infrastructure April 2015

infrastructure@lists.fedoraproject.org

46 participants
79 discussions

Re: Potential freeze-break for pkgdb2
by Pierre-Yves Chibon 07 Apr '15

07 Apr '15

On 2015-04-07 17:21, Ralph Bean wrote: > On Mon, Apr 06, 2015 at 08:35:51AM -0600, Kevin Fenzi wrote: >> On Fri, 3 Apr 2015 14:10:07 +0200 >> Pierre-Yves Chibon <pingou(a)pingoured.fr> wrote: >> >> > Good morning everyone! >> > >> > I have a pull-request pending on pkgdb2 [1] for which I am wondering >> > if it is worth a freeze-break. >> >> ...snip... >> >> > What do people think? Worth a freeze break? (If so could I have +1 as >> > well?) >> >> Sure, we could... +1. >> >> However, I wonder if we shouldn't change that it cc's to me and you... >> perhaps it should go to a role/group? admin@ or something? > > Agreed. +1 as is, but it would be nice to set the NOTIFYEMAIL config > value to some sysadmin group in the future. Ok, once the PR is approved, I'll cut a new pkgdb2 release and push it to stg and prod. We can adjust the NOTIFYEMAIL at the next release, any wish who should be spammed with these emails? Thanks for the feedback, Pierre

1 0

Potential freeze-break for pkgdb2
by Pierre-Yves Chibon 07 Apr '15

07 Apr '15

Good morning everyone! I have a pull-request pending on pkgdb2 [1] for which I am wondering if it is worth a freeze-break. The change: When someone in the packager or fedorabugs group change their email in FAS there must be a bugzilla account with that email, otherwise ACLs cannot be synced between FAS/pkgdb and bugzilla. When a mis-match happens, Kevin and I are getting emailed (1/10 minutes) and we recently changed the script so that it emails (1/h) the user as well. The problem is that while the header are set to put Kevin and I in cc to these emails, we are not in the list of address to send the email to (so to the user it appears we are cc'ed while in fact we did not receive the email). This pull-request fixes this situation. The reason I am wondering about the freeze break is that, the code basically works and in the last weeks we have had twice the situation where the user had a mismatch and (maybe thank to the email we sent) they fixed it themselves. So on the pros side for the freeze-break: - it fixes a bug - it is the only change since the last pkgdb2 release, so I can just cut a new release and to deploy it (no hotfix needed) - it is a simple change On the cons side: - it works as is - we had the situation on March 30th and March 31th but these were the first time since March 23rd and before this March 10th (ie: it does not occur so often) - More email for Kevin and I :-p What do people think? Worth a freeze break? (If so could I have +1 as well?) Thanks, Pierre [1] https://github.com/fedora-infra/pkgdb2/pull/175

3 2

Move to Zanata
by Noriko Mizumoto 07 Apr '15

07 Apr '15

Hi Infra team As you already know, we (FLP) are moving our translation platform from Transifex to Zanata. Most of projects have now successfully been migrated, while I found the following projects left behind. I am wondering if any of them belongs to Infra team and would like to be migrated to fedora.zanata.org instance (in this case, 'Infra' group can be newly added). Could you kindly take a look and advise to me (noriko(a)redhat.com)? * https://fedora.transifex.com/projects/p/dnf/ * https://fedora.transifex.com/projects/p/fedoauth/ * https://fedora.transifex.com/projects/p/fas/ * https://fedora.transifex.com/projects/p/fedora-elections/ * https://fedora.transifex.com/projects/p/fedora/ * https://fedora.transifex.com/projects/p/oscap-anaconda-addon/ Thanks noriko

5 8

Freeze Break Request: add long query logging to db-qa01.qa
by Tim Flink 06 Apr '15

06 Apr '15

We're continuing to have timeout issues in Taskotron staging and my investigation so far seems to indicate the cause to be slow queries in the database. This change does require a database restart but due to the way that Taskotron works, I can do this so that there are no lost jobs and no significant downtime. I'd stop incoming jobs until all queues are empty and record the jobs which would have been scheduled on a machine outside of infra. Once those queues are empty, I'd shut down all the db-using processes, apply the patch, restart the db, start everything back up and enqueue the jobs which would have been scheduled. This template change to the postgresql-server module will only affect db-qa01.qa but will make it look like other postgres servers have a pending change due to the way I've changed the postgresql.conf template. I'm not crazy about doing this during freeze but I'm worried that the timeout problem will start affecting more than stg before long and want to get this figured out before that happens. +1s? Tim diff --git a/roles/postgresql_server/templates/postgresql.conf b/roles/postgresql_server/templates/postgresql.conf index 603f9ea..c9756b8 100644 --- a/roles/postgresql_server/templates/postgresql.conf +++ b/roles/postgresql_server/templates/postgresql.conf @@ -319,9 +319,15 @@ log_rotation_size = 0 # Automatic rotation of logfiles will # fatal # panic (effectively off) +{% if ansible_hostname.startswith("db-qa01") %} +log_min_duration_statement = 500 # -1 is disabled, 0 logs all statements + # and their durations, > 0 logs only + # statements running at least this time. +{% else %} #log_min_duration_statement = -1 # -1 is disabled, 0 logs all statements # and their durations, > 0 logs only # statements running at least this time. +{% endif %} #silent_mode = off # DO NOT USE without syslog or # logging_collector

