update

backend/agents.py

@@ -26,6 +26,12 @@ _active_processes: dict[str, asyncio.subprocess.Process] = {}
 _batch_sem = asyncio.Semaphore(MAX_CONCURRENT_AGENTS)
 _interactive_sem = asyncio.Semaphore(MAX_CONCURRENT_INTERACTIVE)
 
+# OpenCode-Starts serialisieren: gleichzeitig startende Prozesse kollidieren an
+# der internen Session-DB ("database is locked", Exit nach <1s). Der kurze
+# Versatz entzerrt die Starts; danach laufen die Prozesse normal parallel.
+_opencode_start_lock = asyncio.Lock()
+_OPENCODE_START_DELAY = 1.0
+
 # Capability → Claude --allowedTools
 _CLAUDE_TOOLS = {
     "full": "Write,Bash,Read,WebSearch,WebFetch",
@@ -92,14 +98,23 @@ async def run_agent(
         return await _run_claude_cli(agent_key, prompt, timeout, role, capabilities)
 
 
-async def _communicate(agent_key: str, cmd: list[str], stdin_data: bytes | None, timeout: int) -> tuple[int, str, str]:
+async def _communicate(agent_key: str, cmd: list[str], stdin_data: bytes | None, timeout: int, stagger: bool = False) -> tuple[int, str, str]:
     start = time.monotonic()
-    process = await asyncio.create_subprocess_exec(
-        *cmd,
-        stdin=asyncio.subprocess.PIPE if stdin_data is not None else asyncio.subprocess.DEVNULL,
-        stdout=asyncio.subprocess.PIPE,
-        stderr=asyncio.subprocess.PIPE,
-    )
+
+    async def spawn():
+        return await asyncio.create_subprocess_exec(
+            *cmd,
+            stdin=asyncio.subprocess.PIPE if stdin_data is not None else asyncio.subprocess.DEVNULL,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.PIPE,
+        )
+
+    if stagger:
+        async with _opencode_start_lock:
+            process = await spawn()
+            await asyncio.sleep(_OPENCODE_START_DELAY)
+    else:
+        process = await spawn()
     _active_processes[agent_key] = process
     try:
         try:
@@ -154,7 +169,7 @@ async def _run_opencode(agent_key: str, prompt: str, timeout: int, provider: str
         "-f", str(prompt_path),
     ]
     try:
-        rc, stdout, stderr = await _communicate(agent_key, cmd, None, timeout)
+        rc, stdout, stderr = await _communicate(agent_key, cmd, None, timeout, stagger=True)
         return rc, _clean_opencode_output(stdout), stderr
     finally:
         prompt_path.unlink(missing_ok=True)