As an otherwise busy person involved with DNF Counting,
I want that the scripts that are run unattended only send mails when there's something wrong,
so that I'm not disturbed if everything is fine and I can notice anything out of the ordinary better.
Right now, the condense-mirrorlogs.sh and countme-update.sh scripts send mails to a set of people/accounts (root@fpo, mattdm, asaleh, nils) every day. It's annoying and makes it easy to miss important things this way.
condense-mirrorlogs.sh
countme-update.sh
Subject: Cron <root@log01> /usr/local/bin/condense-mirrorlogs.sh > /dev/null
Cron <root@log01> /usr/local/bin/condense-mirrorlogs.sh > /dev/null
Outputting data: /mnt/fedora_stats/combined-http/2021/09/05/mirrors.fedoraproject.org-access.log
Subject: Cron <countme@log01> /usr/local/bin/countme-update.sh
Cron <countme@log01> /usr/local/bin/countme-update.sh
parse-access-log.py --sqlite /var/lib/countme/raw.db /mnt/fedora_stats/combined-http/2021/09/05/mirrors.fedoraproject.org-access.log /mnt/fedora_stats/combined-http/2021/09/06/mirrors.fedoraproject.org-access.log
Metadata Update from @nphilipp: - Issue tagged with: Bug, DNF Counting
@smooge Hey, it may be that I'm not receiving all the spam the various DNF Counting cron jobs send. Is there anything that should be added to the list of things to clean? I remember seeing a mail of yours mentioning the old version of the log syncing script, is the new one still as noisy? I know the script is but I don't know if stuff dropped in etc/cron.daily will send mails if there is any output or only if there's something on stderr or a non-zero exit code.
etc/cron.daily
Metadata Update from @nphilipp: - Issue assigned to nphilipp
PR: fedora-infra/ansible#798
This has been merged, but needs to be deployed still.
And one more fix: fedora-infra/ansible#810
And another one: #55
OK, I've updated mirrors-countme to 0.0.7 on log01, so it seems that the last PR to be merged is: fedora-infra/ansible#810 (which is hot-fixed on log01, but keep the issue open until it's merged for bookkeeping)
Issue status updated to: Closed (was: Open)
Log in to comment on this ticket.