[freeze break request] Make resolv.conf get installed before packages are installed in base.
by Ricky Elrod
In ansible's base role, we install packages before setting resolv.conf.
But if resolv.conf isn't set, DNS queries don't necessarily work (e.g.
when setting up a new virthost for the first time).
This patch moves the resolv.conf task up to the top of the base role to
ensure it's in place before we do anything requiring DNS lookups.
+1's?
-Ricky
diff --git a/roles/base/tasks/main.yml b/roles/base/tasks/main.yml
index 3b71e9c..7af5095 100644
--- a/roles/base/tasks/main.yml
+++ b/roles/base/tasks/main.yml
@@ -4,6 +4,21 @@
# This is the base role for all machines.
# Things in here are things we want to do to every machine no matter what.
#
+
+# XXX fixme # a datacenter 'fact' from setup
+- name: /etc/resolv.conf
+ copy: src={{ item }} dest=/etc/resolv.conf
+ with_first_found:
+ - "{{ resolvconf }}"
+ - resolv.conf/{{ ansible_fqdn }}
+ - resolv.conf/{{ host_group }}
+ - resolv.conf/{{ datacenter }}
+ - resolv.conf/resolv.conf
+ tags:
+ - config
+ - resolvconf
+ - base
+
- name: global default packages to install
yum: state=installed name={{ item }}
with_items:
@@ -111,20 +126,6 @@
- config
- base
-# XXX fixme # a datacenter 'fact' from setup
-- name: /etc/resolv.conf
- copy: src={{ item }} dest=/etc/resolv.conf
- with_first_found:
- - "{{ resolvconf }}"
- - resolv.conf/{{ ansible_fqdn }}
- - resolv.conf/{{ host_group }}
- - resolv.conf/{{ datacenter }}
- - resolv.conf/resolv.conf
- tags:
- - config
- - resolvconf
- - base
-
- name: rsyslog.conf
copy: src={{ item }} dest=/etc/rsyslog.conf mode=644
with_first_found:
9 years, 6 months
[freeze break] Nuke old references to noc03
by Ricky Elrod
I messed up a long time ago when I made noc03 (which became the new
RHEL7 noc01):
[codeblock@lockbox01 puppet]$ host 192.168.1.137
137.1.168.192.in-addr.arpa domain name pointer noc03.vpn.fedoraproject.org.
137.1.168.192.in-addr.arpa domain name pointer
resultsdb01.vpn.fedoraproject.org.
This doesn't matter much anymore because noc03 is now noc01, but I'd
like to fix it up. Before I push the DNS change, I want to remove the
(only) reference to noc03 in puppet, which is on the proxies so requires
+1's. It just nukes two lines that were left over from testing the move
to RHEL7.
Patch follows:
commit b55289c1dba98f8c5cd9b22f34b730427c982ee1
Author: Ricky Elrod <codeblock(a)fedoraproject.org>
Date: Wed Oct 22 03:42:40 2014 +0000
Nuke references to old noc03
Signed-off-by: Ricky Elrod <codeblock(a)fedoraproject.org>
diff --git a/modules/httpd/files/nagios.conf
b/modules/httpd/files/nagios.conf
index ad6b0fb..22fd319 100644
--- a/modules/httpd/files/nagios.conf
+++ b/modules/httpd/files/nagios.conf
@@ -3,6 +3,3 @@ ProxyPassReverse /nagios http://noc01/nagios
ProxyPass /nagios-external http://noc02/nagios-external
ProxyPassReverse /nagios-external http://noc02/nagios-external
-
-ProxyPass /nagios-just-a-test http://noc03/nagios-just-a-test
-ProxyPassReverse /nagios-just-a-test http://noc03/nagios-just-a-test
9 years, 6 months
Alias for list?
by Paul W. Frields
Hi,
We've had a few occasions in the past year where people have sent
email to workstation(a)lists.fp.o instead of the actual Workstation WG
list, desktop@. Would it be possible to alias workstation@ so that
emails there reach the desktop@ list? Does that cause problems that
make it a bad idea?
--
Paul W. Frields http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
http://redhat.com/ - - - - http://pfrields.fedorapeople.org/
The open source story continues to grow: http://opensource.com
9 years, 6 months
[release] Anitya 0.1.10
by Pierre-Yves Chibon
Hi all,
I thought I should keep the good habit.
I just made a new release of anitya and here is the changelog:
* Tue Oct 21 2014 Pierre-Yves Chibon <pingou(a)pingoured.fr> - 0.1.10
- Update to 0.1.10
- Display the number of items returns in the projects, updates and search
templates
- Always display the `Check now` button to admins
- Allow overriding the project's name for the Sourceforge backend
- Use urllib2 for FTP urls instead of requests
- Adjust the default regex for project having src|source before the
version number
- Adjust the default regex for project using srcmin instead of src/source
- Adjust the drupal backend to try '_' instead of '-' if the first attempt
didn't work
- Set the FROM header when making http/ftp queries
- Add keyboard shortcut `e` on the project page to access the edit page
- Add keyboard shortcut `esc` when editing/creating a project to leave
- Adjust the launchpad backend to rely on the project's homepage instead of
its name
Cheers,
Pierre
9 years, 6 months
taskwarrior taskd in infrastructure?
by Matthew Miller
Paul Frields keeps mentioning that this is the greatest thing since the
invention of the calendar, so I thought I'd try it out. Since it seems
rather popular in Fedora Infrastructure, rather than running my own
taskd server, I wonder if there's already a shared one running
somewhere, or plans to do so?
(I see that https://bugzilla.redhat.com/show_bug.cgi?id=1066573 is
stuck in review.)
