Since I think this was more of a GCP-related question, I have asked on Stackoverflow in the meantime: https://stackoverflow.com/questions/60584349/any-idea-why-this-port-is-not-reachable-from-another-gcp-instance
Kind regards, Stéphane Thibaud On Sun, Mar 1, 2020 at 9:27 PM Stéphane Thibaud <[email protected]> wrote: > Hello Gianluca, > > Thank you very much for your answer. Strangely it does not accept the > connection. I am indeed using an internal static IP. In the meantime I did > a portscan from another GCP instance and it looks like the port shows as > 'filtered'. Maybe it is indeed a GCP firewall problem, but I would expect > the default firewall rules to expose the port. > > > Kind regards, > > Stéphane Thibaud > > 2020年3月1日(日) 19:35 Gianluca Bonetti <[email protected]>: > >> Hello Stéphane >> >> GCP should allow internal traffic using the internal IP address (static >> or not), usually a 10.xxx.xxx.xxx, but should deny connections on special >> ports using the external IP address (static or not). >> If you need to use an external IP address you have to add a firewall >> rule, but I think it would be fixed by using the internal IP addresses. >> >> Best regards >> Gianluca Bonetti >> >> Il giorno dom 1 mar 2020 alle ore 08:39 Stéphane Thibaud < >> [email protected]> ha scritto: >> >>> Hello Ignite users,, >>> >>> I suddenly am unable to connect to the Ignite instance. I am running it >>> on GCP with Container OS/Docker. I checked by SSH-ing into the instance and >>> running netstat that port 10800 is listening: >>> >>> Proto Recv-Q Send-Q Local Address Foreign Address >>> State PID/Program name >>> ... >>> tcp 0 0 0.0.0.0:10800 0.0.0.0:* >>> LISTEN - >>> I have allocated a static IP to this instance and connecting to it >>> previously worked (firewall rule should allow internal GCP connections). >>> Now I get 'timeout' error even though I can ping the machine without >>> problem. >>> >>> Any ideas for additional checks? >>> >>> >>> Kind regards, >>> >>> Stéphane Thibaud >>> >>
