Re: koji spin-livecd issues
by Jay Greguske
They are bugs, and a patch was submitted upstream not long ago:
https://fedorahosted.org/koji/changeset?old_path=builder&old=3deb8358e67f...
You'll also want to move the broot.expire() later in the task too. There is a race condition where the output is getting cleaned up before it is moved to a final resting place. The patch I sited above fixes that too.
This code hasn't seen too much exposure, so don't hesitate to ask questions here or to me directly. I'm hanging out on Freenode IRC as jgreguske in #koji or #fedora-devel as well.
Thanks for checking in!
- Jay
----- "Doug Reiland" <dreiland(a)gmail.com> wrote:
> I got a minimal livecd to build (not tested yet) but a to make some
> kojid changes. Are these really bugs or should be handled elsewhere?
>
> I had to include '/dev/shm' in bind_opts
> Also, livecd-creator (imgcreator python package) was calling
> os.random() and that wanted '/dev/urandom', so I had to a mknod for
> that where loop? nodes are created.
>
> Also, I had to include device-mapper in the livecd-build group.
>
> Doug
> --
> buildsys mailing list
> buildsys(a)lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/buildsys
>
14 years, 1 month
Re: koji spin-livecd issues
by Jay Greguske
They certainly can because I'm doing that right now on my test system. Having topurl set is a requirement, but I see I neglected to put that in the wiki HowTo... I'll fix that now. I'm not sure why you're having difficulties with topdir though. Can you describe the error your were seeing and how you worked around it?
- Jay
----- "Doug Reiland" <dreiland(a)gmail.com> wrote:
> Sorry, one more thing.
> More of my initial errors was kojid needing topurl set in kojid.conf
> and it looks like kojid doesn't like topdir and topurl both set.
> Does this imply that no host can belong to both createrepo and livecd channel?
>
> Doug
>
>
> On Thu, Mar 11, 2010 at 10:38 AM, Jay Greguske <jgregusk(a)redhat.com> wrote:
> > They are bugs, and a patch was submitted upstream not long ago:
> >
> > https://fedorahosted.org/koji/changeset?old_path=builder&old=3deb8358e67f...
> >
> > You'll also want to move the broot.expire() later in the task too. There is
> > a race condition where the output is getting cleaned up before it is moved
> > to a final resting place. The patch I sited above fixes that too.
> >
> > This code hasn't seen too much exposure, so don't hesitate to ask questions
> > here or to me directly. I'm hanging out on Freenode IRC as jgreguske in
> > #koji or #fedora-devel as well.
> >
> > Thanks for checking in!
> >
> > - Jay
> >
> >
> > ----- "Doug Reiland" <dreiland(a)gmail.com> wrote:
> >> I got a minimal livecd to build (not tested yet) but a to make some
> >> kojid changes. Are these really bugs or should be handled elsewhere?
> >>
> >> I had to include '/dev/shm' in bind_opts
> >> Also, livecd-creator (imgcreator python package) was calling
> >> os.random() and that wanted '/dev/urandom', so I had to a mknod for
> >> that where loop? nodes are created.
> >>
> >> Also, I had to include device-mapper in the livecd-build group.
> >>
> >> Doug
> >> --
> >> buildsys mailing list
> >> buildsys(a)lists.fedoraproject.org
> >> https://admin.fedoraproject.org/mailman/listinfo/buildsys
> >>
> > --
> > buildsys mailing list
> > buildsys(a)lists.fedoraproject.org
> > https://admin.fedoraproject.org/mailman/listinfo/buildsys
> >
>
14 years, 1 month
Re: koji spin-livecd issues
by Jay Greguske
Yeah I've seen that as well, but it depends on when livecd-creator fails. Generally if it fails in a totally unexpected way (like /dev/urandom being missing) it won't clean up. But if it fails in an expected way, one that has been coded for, then it will.
I'll talk with the livecd-tools upstream maintainers to find out if they know more about the issue and are working on it. If active research/development isn't happening in that space I'll contribute a patch to Koji that forcibly gets it to clean up. It's not an easy problem though: I'm not aware of a good way to identify which mounts are specific to a given task. Do you have any ideas?
- Jay
----- "Doug Reiland" <dreiland(a)gmail.com> wrote:
> OK, thanks.
> FYI, I noticed the loop devices weren't being cleaned up when it
> fails. I noticed because I eventually ran out.
> I checked after my successful run and everything was OK there.
>
>
> On Thu, Mar 11, 2010 at 10:38 AM, Jay Greguske <jgregusk(a)redhat.com> wrote:
> > They are bugs, and a patch was submitted upstream not long ago:
> >
> > https://fedorahosted.org/koji/changeset?old_path=builder&old=3deb8358e67f...
> >
> > You'll also want to move the broot.expire() later in the task too. There is
> > a race condition where the output is getting cleaned up before it is moved
> > to a final resting place. The patch I sited above fixes that too.
> >
> > This code hasn't seen too much exposure, so don't hesitate to ask questions
> > here or to me directly. I'm hanging out on Freenode IRC as jgreguske in
> > #koji or #fedora-devel as well.
