Notices by Masto.host (mastohost@mastodon.social), page 3

@murtezayesil true :)

OK, after a manual server reboot the issue was solved and every instance is back online.

I will try to investigate and identify what could have caused this issue.

Thanks for your patience.

One app server is down and a small number of Mastodon instances are down.
OVH (the hosting company that I use) is already looking into it and I will post here when I have more information.

@stevenroose No problem at all :) I get it but I have no plans to create a relay server.

@stevenroose I do understand your points but I don’t develop Mastodon and I try my best to keep my service as simple as possible as Masto.host is a side project and I have very limited time available for it.

@stevenroose there are instances that run language specific instances and others that post adult content and stuff like that. I believe that a useful relay requires a theme but that is just my personal opinion and I haven't found one that brings value to my usage. That is why I don't use them. I prefer the Federated Timeline to have the curated content from users followed.

@stevenroose Not really. IMO relay servers require moderation or they become useless and I don't have time to moderate content.

Next Friday (7th of June) at 6:00 AM UTC+2 there will be a planned maintenance reboot of an OVH server rack that will affect 3 app servers of Masto.host cloud.

Estimated downtime: 15 minutes

You can follow the issue here: http://travaux.ovh.net/?do=details&id=38765

@jaz @gomez those values are correct and hardcoded in Mastodon code (not set by me). From what I recall, the idea behind those limits is to avoid Federation problems.

@Taris I am working on an automated backup solution. Still don't have an ETA for that but currently only manual backups are available. Meaning upon email requesting a backup, I will send it but just the database is not enough, you need a backup of the media to be able to move the instance and make exactly the same.

@moiety @zigg yep, there was a problem yesterday in OVH data center: https://mastodon.social/@mastohost/102165229141762066

Sorry about that.

@aral I didn't notice any image fails but it could be OVH Object Storage having an issue.

Please, let me know if it repeats.

@masoud @hugo @Gargron

Both 2) and 4) are applied in Mastodon.

I have daily databases running in a remote server of everything both media and databases, It's a nightmare to run but I have it and it works (I tested it).

About four, obviously I will give time in advanced for people to move to a different hosting.

@Matter @Tcjc I run Docker for over 1 year for all Mastodon instances in shared hosting :)

But thanks for the tip

@aral Nop, first time I get any report. When they are in black is because of the size limit of Mastodon, I know that is the way a huge image in bytes will become black.

@frank87 No, sorry. Masto.host is just a side project. I make no marketing as I have very limited free time to dedicate to this project.

@pvols1979 Well, I can tell you that I only use SSH access to solve issues, to do server maintenance and Mastodon updates. So, besides that and if you don't want to do code changes Mastodon I can't really think of something that you would do with SSH access.

@pvols1979 Nop! To provide a fully managed hosting service I created is a shared hosting environment and instances share resources, it would be impossible to give SSH access.

@kev I wish :) you are too kind
@ecosteader

@ecosteader Well, what I see in the screenshot is a pg_restore and when running "RAILS_ENV=production bundle exec rails db:migrate" it shouldn't do a pg_restore.

So, I don't know. Never came across a similar situation.