+1
___
infrastructure mailing list
infrastructure@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/infrastructure@lists.fedoraproject.org
+1.
I have configured rabbitmq to disable logging but it keeps going on..
___
infrastructure mailing list
infrastructure@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/infrastructure@lists.fedoraproject.org
> Greetings.
>
> To fix issue https://pagure.io/fedora-infrastructure/issue/5502
> we need to allow .htaccess files "FileInfo" privs in
> the /srv/web/getfedora.org dir on the proxies so the redirects work.
>
> +1s?
>
> From 5625ce807a7ffd294b145b0288846a7f75a42821 Mon Sep 17 00:00:00 2001
> F
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Looks reasonable and small enough to me.
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Looks sane to me. +1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
> dist-git is currently different infrastructure, I'm not even sure if
> pague.io is hosted in the same DC/network.
It is not. Pagure.io is hosted in OSUOSL, while dist-git and all build infra is
in PHX2.
>
> I like the root src.fp.o/ idea, adding the extra pagure/ just seems
> like extra chara
> On 08/01/2017 09:13 PM, Matt Prahl wrote:
>
> I'd prefer stage and prod to be as close as possible to identical. Also
> people have been writing patches with the assumption that it'll be
> src.fedoraproject.org/pagure/[0]. It's not hard-coded, but it is the
> setting default if unspecified.
The
> I'm confused about something. Can any postfix experts help debug?
>
> On Friday I put a new package-owner-alias cronjob in place on bastion
> (that Matt Prahl wrote). It generates the package owner list from
> pagure over dist-git instead of generating it from pkgdb.
>
>
> https://infrast
> A few fixes for ostree commit signing for f27/rawhide.
>
> [PATCH 1/3] robosig: add in commit signing for f27
> [PATCH 2/3] robosig: sign ostree rawhide with f28 key
> [PATCH 3/3] robosig: add in ostree signing for rawhide ppc64le
These patches look correct to me.
+1
___
> https://pagure.io/atomic-wg/issue/336
>
>
> Each ref only has one commit in it now:
For the record, this is now fixed.
Both the underlying cause and the results.
This was caused by me forgetting to push a fix a while ago that means that
robosignatory-signatomic overwrote the ref.
In the futur
You are correct, updated patch:
commit 9c67c7019b4ea370548ebab4f408b46dc214f243 (HEAD -> master)
Author: Patrick Uiterwijk
Date: Wed Sep 27 17:14:07 2017 +0200
Redirect old docs urls to docs-old
Signed-off-by: Patrick Uiterwijk
diff --git a/roles/fedora-docs/proxy/files/fedora-d
> +1, default timeout cant be 12 hours.
It totally is, and for good reason.
There *are* API calls that take literally hours.
For example, the "Finish and import build" call that koji does as part of a
build step for texlive (a package with lots of subrpms and big subrpms) takes
about 6 hours at
> I have updated pungi on compose-x86-01 compose-x86-02 and branched-
> composer to 4.1.19
>
> the full transaction on all boxes was
>
> [root@branched-composer ~][PROD]# yum --enablerepo updates-testing
...
>
> Complete!
>
> We can downgrade if they have problems. The reason I went and update
> Sure, +1.
>
> It might be good to disable the sync script entirely at first so we can
> examine the output and make sure it looks ok before syncing it anywhere.
The new code will generate it somewhere else (in /mnt/koji/compose/updates/
instead of /mnt/koji/mash/updates/), so that will happen
> Hello,
>
> On Fri, Oct 27, 2017 at 5:27 AM, Chenxiong Qi
>
>
> We will probably need someone else here as well as I do not have permission
> to run pkgs playbook, I think. Otherwise the new updated dist-git package
> (1.4-3)
> should already be in epel7-infra-stg repositories.
FBR filed as
> On Sat, Oct 28, 2017 at 01:41:32AM +0200, Patrick Uiterwijk wrote:
>
> +1 for me.
Thanks
>
> Do you have the link handy for the path of the namespace support in dist-git?
Yep:
https://src.fedoraproject.org/rpms/dist-git/blob/master/f/ensure_gitrepo_namespaced.patch
>
>
> Thanks,
> Pierre
> Greetings.
>
> In order to allow the new rich deps that rust packaging requires we need
> to update rpm on the builders to a version that has backported support
> for 3 new rich deps. It needs these at srpm generation time to
> understand the new rich deps.