--
Matthew Miller
<mattdm(a)fedoraproject.org>
Fedora Project Leader
9 years, 6 months
rhel6 -> rhel7 migrations status
by Kevin Fenzi
I took a look at what rhel6 instances we have left and wrote up a
little status report:
easy:·
These just need a sysadmin-main person to do them. I might try and do
them before freeze, but if someone else wants to, great:
smtp-mm* (one done in prod, just do other 2)
unbound-tummy01 (others already done in prod, 1 left)
not too hard:·
These are the ones where we need a playbook written or where they are
already in ansible, but we need to reinstall them with 7 and make sure
all the needed packages for that application are available there.
fas* (needs playbook)
ns* (needs playbook)
secondary01 (needs playbook)
virthosts (some need outage)
sign-vaults (needs being careful)
backup03 (needs outage and being careful)
blockerbugs* (needs playbook)
(all these are in ansible, just need reinstall for rhel7)
(all these have multiple instances, so no outage hopefully)
badges-web*
download-rdu*
elections*
fedocal*
github2fedmsg*
notifs-web*
nuancier*
packages*
pkgdb*
summershum*
tagger*
value*·
hard:·
These are ones we need outages for or more thought, or work on
packages. I don't think we will be doing any of these until after f21
is out, but over the holidays might be a nice time to do some of them
when things are quiet.
pkgs01 (done in stg, needs testing and outage)
proxies (needs thought about how to setup)
db* (playbooks done, needs outages)
hosted* (needs thinking)
koji* (needs replacement for heartbeat)
kojipkgs* (needs outage)
paste* (needs new paste version in epel7 with unbundling)
ask* (need to try and get askbot working in epel7)
people03 (outage, lots of data to move)
torrent02 (needs whatever torrent software we can stand in epel7)
wiki (waiting on next LTR mediawiki)
impossible:·
These are ones where it's not really possible to move the current thing
to 7, and we are waiting for the next major version.
bodhi* (bodhi1)
collab* (mailman2)
hosted-lists* (mailman2)
mirrorlists (mirrormanager)
releng* (bodhi1)
sundries* (mirrormanager)
kevin
9 years, 6 months
Freeze break request: change backup02 snapshot sizes
by Kevin Fenzi
backup02's backup volume is 100% full.
I'm going to make the snapshots it keeps smaller so I can add that room
to the main backup volume.
I've already confirmed that the smaller size should be fine and
will hold the deltas for those days.
+1s?
kevin
--
diff --git a/modules/scripts/files/drBackup.sh b/modules/scripts/files/drBackup.sh
index 8d4dafb..e737155 100644
--- a/modules/scripts/files/drBackup.sh
+++ b/modules/scripts/files/drBackup.sh
@@ -8,7 +8,7 @@
# Delete snapshot three days ago
/sbin/lvremove -f /dev/BackupGroup01/BackupSnap-$(/bin/date -d '3 days ago' +%a) > /dev/null
sleep 30
-/sbin/lvcreate -s -n BackupSnap-$(date +%a) -L 40G /dev/BackupGroup01/backup > /dev/null
+/sbin/lvcreate -s -n BackupSnap-$(date +%a) -L 25G /dev/BackupGroup01/backup > /dev/null
# Database Backups
echo "rsync db1" | logger
9 years, 6 months
Freeze break request: Fix pungify repo for f21 branched
by Kevin Fenzi
I noticed that the nightly branched pungi jobs are including the old
f20 buildroot override repo instead of the 21 one.
+1s?
kevin
--
diff --git a/roles/koji_builder/files/builders/fedora-branched-pungi-armhfp.cfg b/roles/koji_builder/files/build
index 270458e..da18472 100644
--- a/roles/koji_builder/files/builders/fedora-branched-pungi-armhfp.cfg
+++ b/roles/koji_builder/files/builders/fedora-branched-pungi-armhfp.cfg
@@ -27,5 +27,5 @@ baseurl=http://kojipkgs.fedoraproject.org/mash/branched/armhfp/os
[buildroot]
name=buildroot
-baseurl=http://kojipkgs.fedoraproject.org/repos/f20-build/latest/armhfp/
+baseurl=http://kojipkgs.fedoraproject.org/repos/f21-build/latest/armhfp/
"""
diff --git a/roles/koji_builder/files/builders/fedora-branched-pungi-i386.cfg b/roles/koji_builder/files/builder
index 28ce560..6bfffb4 100644
--- a/roles/koji_builder/files/builders/fedora-branched-pungi-i386.cfg
+++ b/roles/koji_builder/files/builders/fedora-branched-pungi-i386.cfg
@@ -27,5 +27,5 @@ baseurl=http://kojipkgs.fedoraproject.org/mash/branched/i386/os
[buildroot]
name=buildroot
-baseurl=http://kojipkgs.fedoraproject.org/repos/f20-build/latest/i386/
+baseurl=http://kojipkgs.fedoraproject.org/repos/f21-build/latest/i386/
"""
diff --git a/roles/koji_builder/files/builders/fedora-branched-pungi-x86_64.cfg b/roles/koji_builder/files/build
index d6b9a89..56caf8b 100644
--- a/roles/koji_builder/files/builders/fedora-branched-pungi-x86_64.cfg
+++ b/roles/koji_builder/files/builders/fedora-branched-pungi-x86_64.cfg
@@ -30,5 +30,5 @@ baseurl=http://kojipkgs.fedoraproject.org/mash/branched/x86_64/os
[buildroot]
name=buildroot
-baseurl=http://kojipkgs.fedoraproject.org/repos/f20-build/latest/x86_64/
+baseurl=http://kojipkgs.fedoraproject.org/repos/f21-build/latest/x86_64/
"""
9 years, 6 months