> >
> > Thanks for checking in!
> >
> > - Jay
> >
> >
> > ----- "Doug Reiland" <dreiland(a)gmail.com> wrote:
> >> I got a minimal livecd to build (not tested yet) but a to make some
> >> kojid changes. Are these really bugs or should be handled elsewhere?
> >>
> >> I had to include '/dev/shm' in bind_opts
> >> Also, livecd-creator (imgcreator python package) was calling
> >> os.random() and that wanted '/dev/urandom', so I had to a mknod for
> >> that where loop? nodes are created.
> >>
> >> Also, I had to include device-mapper in the livecd-build group.
> >>
> >> Doug
> >> --
> >> buildsys mailing list
> >> buildsys(a)lists.fedoraproject.org
> >> https://admin.fedoraproject.org/mailman/listinfo/buildsys
> >>
> > --
> > buildsys mailing list
> > buildsys(a)lists.fedoraproject.org
> > https://admin.fedoraproject.org/mailman/listinfo/buildsys
> >
>
14 years, 1 month
koji spin-livecd issues
by Doug Reiland
I got a minimal livecd to build (not tested yet) but a to make some
kojid changes. Are these really bugs or should be handled elsewhere?
I had to include '/dev/shm' in bind_opts
Also, livecd-creator (imgcreator python package) was calling
os.random() and that wanted '/dev/urandom', so I had to a mknod for
that where loop? nodes are created.
Also, I had to include device-mapper in the livecd-build group.
Doug
14 years, 1 month
[PATCH] Use tagHistory to grab all tags where the package is tagged
by Dennis Gregorovic
From: dennis <dennis(a)thinkpad.(none)>
---
cli/koji | 8 +++++++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/cli/koji b/cli/koji
index bd8a031..e58d79d 100755
--- a/cli/koji
+++ b/cli/koji
@@ -1870,7 +1870,13 @@ def anon_handle_latest_by_tag(options, session, args):
pathinfo = koji.PathInfo()
for pkg in args:
- tags = [x['name'] for x in session.listTags(package=pkg)]
+ hist = session.tagHistory(package=pkg)
+ tags = {}
+ for x in hist:
+ if not x['active']:
+ continue
+ tags[x['tag_name']] = True
+ tags = tags.keys()
tags.sort()
for tag in tags:
--
1.6.6.1
14 years, 1 month
koji/fedora update management
by Doug Reiland
Is there documentation on how koji can be used to manage
update/testing repositories?
Something like:
add-tag dist-X
add-pkg dist-X pkg1.....
build pkgs for dist-X
lock down dist-X
add-tag --parent dist-X dist-X-testing
add-tag --parent dist-X dist-X-updates
build pkgs for dist-X-testing as changes occur
as pkgs in dist-X-testing reach stable, tag-pkg dist-X-update pkg.
or pkgs in dist-X-testing reach stable, rebuild for dist-X-update
Is this strategy or BKM documented anywhere?
Thanks
14 years, 1 month
kojid - topurl, server's httpd configuration
by Doug Reiland
ok, I saw some old discussion of kojid's topurl setting instead of topdir
After playing with this a bit, I think it is working.
Let me first ask, what should topurl be set to?
So far, I have had to setup aliases in httpd/conf.d/kojiweb.conf for
the following:
Alias /packages/ /mnt/koji/packages/
Alias /remote-koji/ /mnt/koji/
In my remote builder, I have:
topurl=http://<server_name>/remote-koji
server=http://<sever_name>/kojihub
pkgurl=http://<server_name>/packages
Am I doing this correct?
It is getting quiet in this mailing list. I hope I haven't worn out my welcome!
I am really do appreciated the help.
Doug
14 years, 1 month
Proposal: move comps to fedorahosted git
by Bill Nottingham
I'd like to propose moving comps to fedorahosted git.
Why? Because CVS is a pain.
I can work on fixing the automated releng tasks that use comps.
What I'd like to know is if doing this at some point over the
next few weeks (say, post-Alpha) would be a problem for people.
If it is, we can push it off until after F13 ships.
Bill
14 years, 1 month
Bringing a new builder on line
by Doug Reiland
I am trying to configure a new builder.
First, my current setup is 1 system, for everything and using
localhost everywhere.
Now, I am trying to bring online another builder.
I had some fears about ssl stuff, but I just copied my /etc/pki/koji
from "hub" system to new system.
I get a add-host and starting kojid on new system and they communicated.
It appears that filesystem on hub needs to be mounted on each builder.
Is this correct?
Also, it appears that the pathname look the same of all systems. Is
this correct?
For example, on hub: my topdir is /koji/koji. It needs to look like
/koji/koji on all builders, correct?
Is there anything the builders2 shouldn't share, workdir, tmpdir, mock dir?
I kinda got that far and getting builder2 pulling tasks. Now the mock
setup is failing because it is trying to talk to yum repostory via
http://localhost
I have not set anything to localhost on builder2 so it must be pulling
this some the shared data somehow.
Am I doing this right?
I am going to have to redo the hub/server again and don't use localhost
Thanks in advance
14 years, 1 month
