Post Action runs for too long

[RealDyllon]

The Post Run Boostport/setup-cloudflare-warp step runs for approximately 2 minutes on end.

The post-action step appears to be running /usr/bin/warp-cli --accept-tos settings repeatedly for approximately 2 minutes

Sample Step output

Post job cleanup.
/usr/bin/sudo rm /var/lib/cloudflare-warp/mdm.xml

/usr/bin/warp-cli --accept-tos settings
Merged configuration:
(derived)	Always On: true
(network policy)	Switch Locked: false
(network policy)	Mode: WarpWithDnsOverHttps
(local policy)	Disabled for Wifi: false
(local policy)	Disabled for Ethernet: false
(network policy)	Resolve via: ***
(default)	Onboarding: true
(network policy)	Exclude mode, with hosts/ips:
...
(network policy)	Fallback domains:
...
(local policy)	Daemon Teams Auth: true
(network policy)	Disable Auto Fallback: false
(network policy)	Captive Portal: 180
(network policy)	Support URL: [email protected]
(local policy)	Organization: org-name
(local policy)	Auth Client ID: ***
(network policy)	Allow Mode Switch: false
(network policy)	Allow Updates: true
(local policy)	Allowed to Leave Org: false
(network policy)	LAN Access Settings: Allowed until reconnect on a /24 subnet
(network policy)	Profile ID: default

/usr/bin/warp-cli --accept-tos settings
Merged configuration:
(derived)	Always On: true
(network policy)	Switch Locked: false
(network policy)	Mode: WarpWithDnsOverHttps
(local policy)	Disabled for Wifi: false
(local policy)	Disabled for Ethernet: false
(network policy)	Resolve via: ***
(default)	Onboarding: true
(network policy)	Exclude mode, with hosts/ips:
...
(network policy)	Fallback domains:
...
(local policy)	Daemon Teams Auth: true
(network policy)	Disable Auto Fallback: false
(network policy)	Captive Portal: 180
(network policy)	Support URL: [email protected]
(local policy)	Organization: org-name
(local policy)	Auth Client ID: ***
(network policy)	Allow Mode Switch: false
(network policy)	Allow Updates: true
(local policy)	Allowed to Leave Org: false
(network policy)	LAN Access Settings: Allowed until reconnect on a /24 subnet
(network policy)	Profile ID: default

... this continues for ~2 minutes ...

[F21]

Does it eventually finish? The reason it's takes time is because it needs to check if the runner is disconnected from WARP: https://github.com/Boostport/setup-cloudflare-warp/blob/main/lib/setup-cloudflare-warp.js#L179

[giriparus]

Not sure if this is the right approach but it makes sense to me to run the disconnect command explicitly before removing the config file just to be sure. Here's a PR for it.