this post was submitted on 27 Jan 2024
983 points (96.3% liked)

Lemmy.World Announcements

29062 readers
5 users here now

This Community is intended for posts about the Lemmy.world server by the admins.

Follow us for server news ๐Ÿ˜

Outages ๐Ÿ”ฅ

https://status.lemmy.world/

For support with issues at Lemmy.world, go to the Lemmy.world Support community.

Support e-mail

Any support requests are best sent to [email protected] e-mail.

Report contact

Donations ๐Ÿ’—

If you would like to make a donation to support the cost of running this platform, please do so at the following donation URLs.

If you can, please use / switch to Ko-Fi, it has the lowest fees for us

Ko-Fi (Donate)

Bunq (Donate)

Open Collective backers and sponsors

Patreon

Join the team

founded 1 year ago
MODERATORS
983
submitted 9 months ago* (last edited 9 months ago) by [email protected] to c/[email protected]
 

Version 0.19.X Deployment

Hello world!

I'm sure many of you fine folks have been wondering why we have not upgraded to version 0.19.X yet.

The whole team here has been getting asked this question quite a bit, from both members of the community and other instance admins. We want everyone to know, YES, we will upgrade to version 0.19.

We do not have a hard date set for when we will be upgrading as of yet, as we have about 7 associated task items that go along with this upgrade for things we need to double-check and confirm there will be no breakage.

Another big reason why we have not upgraded is due to reports from other instance admins that the front-end GUI can have performance issues. While a large portion of folks access our site using apps on their phones, we would rather not degrade the user experience for the rest of everyone else.

Worst case in terms of UI slowness, we would ask everyone to use either Alexandrite () or Photon () to mitigate the potential performance issues.

One final point to mention to address the delay is that we usually wait 1-2 weeks after a new release before applying it to production. This was complicated by the fact that there were multiple releases right after each other, with each new release having issues we hoped would be patched in the NEXT version. A new version would come up, and we would have to evaluate the stability and performance of it, which would take about 2 weeks, and then ANOTHER version would come out, which we would need to test. This was further complicated by the fact this was during the holiday season.

Version Release Date
0.19.0 2023-12-15
0.19.1 2023-12-20
0.19.2 2024-01-10
0.19.3 2024-01-24

For us, stability is paramount, as we care very much about keeping everyone happy here.

While we can appreciate the rapid release of patches on 0.19, I hope everyone can now understand the delay. If all goes well, we will set a release day in the next week or two, most likely on a Saturday around 1800 UTC.

-jelloeater85

you are viewing a single comment's thread
view the rest of the comments
[โ€“] [email protected] 23 points 9 months ago (6 children)

Hey just FYI, I've been experiencing a postgres memory leak issue on my instance since the upgrade to 0.19.2. Theres a GitHub issue open about it (#4406), not sure if it's something that will affect you but probably worth keeping in mind

[โ€“] [email protected] 2 points 9 months ago* (last edited 9 months ago) (3 children)

We are investigating another possible bug, that sorta spams other instances full of requests ( we say 5k requests per 5 seconds ) kinda like a DOS attack. We only saw this phenomena on 0.19.3 instances, and they are hitting us currently hard with many requests. We will have to block them until it gets fixed on their side.

[โ€“] [email protected] 1 points 9 months ago (1 children)

Hmmmm interesting, i hope im not on the ban list! I haven't noticed any weird traffic like that

[โ€“] [email protected] 2 points 9 months ago (1 children)

Nope you are not ;)

We noticed it that it happens on smaller instances, that probably doesnt have the perfect hardware to run the 0.19 version and crashes => retries federation and they probably dont have a backoff strategy.

Because we noticed the site the domain was leading too wasnt responsive half of the time we tried to look if we can contact the admin.

[โ€“] [email protected] 2 points 9 months ago

LETSGO

Yeah my instance is small but my uptime is pretty good (except this db issue tbh) so that might be why

load more comments (1 replies)
load more comments (3 replies)