Reconciliation timeout

[martadams89]

Issue ##55 still present

Still getting this issue @ChrispyBacon-dev

2025-05-23 22:51:05,071 [INFO] [DockerEventListener] Successfully connected to Docker event stream.
2025-05-23 22:51:20,068 [INFO] [ReconciliationThread] [Reconcile Thread] Starting state reconciliation...
2025-05-23 22:51:20,068 [INFO] [Thread-1] Started reconciliation in background thread ReconciliationThread
2025-05-23 22:51:24,950 [INFO] [ReconciliationThread] [Reconcile] Found 30 running hostnames with DockFlare labels.
2025-05-23 22:55:30,069 [ERROR] [Thread-2] [Reconcile] WATCHDOG: Reconciliation taking too long (250.0s)! Forcing completion.

I think this value should be an env variable so it can be customised, I guess for a large number of records it might take time to parse

For what it's worth I'm using the external Cloudflare tunnel mode...

[ChrispyBacon-dev]

OK thank you for pointing out that you run it in external mode.. I don't have any issues at all but I don't run it in external mode.

After looking into it it really seems related to external mode. With about 30x containers it is a combination not hitting API call limits.

reduce from 0.2s to 0.05s
if USE_EXTERNAL_CLOUDFLARED and i + batch_size < container_count: time.sleep(0.05)
and changed
max_total_time = 480 (8min)

let me know if it works

[martadams89]

Still looks to be problematic, especially when polling other tunnels for records that exist.

The initial reconciliation appeared to work, but when I expanded my other tunnels (one doesn't have any DNS records associated) then it seemed to hang.

I also have multiple DNS Zones on one of my tunnels, and I added the variable to .env for this.