Push gateway errors

I am currently hosting an Rocket Chat instance in Azure on k8s and I am seeing a lot of

rocketchat_logger rocketchat_logger.js:278 e[31mSystem ➔ error Error sending push to gateway (0 try) -> { Error: connect ECONNREFUSED     at Object._errnoException (util.js:1022:11)     at _exceptionWithHostPort (util.js:1044:20)     at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1182:14)    code: 'ECONNREFUSED',   errno: 'ECONNREFUSED',   syscall: 'connect',   address: '',   port: 443 } 

I went a head and proactively added a rule to the network service group to allow outbound access to this IP but still see this failure off and on for periods of time.

I am not able to reproduce the failure from my desktop as I am from azure so i am suspecting rate limiting or black listing. Can any one from the RC team shine some light on this?

That is the wrong ip :frowning: I’m not sure why its resolving incorrectly. But it should be: I wouldn’t recommend hard-coding to this in any sort of workaround as it might change again in future.

Its been changed for a good while. I’d check your dns resolver and see if maybe the cache needs flushed or something.

I kicked over the kube dns services local to the cluster and the correct IP is now resolving. I can consider this resolved and will follow up with the Azure folks to see why their kube dns setup is not honoring TTLs.


1 Like

Might be a general kube dns issue. If you do find an upstream issue definitely let me know. We use a lot of kubernetes and would be good to track the issue.