3 4

Freeze break: update koji bulders iptables for compose-x86-01
by Kevin Fenzi 06 Apr '15

06 Apr '15

We recently setup a new compose-x86-01 instance and retired the old compose-x86-02 one. However, I failed to update the iptables rules on the koji builders allowing them to talk to compose-x86-01. This breaks composes of cloud images at least and likely other things related to the release. +1s to apply and run buildvm:buildhw playbooks? kevin -- diff --git a/roles/base/templates/iptables/iptables.kojibuilder b/roles/base/templates/iptables/iptables.kojibuilder index ab4a6ec..ef2f967 100644 --- a/roles/base/templates/iptables/iptables.kojibuilder +++ b/roles/base/templates/iptables/iptables.kojibuilder @@ -31,9 +31,9 @@ -A OUTPUT -p tcp -m tcp -d 10.5.124.138 --dport 80 -j ACCEPT -A OUTPUT -p tcp -m tcp -d 10.5.124.138 --dport 443 -j ACCEPT -# compose-x86-02.fp.o --A OUTPUT -p tcp -m tcp -d 10.5.125.42 --dport 80 -j ACCEPT --A OUTPUT -p tcp -m tcp -d 10.5.125.42 --dport 443 -j ACCEPT +# compose-x86-01.fp.o +-A OUTPUT -p tcp -m tcp -d 10.5.125.41 --dport 80 -j ACCEPT +-A OUTPUT -p tcp -m tcp -d 10.5.125.41 --dport 443 -j ACCEPT # DNS -A OUTPUT -p udp -m udp -d 10.5.126.21 --dport 53 -j ACCEPT

5 4

Hello World!
by Saurabh Sharma 03 Apr '15

03 Apr '15

Hello World!

1 0

UserKnownHostsFile for copr-*-dev machines
by Miroslav Suchý 03 Apr '15

03 Apr '15

Valentin and me are now playing quite a lot with copr-*-dev as part of new OpenStack testing and I always have to ask somebody to wipe the entry from known_hosts on lockbox otherwise rbac will refuse to connect. Can I suggest to put into ssh_config on lockbox: Host copr-be-dev.cloud.fedoraproject.org copr-fe-dev.cloud.fedoraproject.org UserKnownHostsFile /dev/null Can I get two +1? And can somebody put it there then? -- Miroslav Suchy, RHCE, RHCDS Red Hat, Senior Software Engineer, #brno, #devexp, #fedora-buildsys

2 2

MM2 crawler bisect run
by Adrian Reber 02 Apr '15

02 Apr '15

I have set up a git bisect run on mm-crawler01 in hope to find the commit which fixes curl. This means that mm-crawler01 will scan our mirrors in the next few hours continuously. I just want to mention it here in case one of the mirror admins is asking why their mirror is scanned over and over again. It might also put a high load on the database as the crawler has a DB connection open for each thread. Adrian

1 0

FREEZE BREAK REQUEST: Add DNS zones to ns servers
by Stephen John Smoogen 02 Apr '15

02 Apr '15

https://infrastructure.fedoraproject.org/cgit/ansible.git/commit/?id=408399b diff --git a/roles/dns/files/zones.conf b/roles/dns/files/zones.conf index cfdda70..93d1d00 100644 --- a/roles/dns/files/zones.conf <https://infrastructure.fedoraproject.org/cgit/ansible.git/tree/roles/dns/fi…> +++ b/roles/dns/files/zones.conf <https://infrastructure.fedoraproject.org/cgit/ansible.git/tree/roles/dns/fi…> @@ -357,3 +357,13 @@ zone "fedora.events" { file "/var/named/master/built/fedora.events"; }; +zone "pagure.io" { + type master; + file "/var/named/master/built/pagure.io"; +}; + +zone "pagure.org" { + type master; + file "/var/named/master/built/pagure.org"; +}; + -- Stephen J Smoogen.

4 3

crowdsourcing an interview on git
by Matthew Miller 02 Apr '15

02 Apr '15

Hi everyone! Linux Foundation is running a series on Git for its 10th anniversary. They asked me a few questions, and I thought it might be even better to get a community answer from the whole infrastructure team. So, if you have anything interesting to say to any of these, say it, and I'll wrap up the responses into a unified Fedora Whole. I need to respond by Friday. * Why does Fedora use Git? * What makes Git such a great tool? * How many developers do you have collaborating on git? * How much do you personally use it? (estimates are great) * What's Fedora's most active git repo right now and why? * What is your favorite pro tip for using git? * Any git success stories you can share? * Anything else you'd like to say to mark the 10-year anniversary? -- Matthew Miller <mattdm(a)fedoraproject.org> Fedora Project Leader

5 4

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

infrastructure April 2015