Intermittent connection issue after connecting to the remote host using tailscale

[eman-cirrusgo]

Hello,

We are using this step within our GitHub workflow, the ping to a remote server after this step sometimes works and sometimes it, doesn't, how can I check further where is the issue?

      - name: Connect to Tailscale
        uses: tailscale/github-action@v2
        with:
          oauth-client-id: ${{ secrets.TS_OAUTH_CLIENT_ID }}
          oauth-secret: ${{ secrets.TS_OAUTH_SECRET }}
          tags: tag:xxx
          version: 1.52.0
      - name: ping the server
        run: |
          ping -c 5 $SSH_HOST
        env:
          SSH_HOST: ${{ secrets.CICD_SSH_HOST }}   

[grelland]

Don't have an answer, but can confirm we are periodically experiencing the same issue on GitHub Actions hosted runners which connect to our Tailnet. Interested in learning more about how to debug it.

[lukasmrtvy]

Related https://github.com/tailscale/github-action/issues/107

[eman-cirrusgo]

I keep getting this in the debug status, and it's becoming to happen frequently...

# Health check:
#     - no DERP home
pong from *** (100.x.x.x) via DERP(dbi) in 632ms
pong from *** (100.x.x.x) via DERP(dbi) in 220ms
pong from *** (100.x.x.x) via DERP(dbi) in 221ms
pong from *** (100.x.x.x) via DERP(dbi) in 223ms
pong from *** (100.x.x.x) via DERP(dbi) in 226ms
pong from *** (100.x.x.x) via DERP(dbi) in 220ms
pong from *** (100.x.x.x) via DERP(dbi) in 220ms
pong from *** (100.x.x.x) via DERP(dbi) in 221ms
pong from *** (100.x.x.x) via DERP(dbi) in 221ms
pong from *** (100.x.x.x) via DERP(dbi) in 221ms
direct connection not established

[kim-anchorzero]

For others who have made the jump to v3 of the action, there was a default timeout of 2 minutes added to the longevity of the connection where the previous was 5m. https://github.com/tailscale/github-action/commit/16d7e0b7812f55e668628a71630a09956522baa9#diff-1243c5424efaaa19bd8e813c5e6f6da46316e63761421b3e5f5c8ced9a36e6b6R135

[oxtoacart]

v4.0.0 of the action now includes a ping parameter that you can use to specify which devices need to be reachable before your CI job proceeds. We are hopeful that this will resolve your issue. If it does not, please let us know by reopening the ticket.