Re: Potential freeze-break for pkgdb2
by Pierre-Yves Chibon
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
9 years, 1 month
Potential freeze-break for pkgdb2
by Pierre-Yves Chibon
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
9 years, 1 month
Freeze Break Request: add long query logging to db-qa01.qa
by Tim Flink
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
9 years, 1 month
Freeze break: update koji bulders iptables for compose-x86-01
by Kevin Fenzi
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
9 years, 1 month
UserKnownHostsFile for copr-*-dev machines
by Miroslav Suchý
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
9 years, 1 month
MM2 crawler bisect run
by Adrian Reber
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
9 years, 1 month
crowdsourcing an interview on git
by Matthew Miller
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
9 years, 1 month
