diff options
| author | Adam Malczewski <[email protected]> | 2026-06-25 18:11:17 +0900 |
|---|---|---|
| committer | Adam Malczewski <[email protected]> | 2026-06-25 18:11:17 +0900 |
| commit | 7626c7f3adf940ee871c4fd2ba2d342f19d9d60b (patch) | |
| tree | 17af00238fd8020bf27a482415f37fb79baf0b59 /packages/lsp/src/tool.ts | |
| parent | c1bc7bfaaca7bdf4d9b2973f5dc88605217a7866 (diff) | |
| download | dispatch-7626c7f3adf940ee871c4fd2ba2d342f19d9d60b.tar.gz dispatch-7626c7f3adf940ee871c4fd2ba2d342f19d9d60b.zip | |
fix(lsp): stop per-edit hangs on dead/slow servers (10s cap + skip + self-heal)
The LSP diagnostics path hung up to 60s per edit whenever a configured Ruby
language server was dead or slow (the reported Steep langserver case): a
killed/crashed server was never detected (stayed "connected" forever), servers
were queried sequentially with a 60s budget each, and a corrupted-but-alive
server (Steep's ~3h phantom-SyntaxError drift) had no recovery.
Four fixes, all in packages/lsp/ (the tool-edit-file call site lowered to 10s):
1. Dead-process detection: SpawnedProcess.onExit (Bun proc.exited) + stdout-end
defence flip the client to error, dispose the rpc, kill the proc. The manager
re-spawns a fresh server after the 30s backoff. Dead servers are now skipped
(0s) instead of polled for 60s.
2. Concurrent fan-out + 10s hard cap: new aggregateDiagnostics queries all
matching servers at once, each capped at 10s. A non-responder is skipped
with "LSP took too long (>10s), skipped — raise this to the user" instead of
blocking the fast server's results. Replaces the vague "unusually long"
warning (now structurally impossible: slow is always false).
3. Corruption self-heal: a detector flags a server re-emitting identical
non-empty diagnostics despite the file changing; after 5 repeats the client
is marked broken and re-spawned. Clean files never trip it. (Acknowledged
false-positive risk on persistent unfixed errors; CLI type-check gate stays
authoritative.)
4. sendRequest timeout: hover/definition/references cap at 10s so they can't
hang the turn against a dead server; the initialize handshake keeps its 45s
race.
Verification: typecheck clean; 1573 tests pass (96 files), +15 new LSP tests
(86 in packages/lsp); biome clean. No kernel/contract changes; onExit is
internal to packages/lsp.
Diffstat (limited to 'packages/lsp/src/tool.ts')
| -rw-r--r-- | packages/lsp/src/tool.ts | 43 |
1 files changed, 21 insertions, 22 deletions
diff --git a/packages/lsp/src/tool.ts b/packages/lsp/src/tool.ts index 8d282ec..be0d269 100644 --- a/packages/lsp/src/tool.ts +++ b/packages/lsp/src/tool.ts @@ -6,6 +6,7 @@ import { extname, resolve } from "node:path"; import type { ToolContract, ToolExecuteContext, ToolResult } from "@dispatch/kernel"; +import { aggregateDiagnostics } from "./aggregate.js"; import type { LspManager } from "./manager.js"; type Operation = "diagnostics" | "hover" | "definition" | "references" | "documentSymbol"; @@ -157,6 +158,8 @@ export function createLspTool(manager: LspManager): ToolContract { switch (operation) { case "diagnostics": { + // 10s hard ceiling per server (same policy as the edit path). + const DIAGNOSTICS_TIMEOUT_MS = 10_000; // Query ALL connected servers whose extensions match this file. const matching = statuses.filter( (s) => s.state === "connected" && s.extensions.some((ext) => ext === fileExt), @@ -179,31 +182,27 @@ export function createLspTool(manager: LspManager): ToolContract { if (!client) { return { content: "Language server client not available.", isError: true }; } - const result = await client.waitForDiagnostics(absolutePath); + const result = await client.waitForDiagnostics(absolutePath, { + timeoutMs: DIAGNOSTICS_TIMEOUT_MS, + }); + if (result.timedOut) { + return { + content: `⚠️ [${connected.name}] LSP took too long (>10s), diagnostics skipped — please raise this to the user.`, + }; + } return { content: result.formatted || "No diagnostics found." }; } - // Query each matching server and merge results, tagged by source. - const parts: string[] = []; - let anyTimedOut = false; - for (const s of matching) { - const client = manager.getClient(s.id, s.root); - if (!client) continue; - const result = await client.waitForDiagnostics(absolutePath, { timeoutMs: 60_000 }); - if (result.timedOut) anyTimedOut = true; - if (result.slow) { - parts.push( - `⚠️ LSP is taking unusually long. If this happens more than once, raise it to the user.`, - ); - } - if (result.formatted) { - parts.push(`[${s.name}]\n${result.formatted}`); - } - } - if (anyTimedOut && parts.length === 0) { - parts.push("Diagnostics timed out (server may still be indexing)."); - } - return { content: parts.length > 0 ? parts.join("\n\n") : "No diagnostics found." }; + // Query matching servers concurrently, each capped at 10s; + // a non-responding server is skipped with a notice. + const agg = await aggregateDiagnostics( + (id, root) => manager.getClient(id, root), + matching, + absolutePath, + DIAGNOSTICS_TIMEOUT_MS, + {}, + ); + return { content: agg.formatted || "No diagnostics found." }; } case "hover": { const client = await getFirstMatchingClient(manager, statuses, fileExt); |
