I am not allowed to modify the message so I must reply here.einar.hjortdal
ParticipantI have manually edited the /etc/resolv.conf inside the container and the issues were resolved.
I have now a new issue here: https://pastebin.com/x8kaGuuV
It looks like it cannot find the required packages to build.P.S. I couldn’t submit more information before, somehow the forum limited me
-
This reply was modified 1 month, 3 weeks ago by
einar.hjortdal.
Attachments:
You must be logged in to view attached files.einar.hjortdal
ParticipantThe docker host is openmamba. You were right, restarting the daemon seemed to fix the issue. But this fix was short-lived, after loading the list of packages and attempting an “unpack and edit”, this issue reappears.
[root@buildvm /]# curl https://push.openmamba.org/pub/openmamba/base/SRPMS.base/ curl: (6) Could not resolve host: push.openmamba.org (Timeout while contacting DNS servers)But it does resolve the google domain.
[root@buildvm /]# curl google.com <HTML><HEAD><meta http-equiv="content-type" content="text/html;charset=utf-8"> <TITLE>301 Moved</TITLE></HEAD><BODY> <H1>301 Moved</H1> The document has moved <A HREF="http://www.google.com/">here</A>. </BODY></HTML>I am suspecting something is up with resolvectl and docker
-
This reply was modified 1 month, 3 weeks ago by
Silvan.
-
This reply was modified 1 month, 3 weeks ago by
einar.hjortdal.
Attachments:
You must be logged in to view attached files.einar.hjortdal
Participant[root@buildvm /]# curl https://push.openmamba.org/pub/openmamba/base/SRPMS.base/
curl: (6) Could not resolve host: push.openmamba.org (Timeout while contacting DNS servers)You guessed right
einar.hjortdal
ParticipantHi thank you for the reply. I am experienced with rpmbuild. I imagine the /var/autodist/RPM is the usual rpmbuild directory
The build environment is selected just like in your screenshot, but I have no spec to select. I have F5’d but no changes.
Attachments:
You must be logged in to view attached files.einar.hjortdal
ParticipantI have the container running on my system and the web interface open in my browser.
What I’d like to do is to add a patch the the kernel and get it to package it.
At the top of the web interface I select the base repository, and the SRPMs select appears, but it has no options. What should I do?
I also have spawned a shell in the running container according to the instructions, I suspect I have to setup something else in it through it.
I have also forked the kernel repository on the gitea, to version my changes.
einar.hjortdal
Participantdiscoveryis a strict requirement of the KDE Plasma desktop environment as it is provided by openmamba so it is by design a requirement of thedesktop-base-kdemeta package.This is what I wanted clarity on.
On other distributions I’ve used previously, discovery is not a strict requirement for plasma.
If this is how it is for openmamba then it is how it is, despite how unnecessary it may seem to me.einar.hjortdal
ParticipantIs it normal for desktop-base-kde and kwin-x11 to be removed when a user tries to remove discovery? Shouldn’t those packages be independent?
einar.hjortdal
ParticipantI have reinstalled openmamba and I think I understand what is happening:
After installing I tried to remove discover withsudo dnf remove discover$ sudo dnf remove discover
Dependencies resolved.
================================================================================================================================================================================
Package Architecture Version Repository Size
================================================================================================================================================================================
Removing:
discover x86_64 6.4.0-1mamba @rolling 6.9 M
Removing dependent packages:
desktop-base-kde x86_64 6.0-7mamba @rolling 35 k
discover-notifier x86_64 6.4.0-1mamba @rolling 341 k
mambatray x86_64 5.3-1mamba @System 51 k
Removing unused dependencies:
kwin-x11 x86_64 6.4.0-1mamba @rolling 12 M
libkwin-x11 x86_64 6.4.0-1mamba @rolling 12 MTransaction Summary
================================================================================================================================================================================
Remove 6 PackagesFreed space: 32 M
Is this ok [y/N]: n
Operation aborted.I had
discoveruninstalled before the plasma 6.4 update.-
This reply was modified 5 months ago by
einar.hjortdal.
June 3, 2025 at 3:30 pm in reply to: Kernel 6.12.30 upgrade then stuck at `starting openmamba` #29355einar.hjortdal
ParticipantI think the recent merges fixed the issue
June 3, 2025 at 12:06 pm in reply to: Kernel 6.12.30 upgrade then stuck at `starting openmamba` #29353einar.hjortdal
ParticipantIt must be the nvidia_470 package. I’ve uninstalled I can successfully start kernel 6.12.30
However, if I install nvidia_470 while running kernel 6.12.30, the problem reappears on the next reboot.June 3, 2025 at 10:07 am in reply to: Kernel 6.12.30 upgrade then stuck at `starting openmamba` #29352einar.hjortdal
ParticipantHi, does access to any virtual console work by pressing CTRL-ALT-F2..F8? This would allow retrieving log information from the console to identify the cause of the issue. If not, the only way to get useful information would be to start the kernel with the “debug” menu choice (or removing “quiet splash” from Grub command line). Did you by any chance install the “nvidia” proprietary driver?
CTRL-ALT-F2..F8 do not seem to work. Starting the kernel with debug messages doesn’t show any useful messages, the only line shown is the
efi stub loaded initrd.I do have the
nvidia_470package installed on 6.12.20 and it still seems to work when booted with this kernel. I had my own suspicion that the cause of the issue is that, but I have no proof.einar.hjortdal
Participantopenmamba is in fact a single maintainer driven distribution
I had no idea. Damn! You’re doing an incredible job.
The same may be applied to .spec files
This is what I had in mind, I had contributed to OpenMandriva a couple years ago and they accepted .spec files and relative patches. Their build system builds the package by pulling sources and compiling when necessary, then puts it on the repository. No uploading of binaries.
A contributor might want to start learning and creating packages for openmamba in his local environment and if you express that you have this interest the discussion might evolve into discussing the technical details on how to do this.
I would like that. And if necessary, I may host my own user rpm repository, with packages built according to openmamba standards.
-
This reply was modified 1 month, 3 weeks ago by
-
AuthorPosts

