Als je een eigen server hebt (op deze server zijn 2 andere gebruikers) krijg je best een mooi inzicht in je eigen-post gedrag, door in de moderator interface te kijken.
Als je een eigen server hebt (op deze server zijn 2 andere gebruikers) krijg je best een mooi inzicht in je eigen-post gedrag, door in de moderator interface te kijken.
Mise à jour terminée...
#MastoTn #MastoAdmin
Just one more month before https://strangeminds.social is shut down permanently
Lancement de la mise à jour de notre instance Mastodon (mastodon) de 4.5.10~ynh1 à 4.5.11~ynh1
#MastoTn #MastoAdmin
https://www.patreon.com/WelshPixie1/posts/couple-of-tools-160783243
New Patreon post: A couple of tools for battling AI/bot sign-ups.
I think I can confidently say the transition from failing #footiMac trying to handle too many tasks and especially the mastodon database, to a tag team between footiMac and the ten-year younger #newoldiMac running the db and redis process under Colima and docker is complete and successful.
It has been solid for 24 hours now. It’s responsive. It’s almost fast. 😂 even have elasticsearch reactivated.
And the best part: I learned more about docker and feel much more comfortable with it.
In the span of just a week I was left twice with Sidekiq not doing it's job, building up some serious queues and unprocessed events.
I still don't know what caused it, but I suspect that perhaps the DoS bug fixed in Mastodon 4.5.10 was the culprit.
Having now updated to version 4.5.11 it was also the excuse I needed to do some other performance tuning such as optimizing event workers and distributing the load across multiple workers for Sidekiq queues.
Additionally I made some changes to postgres tuning parameters to ensure a more appropriate configuration for the server it was running on; I had not done that before.
All in all, not necessarily downtime but the outcome was pretty much the same anyway.
TL;DR - Some unexpected "downtime" possibly due to a bug in Mastodon, upgraded to 4.5.11, tuned the db, tuned Sidekiq, we're back and kickin' again. I think.
Any one else having #Cloudflare timeout issues?
The file transfer from the VPS to the new server took longer than expected, and I ended up falling asleep - it was midnight, after all. My apologies for the extended downtime. Everything is back to normal now.
#Mastodon #ServerMigration #Downtime #SysadminLife #MastoAdmin #Fediverse #Bangladesh
Bill C-34, the new Digital Safety Act / Safe Social Media Act.
If this passes, this will have sweeping repercussions for the Canadian fediverse.