Dokploy/dokploy · error · Error

Failed to fetch models: ${errorText}

Error message

Failed to fetch models: ${errorText}

What it means

Thrown after a successful Docker removal attempt when the database DELETE on the network table returns no row — i.e. no network row with that networkId exists. NOT_FOUND (not the Docker errors above), and thrown only when .returning() is empty.

Source

Thrown at apps/dokploy/server/api/routers/ai.ts:138

							{
								id: "MiniMax-M2.7",
								object: "model",
								created: Date.now(),
								owned_by: "minimax",
							},
						] as Model[];
					default:
						if (!input.apiKey)
							throw new TRPCError({
								code: "BAD_REQUEST",
								message: "API key must contain at least 1 character(s)",
							});
						response = await fetch(`${input.apiUrl}/models`, { headers });
				}

				if (!response.ok) {
					const errorText = await response.text();
					throw new Error(`Failed to fetch models: ${errorText}`);
				}

				const res = await response.json();

				if (Array.isArray(res)) {
					return res.map((model) => ({
						id: model.id || model.name,
						object: "model",
						created: Date.now(),
						owned_by: "provider",
					}));
				}

				if (res.models) {
					return res.models.map((model: any) => ({
						id: model.id || model.name,
						object: "model",
						created: Date.now(),

View on GitHub (pinned to 546686ea35)

Solutions

1. Treat a 404 NOT_FOUND from removeNetwork as idempotent success if the goal is deletion
2. Refresh the network list in the client after deletion to avoid replaying stale IDs
3. Guard against double-submission in the UI (disable button while pending)
4. If it persists, confirm the networkId being sent matches an existing row

Example fix

// before
await api.network.remove.mutate(id);
// after
try {
  await api.network.remove.mutate(id);
} catch (e) {
  if (e?.data?.code === 'NOT_FOUND') return; // already deleted
  throw e;
}

const exists = await db.query.networks.findFirst({
  where: eq(network.networkId, id),
});
if (!exists) return; // nothing to do

try { await removeNetwork(id); } catch (e) { if (e?.data?.code === 'NOT_FOUND') return; throw e; }

When it happens

Trigger: Calling removeNetwork twice in a row (second call finds no row); deleting a network whose DB row was already removed; passing a stale or wrong networkId; concurrent deletion racing this call.

Common situations: UI retry after a timeout where the first request actually succeeded; duplicated delete buttons firing; state desync between client cache and server.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

• INTERNAL_SERVER_ERROR
• BAD_REQUEST
• INTERNAL_SERVER_ERROR
• NOT_FOUND
• NOT_FOUND

AI-assisted analysis of Dokploy/dokploy@546686ea35 (2026-08-27). Data as JSON: /api/errors/c7afdd3881807b4c. Report an issue: GitHub.