cloudflared receiving incorrect tunnel ID while CNAME updates correctly

[valkmit]

Trying to get to the bottom of a very strange issue...

I've deployed this as a helm chart using a tunnel created with ID A.

Then, I deleted the tunnel and uninstalled the helm chart, and then created tunnel with ID B.

I install the helm chart again with tunnel ID B, but the cloudflared pod is attempting to set up the tunnel with ID A (which fails)

Interestingly it's able to set the CNAME records correctly to tunnel ID B, but it tries to establish the tunnel to ID A.

My only guess is that there's some kind of resource that's not being cleaned up.

cloudflared logs:

2024-12-24T14:33:00Z INF Starting tunnel tunnelID=9d6ec092-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-12-24T14:33:00Z INF Version 2024.12.2 (Checksum cb61fcb41380efff4e2ef0a1eeea23bcddff360e8703f99dbb5c24da1de937b9)
2024-12-24T14:33:00Z INF GOOS: linux, GOVersion: go1.22.5-devel-cf, GoArch: amd64
2024-12-24T14:33:00Z INF Settings: map[metrics:0.0.0.0:44483 no-autoupdate:true p:auto protocol:auto token:*****]

ingress controller logs:

2024/12/24 14:20:49 main/tunnel-client: "caller"={"file"="tunnel-client.go" "line"=162} "level"=0 "msg"="update DNS record" "id"="8ca2476c6fd2e07f705c6a0b071dee28" "type"="CNAME" "hostname"="XXXXXXXXXX" "content"="f0de3570-XXXX-XXXX-XXXX-XXXXXXXXXXXX.cfargotunnel.com"

[valkmit]

I managed to fix this issue by deleting the controlled-cloudflared-connector deployment manually, uninstalling the helm chart, and then reinstalling it. Not sure what the actual underlying issue was