I wasn't the first one to notice it. Mint was the first one I think.
It's very unpredictable on when it will happen and very likely hardware dependent. If you are running on a garbage tier VPS like I do or slow storage, you are more likely to encounter it. Restarting Pleroma will always fix it, or if you have IEx access to Pleroma, running `Oban.drain_queue(queue: :federator_incoming)` might fix it. The latter did not work for me as Elixir went into some weird state where it refused all localhost connections and somehow dropped all references to Pleroma and Oban during runtime.
If you have monitoring on your server, did you notice an increase in disk usage by Postgres? That's the only lead I currently have (Oban became more resource intensive).
Phantasm /
npub1sh…yrctv
2024-08-14 15:17:41
in reply to nevent1q…z9wl
Author Public Key
npub1shvrwz9qfeqrwzl399gke5yx8kpkjw8xw4ct0f4g9kq4rf24x4tqryrctvPublished at
2024-08-14 15:17:41Event JSON
{
"id": "1e5653ce9864950294703f21f6d0654192c3f092a857bec594d8de3eba8c8185",
"pubkey": "85d83708a04e40370bf129516cd0863d836938e67570b7a6a82d8151a5553556",
"created_at": 1723648661,
"kind": 1,
"tags": [
[
"p",
"85d83708a04e40370bf129516cd0863d836938e67570b7a6a82d8151a5553556"
],
[
"p",
"f02ea9d1659562b42fd2851118625739949550d7515f3e0e2283382fe0056a94"
],
[
"e",
"327e6099a9d45cc43585a73cd289b090fc93f6e97171aaaf61f2d8be0420ce1c",
"",
"root",
"f02ea9d1659562b42fd2851118625739949550d7515f3e0e2283382fe0056a94"
],
[
"p",
"aa1985f61d07ccdc14c11870259eb71306255030c27386bfba1c6c097e9964e9"
],
[
"e",
"0bef75e1abff473e0a4721a84180840528101f27d74a8eacc03049c252e9b5fe",
"",
"reply",
"aa1985f61d07ccdc14c11870259eb71306255030c27386bfba1c6c097e9964e9"
],
[
"proxy",
"https://fluffytail.org/objects/e4677d27-d23c-4495-bba8-2f37e977ad6e",
"activitypub"
],
[
"L",
"pink.momostr"
],
[
"l",
"pink.momostr.activitypub:https://fluffytail.org/objects/e4677d27-d23c-4495-bba8-2f37e977ad6e",
"pink.momostr"
],
[
"-"
]
],
"content": "I wasn't the first one to notice it. Mint was the first one I think. \r\n\r\nIt's very unpredictable on when it will happen and very likely hardware dependent. If you are running on a garbage tier VPS like I do or slow storage, you are more likely to encounter it. Restarting Pleroma will always fix it, or if you have IEx access to Pleroma, running `Oban.drain_queue(queue: :federator_incoming)` might fix it. The latter did not work for me as Elixir went into some weird state where it refused all localhost connections and somehow dropped all references to Pleroma and Oban during runtime.\r\n\r\nIf you have monitoring on your server, did you notice an increase in disk usage by Postgres? That's the only lead I currently have (Oban became more resource intensive).",
"sig": "8fb22f0a7e4ab9f12b90b6e381a93be7e43013d94009395dd4a4f21f9f60efd58877e7843d5cc1ebe6c89330a82c6c4396fd5856ad1cb652dde63cb8320d62cd"
}