Notices by Masto.host (mastohost@mastodon.social)

@dalias @pwaring I understand and thank you :)

But I agree that more should be done to make Mastodon more friendly for smaller instances. Media storage in larger instances is usually not a problem if you have a clean-up like I have for everyone on Masto.host.

The cost of media storage on large instances is a small percentage, as remote media only needs to federate once for all users on a large instance. However, in a small instance, it's significant, and more should be done to address this.

@dalias @pwaring No, there isn't. But that is something that would require Mastodon to support, not something I can easily do on my end.

I don't develop Mastodon. I'm not even much of a code contributor because I know virtually nothing about Ruby/Rails.

Even if that were possible, I'm not sure how much it would impact the user experience. Imagine opening the app (or every time you scroll) and having to wait for media to be fetched remotely and cached locally on the fly.

@dalias Anyone can set 'Media cache retention period' in their Mastodon administration settings.

Not only that, as I have in place auto purge of media cache https://masto.host/re-mastodon-media-storage/ for anyone hosted on Masto.host.

Financially, it is much better to have someone pay me a small amount for years than to have them churn after a couple of months.

Don't think it's a question of the right incentives.

@pwaring

@pwaring Yes, there is an automatic purge in place. You can read https://masto.host/re-mastodon-media-storage/ for details.

As it mentions there, "it’s normal for the media storage to grow very fast in the first few days, but after 7 days, the growth should start to slow down."

All is back online, and timelines are catching up.

Sorry for the trouble this caused.

The only information I have regarding the issue at the moment is: https://public-cloud.status-ovhcloud.com/incidents/wnjcmhtl1ffz

Any issues or questions, please let me know.

Thanks 🐘

Service is slowly coming back online.

There is currently a major incident impacting the OVH data centers and services are down. I am trying to get more information and will update once I know more.

Database Server 14 had a problem.

It stopped responding at around 12:05 AM UTC. I had to issue a manual restart, and at 12:25 AM, it was back online.

So, there was around 20 minutes of downtime for any instance using that server.

Everything looks good now, and I will investigate further to understand what caused the problem.

I am sorry for the trouble this caused.

Thanks 🐘

The Object Storage authentication issue is starting to work.

Some servers may have many background jobs that need to catch up but hopefully it should all be back to normal soon.

There is currently an Object Storage authentication problem that is causing instances to slowdown or to become unresponsive.

The issue is identified and is being addressed: https://public-cloud.status-ovhcloud.com/incidents/9myc4g6tfvlb

Really sorry for the trouble this is causing.

Upgrade installed!

Everyone running on Masto.host is now on Mastodon v4.3.5.

Any issues or questions, please let me know.

Thanks 🐘

I will be starting the installation of Mastodon v4.3.5 on Masto.host.

There will be about 15 seconds of downtime during the upgrade.

Changelog: https://github.com/mastodon/mastodon/releases/tag/v4.3.5

I will be starting the installation of Mastodon v4.3.4 on Masto.host.

There will be about 15 seconds of downtime during the upgrade.

Changelog: https://github.com/mastodon/mastodon/releases/tag/v4.3.4

I'm doing some maintenance in the Masto.host infrastructure. This will cause a reboot on all hosted instances and about 15 seconds of downtime.

No other impact is expected.

@augustocc parabéns 🎉❤️

Upgrade installed!

Everyone running on Masto.host is now on Mastodon v4.3.3.

Any issues or questions, please let me know.

Thanks 🐘

I will be starting the installation of Mastodon v4.3.3 on Masto.host.

There will be about 15 seconds of downtime during the upgrade.

Changelog: https://github.com/mastodon/mastodon/releases/tag/v4.3.3

@ironchamber Well, the backup will only be useful in a last case scenario. There is no easy way to restore it without losing all federation since the backup was generated until the restore happens.

@ironchamber Regarding being safe, it's safe to set a 'Content cache retention period' as long as you are OK with the consequences of losing all cached posts from remote users older than the time you set in there. That signifies that if you favorited, boosted, bookmarked, received a reply from a user that is remote and that is older than that time, that will be lost and unrecoverable.

@ironchamber The easiest way to reduce the database size is to set a 'Content cache retention period', that has consequences like I explain here https://masto.host/mastodon-content-retention-settings/ but that can slow down the growth and even reduce the database size, depending on what you set it to.

Feel free to reach out if you have any questions or if there is something I can help with.