>
> ignatenkobrain has created a rpm
> Good Morning Everyone,
>
> There was a report that releases.pagure.org doesn't serve .msi binaries
> properly.
>
> I believe this diff should fix it:
>
> roles/pagure/frontend/templates/0_pagure.conf | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/roles/pagure/frontend/templates/0
> We need
> https://pagure.io/pungi/c/c89f0334579aee68286d07d4b57f4df67510166c?branch...
> patch to create live images for arm in modularity, which is included in Pungi
> 4.1.20
> release. So, I would like to update the branched composer and try it out. And
> if possible
> include it in F27 Beta
> If we happen to miss a sync for whatever reason there could be
> gaps in the ostree history. Adding --depth=-1 will make sure that
> the entire history for the ref is in the target repo.
> ---
> roles/bodhi2/backend/files/new-updates-sync | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> ---
> roles/bodhi2/backend/files/new-updates-sync | 11 ++-
> 1 file changed, 10 insertions(+), 1 deletion(-)
>
> diff --git a/roles/bodhi2/backend/files/new-updates-sync
> b/roles/bodhi2/backend/files/new-updates-sync
> index a241eeb..35c9d58 100755
> --- a/roles/bodhi2/backend/files/n
> The diff looks ugly but all I did was copy the updates
> entries and s/updates/testing/ and then I deleted the one
> existing testing entry that was below f27-workstation.
> ---
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.o
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
> For Fedora Atomic we release content every two weeks. A lot of ducks
> have to be in a row in order to do this and often times behind the
> scenes work is time sensitive. I really need access to releng/infra
> so that I can be more efficient in doing this. I'd like to be able to:
Okay, here we g
> Sync out the stable ref too, which is just aliased to
> the updates ref at this point.
> ---
> roles/bodhi2/backend/files/new-updates-sync | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/roles/bodhi2/backend/files/new-updates-sync
> b/roles/bodhi2/backend/files/new-updates-sync
> in
Okay, let's do this thing.
Can you please push this change to ansible?
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
I asked Dusty to push a patch to fix the f26 and f27 dest directories.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
> On Tue, Mar 6, 2018 at 11:34 PM, Patrick Uiterwijk
>
> Comment inline below.
>
>
> I think aarch64 should be up with the ['x86_64', 'armhfp',
> 'source'] list now
Right, imagine that I've done that. (lines of patch following).
+'to': [{'arches': ['x86_64', 'aarch64'
Hi,
Pungi 4.1.22 moved from py2 to py3 but is needed for modular update composes,
so can I get +1s to update pungi and move the deps over to their py3 versions?
These deps are not deps of the main package because they're only needed when
running module updates.
I think the biggest chance of iss
This FBR is retracted.
The issue is not fixed in the latest pungi, and will need emergency Bodhi
patches instead, together with other fixes.
So, instead, I'll be sending out an FBR for a Bodhi update soon.
___
infrastructure mailing list -- infrastructur
For F28+ modular, the modules are signed by the keys of the underlying platform
rather than the fedora-modular key.
+1s for this patch to set the correct key?
diff --git a/roles/bodhi2/backend/templates/pungi.module.conf.j2
b/roles/bodhi2/backend/templates/pungi.module.conf.j2
index 77eceb500..
> Hi,
>
> The current threshold are a little bit low and we receive warning when
> there are a lot of package updates. I have already pushed the commit
> thinking that this was impacting only packages hosts which is not the
> case :)
>
> Looking for retroactive +1 ?
>
> Thanks
> Clément
>
+1 T
> We were having issues with modular rpms ending up in Everything repo and also
> missing
> Container variant, which should be fixed by pungi-4.1.22-7.fc27 update.
>
> koji build: https://koji.fedoraproject.org/koji/buildinfo?buildID=1055890
+1
>
> Fixes:
> https://pagure.io/pungi/issue/862
>
Hi all,
So, in ansible commit 2cce3883a34e5366e822fd581455f63344630c8c, Kevin disabled
the graphite plugin for collectd.
Unfortunately, at the same time, he removed the restart handler for collectd,
which means that all of our boxes are still running with the old config, slowly
filling up their
> Greetings.
>
> It was reported to me that a number of new packages were not showing up
> in grokmirror on src.fedoraproject.org. On looking I noted that new
> packages did not have the chained post-receive hook we normally put in
> place. The cause seems to be a typo in the command.
>
> I'd lik
> The f28 repos we need to tap into are not in the same locations
> as for post release (f26/f27) runs. We also don't need to include
> the 'updates' repo because there is no updates repo for f28 yet.
