Maintenance has concluded!

Whoops-- Turns out, no it wasn’t. I didn’t realize public channels weren’t searchable by default over federation XD

The #pawb:pawb.social space and its channels should be public to everyone + federated; If you’re still getting issues, let me know!

Yes! We’ve got one up at https://matrix.pawb.social/ via web, or pawb.social via apps. You should be able to sign in using your Lemmy (pawb.social), furry.engineer, or pawb.fun account :3

Apologies-- I’m still trying to find a way of properly integrating the images back into Pict-rs, but I’m going to look at making a stop-gap to make the previous images visible again via Nginx. I suspect this could cause some image conflicts since the old image IDs may / may not be re-used, but it should at least fix some of the broken image icons in the meantime.

We haven’t yet got the images reimported so existing images and profile pictures will still be broken. The base URL without the format link would have been cached most likely, but the format link would error as pict-rs isn’t aware of the file you’re trying to reference.

For now, reuploading things like profile pictures is my recommendation rather than waiting for us to get the images reimported.

We did discuss briefly with Asonix but they weren’t sure it’d be possible, so we’re exploring options using an external database for pict-rs but that seems relatively new.

tl;dr we have the files, but are still working on getting them reimported!

Just found out this was happening and sad I missed it, but must say… seeing the results of y’all’s work is damn impressive! There was also stats linked on the canvas@toast.ooo community: https://pawb.social/post/28021100, so for the name I saw first at #2 out of all 638 users, well done @tblFlip@pawb.social with 4598 pixels and thank you to everyone who contributed and organized \o/

Strangely… I was tweaking with Cloudflare rules again but I didn’t think I changed anything that should affect that… Please do let me know if it occurs again, but I’m glad it’s fixed now x3

Investigating – I’m not able to find anything obvious in our logs, and it doesn’t look like requests from Jerboa are getting rejected per se. When it occurs again, could you try to access the same content via a browser and see if a message comes up from Cloudflare?

Paradoxically, Voyager is what I use myself, so it was allowlisted! There were a ton of rules added during the DDoS, so I’ve trimmed it down a bit 😅

Appreciate this… I’m wondering if there’s some kind of race condition that’s causing the DB (or your client) to retry posting. Will investigate once I’m back from a work trip early next week!

Good question, and I’ve not thought about it previously other than that we would remove content if the original artist requested us to do so. Though, we don’t have a formal list of DNP content, we just expect that users post content they’re permitted to or that is already otherwise shared on other sites (like e621).

I’m not against a DNP list, but I think I’d lean more towards respecting an existing DNP like https://e621.net/wiki_pages/85. Though, if we did end up getting to the point of having a lot of DNPs that don’t appear on their list, we’d probably move to making our own.

I’ll bring this up to the rest of the team too for feedback!