| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
This is currently the only subdomain with an SPF record and is therefore
the only one from which some mail providers will accept e-mails these
days. Having _noreply in the name makes it more obvious that a reply
will go nowhere.
|
|
|
|
|
| |
If the process info can't be read, just skip it instead of displaying
"[: : integer expression expected"
|
|
|
|
|
|
| |
Remove errors.log or the report won't be generated. Also, log stderr to
the journal or else the normal stderr logs will cause a spurious cron
e-mail to be sent.
|
|
|
|
| |
Apache doesn't support comments here.
|
|
|
|
|
|
|
| |
All these point to valid https: resources, but there is a small chance
that some unusual interaction will cause it not to work. Some of these
changes also won't take effect until the server is restarted, so we'll
need to keep this in mind if failures occur long from now.
|
|
|
|
| |
These are for use by humans only.
|
|
|
|
|
| |
The report will be available at
https://pkgsubmit.mageia.org/spec-rpm-mismatch.html
|
|
|
|
|
|
|
| |
The test jobs seem to be working as desired, so make them actually
start deleting the old files every 4 hours. Use -ignore_readdir_race in
all of them to avoid errors when schedbot cleans the files in the middle
of a run (an unlikely situation because of the clean-up times involved).
|
|
|
|
| |
Follow-up to 82f3da50
|
| |
|
|
|
|
|
|
|
| |
Even if 9/nonfree/backports_testing/ hasn't had any builds recently, it
shouldn't be deleted. Only delete directories deeper in the hierarchy.
Follow-up to f7e017e8
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It's now confirmed that tidy() has been creating huge (2.6 GiB) state
files that the Puppet agent loads before every agent run, which causes
runs to take up to 4 days each and use of all RAM on the server.
Cleaning files using find is more straightforward and efficient and
avoids this problem.
The tidy() functions are disabled here and the cron jobs aren't actually
deleting files yet, so a follow-up commit will enable deleting
imminently, once testing shows it will work.
Follow-up to 59d57245
|
|
|
|
| |
This hasn't been used in over a decade (removed in commit 18854eb0).
|
|
|
|
|
| |
No code changes are needed. Also, drop support for older than mga7 since
the two nodes that need this are running that.
|
|
|
|
|
|
| |
There are no template substitutions needed in these files, so allowing
them opens the danger of substitutions happening unknowingly with future
changes to these files.
|
|
|
|
|
|
| |
It now also passes pytype and flake8 checks.
Also, improve logging in the case of errors
|
|
|
|
|
|
|
| |
Puppet complained with "Files must be fully qualified" using the file()
function.
Follow-up to 0ea383bf2
|
|
|
|
|
|
|
| |
The Puppet method for doing this seems to be what's causing its memory
usage to reach the size of the physical RAM in the system and take up to
4 days to complete a Puppet run. The test will show the files being
deleted and how long it takes, but won't actually delete them.
|
|
|
|
| |
The service doesn't support TLS.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This depends on a newer version of pybugz that's Python 3 compatible
(tested with 0.14) and git_multimail.py, which has already been updated.
Replace token support with API key support, as per the latest pybugz
(and Bugzilla). If an API key is found, it will be used and if not found
it will fall back to username/password. No attempt is made to try to
create an API key in the same way that a token was minted before. Use a
different file name for an API key for coexistence with a token, which
is still used by other programs.
Add a debug flag for enabling more logging to better see when things go
wrong.
Create variables for configuration items.
Log a message when an i18n e-mail is sent.
Do a few little code cleanups.
|
| |
|
|
|
|
|
| |
Don't try to perform template variable substitution because there are
strings in the file that look like substitutions but aren't.
|
| |
|
| |
|
|
|
|
|
| |
This file had what looked like an ERB delimiter eaten by Puppet so git
didn't see the right percent placeholder to get the address.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This is from https://github.com/mhagger/git-multimail
I've reapplied the changes which avoid some unnecessary error messages
when the summary email address is not configured (used for i18n mails).
This latest upstream version also supports Python 3 although the shebang
is left at Python 2.
|
| |
|
|
|
|
|
|
|
| |
Sometimes someone moves a package manually and we end up with several
versions of a package.
In this case we should delete the src.rpm and handle its binary
packages like if it was gone.
|
|
|
|
|
| |
Slow systems/networks might be the reason for the occasional missing
data causing previous crashes.
|
|
|
|
| |
Also, use CSS to highlight the colour summary.
|
| |
|
| |
|
|
|
|
|
| |
This happens every once in a while and results in a cron job failure
e-mail.
|
| |
|
|
|
|
| |
It now also passes pytype and (mostly) flake8 checks.
|
| |
|
|
|
|
| |
This method should work in the old Puppet 3.6 we're using
|
|
|
|
|
| |
This is only currently used by xymon, but it looks like it could be used
to simplify a few other vhosts.
|
|
|
|
| |
This fixes commit eb04f63d
|
|
|
|
| |
Puppet insists on it.
|
|
|
|
|
| |
The cron job uses the slapd file, not the ldap file to configure itself,
so disable the job there instead. This updates commit a27358df4
|
|
|
|
|
|
| |
This only affects mailing lists from @ml.mageia.org and not user e-mails
(from @mageia.org). The mailing list is configured to rewrite the sender
of each message as X-owner@ml.mageia.org
|
|
|
|
|
|
|
| |
The reverse proxy test hard-coded the address of the obsolete alamut
server. Enable SNI on probes so that friteuse returns the correct
certificate. Get the addresses from the configuration instead of
hard-coding them.
|
|
|
|
|
|
|
|
| |
This is try #2 after e0230ba1 was reverted. This time, the sudoers file
is given a name that will cause it to be sorted before any other rules,
allowing those other rules to override it. This is on the assumption
that the revert was needed because a NOPASSWD: rule wasn't taking effect
because this one (which requires a password) took precedence.
|
|
|
|
|
| |
The master is backed up and is the source of reference, and the slave
backup was just throwing up errors due to missing packages.
|
|
|
|
|
|
|
| |
We can have only one group so mga-packagers was made a member of
mga-qa-committers.
Should fix mga#32424
|
|
|
|
| |
This should stop an unnecessary cron warning e-mail.
|