Hi,
I am trying to build a package for CentOS that requires some packages
from epel + the developer toolset to be built.
I added the repositories in the project meta configuration - which looks
like this:
<repository name="CentOS_7">
<path project="security:zeek" repository="CentOS_7"/>
<path project="Fedora:EPEL:7" repository="standard"/>
<path project="CentOS:CentOS-7" repository="scl-rhscl"/>
<path project="CentOS:CentOS-7" repository="standard"/>
<arch>x86_64</arch>
</repository>
The spec file then require some packages from the repository,
speficically:
BuildRequires: epel-release kernel-devel cmake3 devtoolset-7
After trying this, the build has been blocked with the status
“downloading 6 dod packages”.
Could anyone point me to what I am doing wrong?
Link to build:
https://build.opensuse.org/package/show/home:0xxon:branches:security:zeek-c…
Link to project meta:
https://build.opensuse.org/projects/home:0xxon:branches:security:zeek-cento…
Thanks a lot,
Johanna
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Hi all,
last weeks my instance (2.10) had strange problems, often everything was slow / sluggish but the OBS server was more or
less idle (I have thrown enough hardware at the problem I guess, 16GB of RAM and 10 VCPUs on a not totally up-to-date
intel xeon x5670).
Even the webfrontend was not responding and "osc st" for example took many minutes to eventually finish.
Investigating, I found lots of "rpc timeout" messages in all services log files.
Just restarting some of them (src server, scheduler) seemed to fix the issue often, but then it reappeard a day or a
week later.
Finally I just rebooted the whole server, now everything is fine again.
I also enabled $logtimeout =1 in BSRPC.pm, even before rebooting.
It looked like communication with bs_srcserver was the one that timed out, and from my hopeless tries in stracing the
issue I'd guess some lock around the lastevents file was the problem.
I think that this was caused by restarting services in the wrong order, which was finally "fixed" by rebooting the server.
Is there a recommended way to restart all services without rebooting the server?
Or is a reboot the simplest way to achieve that ;-)
Thanks,
seife
--
Stefan Seyfried
"For a successful technology, reality must take precedence over
public relations, for nature cannot be fooled." -- Richard Feynman
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
This is just to raise the question to simply initiate a discussion, why not have AlpineLinux on OBS too?
On the positive side, it should be relatively easy to adapt since it is really rather similar to Arch, which is already supported on OBS, but simpler.
On the other hand, unlike Arch, it also has 6 month releases that are supported for 2 years. I do appreciate those would accumulate over time. It also has a lot more architectures supported. So I do appreciate that it could also become a maintenance time sink in OBS.
Also, I have no sense of the real size of the AlpineLinux using community, and I suspect most use it from the docker and container perspective, which is a valid and potential community that could benefit from OBS. In my particular case, I do use Alpine very broadly, and specifically for headless arm/IoT devices.--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Dear OBS Maintainers,
I am running a deployment of OBS on a Kubernetes cluster. In
particular, I am running a horizontal pod autoscaler, that creates
more worker pods as-needed and removes them when they have been idle
for long enough. This works reasonably well, but leaves a lot of
'dead' workers lingering in the build status page, since Kubernetes
assigns new unique pod names every time a pod is created. Is there a
way to configure OBS to assume a worker is not coming back if it has
not been reachable some time?
Best regards,
Moritz
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Dear all,
I would like to use osc with use_keyring without a windowmanager i.e. on
a server without X/wayland.
Is this possible as according to the documentation [1]. use_keyring only
supports (For KDE the KWallet is used, for GNOME it is Seahorse.).
I've currently have a windowmanager installed on my server VM but I
don't want this anymore as it makes the server slower, uses more
resources and uses more power.
Regards,
Joop.
[1]
https://openbuildservice.org/help/manuals/obs-user-guide/cha.obs.osc.html
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Hi,
we had a serious regression in spec file parser of OBS
since yesterday.
We need therefore a cold start of the schedulers to get into
guaranteed clean state again.
sorry
adrian
--
Adrian Schroeter <adrian(a)suse.de>
Build Infrastructure Project Manager
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany
(HRB 247165, AG München), Geschäftsführer: Felix Imendörffer
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
On Freitag, 31. Januar 2020, 12:27:42 CET Aleksa Sarai wrote:
> On 2020-01-31, Adrian Schröter <adrian(a)suse.de> wrote:
> > we had a serious regression in spec file parser of OBS
> > since yesterday.
> >
> > We need therefore a cold start of the schedulers to get into
> > guaranteed clean state again.
>
> I hate to add more fuel to the fire -- but it looks like OBS is down
> now? I get "Sorry, our script crashed." errors after a really long
> timeout.
The storage was not able to handle the load of all the scheduler
cold starts...
We do it incremental now.
--
Adrian Schroeter <adrian(a)suse.de>
Build Infrastructure Project Manager
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany
(HRB 247165, AG München), Geschäftsführer: Felix Imendörffer
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Hi all,
Apologies if this is the wrong place to send.
I've been having an issue where packages built in OBS suddenly depend on
python3dist(bs4) instead of python3-beautifulsoup4 or another package
that exists, does anyone know why this might be? I assume it's something
wrong in my spec file. This never used to happen though, and now it
happens for both Fedora 30 and 31.
Can anyone give me any advice?
Thanks,
Hamish McIntyre-Bhatty
Hi
did something change in the Centos8 Repositories used by the OBS?
Just checked and almost every package complains about not providing
libperl.so.5.26 and perl-Encode(x86-64). Howevery none of these packages
has these dependencies.
Regards
Mathias
--
Mathias Radtke
Save the Date: opsi-Konferenz
2 Tage Workshops, Konferenz und Best Practice bei der
opsiconf in Mainz am 2./3.März 2020
weitere Infos und Anmeldung unter opsiconf.org
opsi - Open PC-Server-Integration
das Open Source Client-Management-System von uib gmbh
Birgit Hubal
Vertrieb / Projektkoordination
eMail: b.hubal(a)uib.de
Tel. +49 6131 / 27561-18
Fax +49 6131 / 27561-22
uib gmbh
Bonifaziusplatz 1B
55118 Mainz
Internet:
https://uib.dehttps://opsi.org
Geschäftsführer: Dr. Detlef Oertel, Erol Ülükmen
Handelsregister: Amtsgericht Mainz HRB 6942
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org
Hi,
I am a bit new to osc and OBS and for now I'm just building RPMs
locally for openSUSE TW using 'osc build'
Under my home:<account> folder, I have 2 packages: libfoo and foo.
I created a .spec for libfoo and have it built successfully with 'osc
build'.
Now I want to build foo and made a .spec for it. It depends on libfoo
and the
.spec references it but I can't understand how to have the RPMS
generated for libfoo
to be installed in the chroot for foo which are necessary for it to compile.
I can manually install libfoo RPMs with 'osc shell' but they are removed
when
building foo (and --noinit does not help).
My guess is that this approach is wrong.
So what is the proper workflow to build dependent packages locally ?
Eventually I plan to have it built by OBS but I would like to avoid it
without
having stuff working locally first (if possible)...
--
To unsubscribe, e-mail: opensuse-buildservice+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-buildservice+owner(a)opensuse.org