>
This is actually an f28 repo, but I had made a mistake moving things around (I
accidentally m
> On 03/09/2018 04:25 PM, Patrick マルタインアンドレアス Uiterwijk wrote:
>
>
> ok I'm using a pull request for this:
>
> https://gitlab.com/dustymabe/fedora-infra-ansible/merge_requests/1/diffs?w=1
>
> here is the updated patch if you want to fight with pulling it in f
Hi all,
The filenames for atomic workstation treefiles changed from -ostree- to
-atomic-.
Can I get +1s?
Patrick
diff --git a/roles/bodhi2/backend/templates/pungi.rpm.conf.j2
b/roles/bodhi2/backend/templates/pungi.rpm.conf.j2
index ff31d400e..1561360d6 100644
--- a/roles/bodhi2/backend/templa
> Hi,
>
> After the last change for updating fedimg on stg. The config key in
> the prod also broke. To fix this issue I would like to apply the
> following patch so the it can listen to pungi messages.
>
> ---
> roles/fedimg/templates/fedmsg.d/fedimg.py | 4 +---
> 1 file changed, 1 insertion(+
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1, though I thought we'd set fedmsg CRL as "content", meaning it wouldn't fall
under freeze.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Hi,
Last Friday, pungi got updated on compose-x86-01, but that requires
fedmsg-1.1.0 (there's currently 1.0.1 on that box) because of json encoding
issues.
+1 to run "dnf update --refresh fedmsg" on compose-x86-01?
Patrick
___
infrastructure mailing l
Looks reasonably self-contained to me and shouldn't break release stuff, so
that's fine, +1 on the change.
I just think it would be a good idea to inform the maintainers of these
packages of this?
___
infrastructure mailing list -- infrastructure@lists
Sure, +1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Sure, +1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
> This is an attempt to fix https://pagure.io/dusty/failed-composes/issue/176
>
> I'd like to get patrick and one other person to review these changes.
Sure, the changes look fine and easily revertable.
Also, all these artifacts are failable.
So +1.
___
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
No, because the logger.info statements also output things like what the current
state is, and we explicitly set logging to ERROR to only catch the errors.
I want these specific output messages to print whatever the log level is set to.
___
infrastructure
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Looks reasonable to me, and easy to revert.
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Hi,
So, for now (until https://github.com/fedora-infra/bodhi/pull/1986 gets
finished and merged), it might happen that Bodhi stages a repo a second time
(if it crashed while waiting for sync), which new-updates-sync doesn't notice,
meaning releng needs to manually clean up the state file.
This
Hi,
I have had to update libmodulemd on bodhi-backend01, because it requires the
"modulemd_objects_from_file" function in libmodulemd, which was added in
version 1.2, but it is lacking a (versioned) requirement on libmodulemd, and
due to the freeze we still had modulemd v1.1.
This should not ha
So, I guess that the current patch will work just fine, so +1 on that, but it
does make updates out of sync with updates-testing.
Before changing these paths, we need to ask adamw whether this will break his
tooling, and possibly coordinate with him.
__
+1 looks fine to me.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines:
Let's give this a try.. +1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guide
Hi all,
Can I get +1s for the following DNS repo change?
This will make it simpler to debug issues with the candidate-registry from OSBS.
Patrick
commit 103f8de1f81264568d74e256ea45588d553ddd31
Author: Patrick Uiterwijk
Date: Tue Sep 25 18:52:47 2018 +
Only use proxy101 for internal
> +1 for temp fix. Do we need to make a host proxy111 which is an A
> record of proxy101 and proxy110 so we can do failover?
> On Tue, 25 Sep 2018 at 14:55, Patrick マルタインアンドレアス Uiterwijk
> https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproj
Can I get +1s for the following patch?
Explanation as to why this specific patch is in the commit message.
commit d2688610419973cd519ef8f8f581a03215aaf73c (HEAD -> master)
Author: Patrick Uiterwijk
Date: Tue Sep 25 21:25:00 2018 +0200
Only do OCSP stapling on the proxies
The act
Hi all,
Can I get +1s for the patch underneath?
This adds proxy101/proxy110 to the whitelist for OSBS for candidate-registry
pushing.
Patrick
commit ef9fa10389a6a60843c96e4855dbdcb604583baa (HEAD -> master)
Author: Patrick Uiterwijk
Date: Tue Sep 25 21:34:39 2018 +0200
Whitelist proxy1
> No, I am hoping that once I figure out what's going on, we can drop this temp
> workaround.
