Zum Inhalt der Seite gehen


#Forgejo was upgraded to version 9 now. The upgrade went smoothly and there are three months of improvements now available to you. 🥳

You can read about the release on the Forgejo blog: forgejo.org/2024-10-release-v9…

Thanks to the team for the great work.
As usual: report any bugs you might find to our community issues, or if you are an advanced user, kindly forward the reports directly to codeberg.org/forgejo/forgejo/i…

Thanks to the Forgejo contributors!

Als Antwort auf Codeberg.org

I notice a somewhat longer loading time than usual.
I assume it's a combination of the backend starting to come back online completely and many requests at once?
Als Antwort auf Codeberg.org

The update itself went smoothly, however, we're really struggling with crawling again.

This time, some actor is hitting the API searching for random keywords. We cannot detect and block the offending IP ranges quickly enough.

Initially, we started hunting for a potential regression in Forgejo that causes the slowness, but then we noticed the new access pattern that seems to be the real cause instead and just coincidences with the update.

Dieser Beitrag wurde bearbeitet. (1 Monat her)
Als Antwort auf Codeberg.org

thanks for your continued effort on this. I ❤️ Codeberg.

Do you happen to have a US non-profit entity to enable tax deductible donations for US taxpayers?

Dieser Beitrag wurde bearbeitet. (1 Monat her)
Als Antwort auf

@kaleb No, we don't have such an institution. But I'm sure you'll find another lovely US-based non-profit that appreciates your donation and uses it for the benefit of the free software ecosystem.
Als Antwort auf Codeberg.org

Some small bugs were already discovered and fixed by the team.

If you would like to improve Forgejo, take a look at the good first issues codeberg.org/forgejo/forgejo/i… and check out the recently improved #Forgejo contributor documentation: codeberg.org/forgejo/forgejo/i…

Als Antwort auf minecraftchest1

@minecraftchest1 It's complicated by the fact that reverse proxy and access logs are managed in different containers. There is rate-limiting on the reverse proxy, but most crawlers distribute over many IPs. So what we basically need is a way to group by IP range and block the whole range instead of single IP addresses.

Diese Webseite verwendet Cookies. Durch die weitere Benutzung der Webseite stimmst du dieser Verwendung zu. https://inne.city/tos