nekomimi.pet / wisp.place-monorepo
Monorepo for wisp.place. A static site hosting service built on top of the AT Protocol. wisp.place
87
fork

Configure Feed

Select the types of activity you want to include in your feed.

Site not updating #34

open opened by
bad-example.com

Hello! I love wisp.place!

I have a custom domain configured for hubble.microcosm.blue (though this happened before when I used a wisp subdomain), and updates to the site (uploaded from web) haven't appeared yet after a few hours. (when it happened a while ago, i waited many hours, then eventually deleted and recreated the site to get it updated)

I did hit the sync api endpoint and got an OK but i assume that's just on the atproto side.

If it's just a long cache timeout then no worries, just wondered if something might be up!

[edited to fix title -- at first i had stale files on my PDS after upload and suspected it might be related. didn't repro and they subsequently updated ok]

nekomimi.pet

updates should be able to immediately clear the cache. I redid how i was doing the caching completely a few weeks ago and it was broken for a bit but it seems sometimes it still is. ill check my logs. maybe my firehose consumer was offline during the bit and didnt backfill correctly which i did push fixes for that sort of issues just now

bad-example.com (author)

worked after a CLI deploy! thank you!

nekomimi.pet

okay found the cause of this, i was stupidly caching the file CIDs from website uploads into the db and then the firehose would get the event, and see the CIDs were seemingly unchanged, and then not cache. oops. i really want to guarantee read after write in case the PDS does their own things to the blobs so ill just remove the pre-upsert from main app

bad-example.com (author)

ooooooooooooh ha that's kind of a cool bug.

sign up or login to add to the discussion
Labels

None yet.

assignee

None yet.

Participants 2
AT URI
at://did:plc:hdhoaan3xa3jiuq4fg4mefid/sh.tangled.repo.issue/3mhj6kz63qk22