{
  "is": "issue",
  "title": "Intermittent errors on the website",
  "createdAt": "2026-11-04 19:40:00 +0000 UTC",
  "permalink": "https://status.acoustid.org/issues/2026-11-04-website-errors/",
  "severity": "disrupted",
  "resolved": "false",
  "informational": "<no value>",
  "resolvedAt": "<no value>",
  "affected": ["Website"],
  "filename": "2026-11-04-website-errors.md",
  "body": "\u003cp\u003e\u003cstrong\u003eOngoing and getting worse.\u003c/strong\u003e Since about 19:40 UTC on 4 November the website\nhas been returning errors instead of loading pages. It began as separate\nepisodes with normal service in between, but has become much more frequent\novernight — around half of page loads are currently failing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe API is not affected.\u003c/strong\u003e Fingerprint lookups and submissions have continued\nto work normally throughout, so applications such as Picard are unaffected. This\nis limited to browsing the website itself.\u003c/p\u003e\n\u003cp\u003eAffected pages fail outright rather than showing wrong information, so there is\nnothing to re-check — anything that loaded was correct. If a page errors,\nreloading it a little later usually works.\u003c/p\u003e\n\u003cp\u003eNo data has been lost and nothing in the database is affected.\u003c/p\u003e\n\u003cp\u003eThe cause is an internal limit on how many pages the website can look up at\nonce being reached, which makes further requests fail until the load eases. We\nare working on raising that limit.\u003c/p\u003e\n\u003cp\u003eProgress:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e02:00 UTC, 5 November\u003c/strong\u003e — substantially worse. Roughly half of all page\nloads failed over the last hour, up from about one in ten earlier in the\nevening, and the errors are now near-continuous rather than coming in\nseparate episodes. The API is still completely unaffected.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e23:00 UTC, 4 November\u003c/strong\u003e — the episodes are becoming more frequent and more\nsevere: three in the last forty minutes, with up to about two in five page\nloads failing at the worst moments. The API remains unaffected.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e22:00 UTC, 4 November\u003c/strong\u003e — quiet for the last half hour, but the problem has\nrecurred several times this evening, so this is left open rather than marked\nresolved.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e19:40 UTC, 4 November\u003c/strong\u003e — first episode begins.\u003c/li\u003e\n\u003c/ul\u003e\n"
}