NgTech-LTD/gns3-server
2
0
Fork 0
mirror of https://github.com/GNS3/gns3-server.git synced 2026-10-24 20:02:35 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix: Add exponential backoff to compute reconnection attempts

Browse Source 
When a remote compute is unreachable, the controller now uses exponential
backoff for reconnection attempts: 5s, 10s, 20s, 40s, 80s, then caps at
300s (5 minutes). Previously it retried every 5 seconds indefinitely.

User-initiated operations (open project, start node) still trigger an
immediate connection attempt, so recovery is not delayed in practice.

Related: #2704
This commit is contained in:
YueGuobin 2026-05-07 11:17:55 +08:00
parent 255d1e4057
commit efe6913d69
No known key found for this signature in database
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
gns3server/controller/compute.py
View File

@ -383,14 +383,16 @@ class Compute:
if report_failed_connection: if report_failed_connection:
raise raise
log.warning(f"Cannot connect to compute '{self._id}': {e}") log.warning(f"Cannot connect to compute '{self._id}': {e}")
# Try to reconnect after 5 seconds if server unavailable only if not during tests (otherwise we create a ressource usage bomb) # Try to reconnect if server unavailable only if not during tests (otherwise we create a ressource usage bomb)
if not hasattr(sys, "_called_from_test") or not sys._called_from_test: if not hasattr(sys, "_called_from_test") or not sys._called_from_test:
self._connection_failure += 1 self._connection_failure += 1
# After 5 failure we close the project using the compute to avoid sync issues # After 10 failures we close the project using the compute to avoid sync issues
if self._connection_failure == 10: if self._connection_failure == 10:
log.error(f"Could not connect to compute '{self._id}' after multiple attempts: {e}") log.error(f"Could not connect to compute '{self._id}' after multiple attempts: {e}")
await self._controller.close_compute_projects(self) await self._controller.close_compute_projects(self)
asyncio.get_event_loop().call_later(5, lambda: asyncio.ensure_future(self._try_reconnect())) # Exponential backoff: 5s, 10s, 20s, 40s, 80s, then cap at 300s
delay = min(5 * (2 ** (self._connection_failure - 1)), 300)
asyncio.get_event_loop().call_later(delay, lambda: asyncio.ensure_future(self._try_reconnect()))
return return
except web.HTTPNotFound: except web.HTTPNotFound:
raise ControllerNotFoundError(f"The server {self._id} is not a GNS3 server or it's a 1.X server") raise ControllerNotFoundError(f"The server {self._id} is not a GNS3 server or it's a 1.X server")