Hi all,
How would I go about identifying why all these 400 (and some 499) errors are popping up in my server’s logs?
[26/Sep/2024:17:12:42 -0400] "POST /inbox HTTP/1.1" 400 137 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:43 -0400] "POST /inbox HTTP/1.1" 400 133 "-" "Lemmy/0.19.3-4-gc98049af6; +https://lemmy.world/"
[26/Sep/2024:17:12:44 -0400] "POST /inbox HTTP/1.1" 400 137 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:44 -0400] "POST /inbox HTTP/1.1" 499 0 "-" "Lemmy/0.18.4; +https://beehaw.org/"
[26/Sep/2024:17:12:44 -0400] "POST /inbox HTTP/1.1" 200 0 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:44 -0400] "POST /inbox HTTP/1.1" 200 0 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:52 -0400] "POST /inbox HTTP/1.1" 499 0 "-" "Mbin/1.7.1 (+https://fedia.io/agent)"
[26/Sep/2024:17:12:53 -0400] "POST /inbox HTTP/1.1" 400 139 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:53 -0400] "POST /inbox HTTP/1.1" 400 140 "-" "Lemmy/0.19.3-4-gc98049af6; +https://lemmy.world/"
[26/Sep/2024:17:12:54 -0400] "POST /inbox HTTP/1.1" 400 135 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:12:59 -0400] "POST /inbox HTTP/1.1" 499 0 "-" "Lemmy/0.18.4; +https://beehaw.org/"
[26/Sep/2024:17:12:59 -0400] "POST /inbox HTTP/1.1" 499 0 "-" "Lemmy/0.18.4; +https://beehaw.org/"
[26/Sep/2024:17:13:00 -0400] "POST /inbox HTTP/1.1" 499 0 "-" "Mbin/1.7.1 (+https://fedia.io/agent)"
[26/Sep/2024:17:13:03 -0400] "POST /inbox HTTP/1.1" 400 137 "-" "Lemmy/0.19.6-beta1; +https://lemmy.ml/"
[26/Sep/2024:17:13:04 -0400] "POST /inbox HTTP/1.1" 400 132 "-" "Lemmy/0.19.3-4-gc98049af6; +https://lemmy.world/"
[26/Sep/2024:17:13:04 -0400] "POST /inbox HTTP/1.1" 200 0 "-" "Lemmy/0.19.3-4-gc98049af6; +https://lemmy.world/"
I don’t see a way to get Lemmy server to spit out any helpful info. I’ve tried setting RUST_LOG=verbose in the env, but no luck, still no log output.
Oh. I subscribed to this post because I hoped someone would be able to give an answer too.
It’s been a day, so before I forget and on the basis that some answer is better than none at all, I’ll have a crack:
400 Bad Request isn’t much to worry about, it doesn’t mean anything is malfunctioning and it can happen for a gazillion reasons. One is you’ve joined a new community and someone Likes a comment you don’t have (particularly if it’s nested in other comments you also don’t have). Another is if someone Likes a post or comment by a user on an instance that you’ve defederated from (your instance is defed’d from lemmygrad and hexbear whereas lemmy.ml isn’t)
As for 499, that seems to be a client issue, and that client mostly seems to be beehaw, who are stuck on an old Lemmy version and being increasingly wonky (in the other direction, they often reply in HTML rather than JSON and randomly decide that their communities’ inboxes don’t exist, so I wouldn’t worry about stuff from them either)
It ended up being an outbound connectivity issue due to firewall changes. Doh. The thing that confused me is Lemmy.ml was still making it through without an outbound response.