And the root cause is found and fixed now:
https://lists.fedoraproject.org/archives/list/infrastructure@lists.fedoraproject.org/thread/7IGK2QIBT3JU3GD77UDDM5VZ4TWKHNQ4/
__
Hi,
Can I get +1s to the following patch?
This would enable keyhelper.py on pagure.io, which is an alternative to
authorized_keys file, which need a gitolite recompile to get changes activated.
If anything goes wrong with it, sshd will fall back to the authorized_keys
file, which will still be u
+1 this should be simple enough to back out if it fails.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/c
Hi,
This is the next step of preparing pagure.io for repoSpanner: switch the entry
command to aclchecker, which is a small script that calls either repoBridge for
repos on repoSpanner (none at this moment) or gitolite for all others.
This adds some configuration for repobridge, but that won't be
(Actually, until a future PR where I apply the correct sshd_Config to pagure01,
it won't even change anything on prod for now other than the existance of these
config options.)
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To
Hi all,
Can I get +1s for these last two patches, which should make Pagure.io fully
repoSpanner compatible?
It switches the prod sshd to the new sshd_config with keyhelper, and the other
patch disables the scriptalias, which means it hits the code in Pagure that
multiplexes this to gitolite (fo
Hi all,
This should be my last FBR for now, as with this I tested everything in
staging, and I believe it works there (SSH pull/push and HTTPS pull, we don't
have HTTPS push for pagure.io (yet)).
This basically sets the username for anonymous access to something that would
never be a valid user
ack, +1.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fed
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedorapro
Hi all,
Can I get +1s to make Nagios check the ticketkey age and to make it stop email
me hourly?
Thanks,
Patrick
commit 3fc57e699b9c524d5683fcc39231f7b1f425bbb9 (HEAD -> master)
Author: Patrick Uiterwijk
Date: Sat Oct 20 15:36:00 2018 +0200
Enable nagios checks for ticketkey, and stop
Hi all,
Can I get +1s for the patchset to allow clime to ack service and host issues in
nagios, and to make the initial COPR notifs go to him?
Patrick
From 1994c662f8eabcec78f24a8d7b2edc155d25bac2 Mon Sep 17 00:00:00 2001
From: Patrick Uiterwijk
Date: Sat, 20 Oct 2018 11:19:57 +0200
Subject:
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedorapro
> Is there any particular reason?
I am going to guess that this is since today is a national holiday in the US:
Thanksgiving[1].
[1]: https://en.wikipedia.org/wiki/Thanksgiving
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
> This is a temporary hack to workaround https://pagure.io/pungi/issue/1092.
> Note this also just re-uses the f27 logic, which is no longer needed.
> ---
> roles/bodhi2/backend/templates/pungi.rpm.conf.j2 | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/roles/bodhi2/b
Given the minimal impact of the latest patch unless the config is updated to
use the F30 config, +1 from me.
Also, if it does in fact break stuff, it's easy enough to revert.
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To un
+1
> From: Kevin Fenzi
> In recent months/years, pungi has started to use more and more memory when
> gathering
> package sets.
> So, give it enough memory to avoid the OOM killer killing composes.
>
> Signed-off-by: Kevin Fenzi ---
> inventory/host_vars/compose-x86-01.phx2.fedoraproject.org
> From: Kevin Fenzi
> We need to add these in so pxe sees the right thing to boot them.
>
> Signed-off-by: Kevin Fenzi ---
> .../files/dhcpd.conf.noc01.phx2.fedoraproject.org| 16
>
> 1 file changed, 16 insertions(+)
>
> diff --git a/roles/dhcp_server/files/dhcpd.con
> I am retitleing as I should have said what the patch was for and make sure
> it had agreed upon subject FBR for filtering
>
> On Wed, 24 Apr 2019 at 07:06, Stephen John Smoogen wrote:
+1
___
infrastructure mailing list -- infrastructure@lists.fedorap
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedorapro
+1
___
infrastructure mailing list -- infrastructure@lists.fedoraproject.org
To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedorapro
Hi all,
Could I get +1s for the following Ansible patch?
This adds the domain distro.builders (used for tiny.distro.builders for the
Fedora Minimization Content Resolver) to the nameserver zones configuration.
This domain is being transferred to Red Hat/Fedora control.
Thanks in advance,
Patrick
86 matches
Mail list logo