Hi,
I have storm cluster that basically works fine. Now I have to add two
more supervisors to this cluster and when I'm doing that I get this
exception (on supervisor machine that is being added):
2018-03-23 10:17:45.539 o.a.s.d.supervisor [INFO] Downloading code for
storm id topo1
...
2018-03-23 10:17:55.629 o.a.s.d.supervisor [WARN] There was a connection
problem with nimbus. #error {
:cause Connection refused (Connection refused)
The only difference between already working supervisors and new
supervisor is that the new one is in different network.
But:
- new supervisor machine is listed in UI;
- I can telnet nimbus machine from new supervisor machine on port 6627;
- tcpdump used on new supervisor machine tells me that new supervisor
machine is using 6627 port only (there is no other communication with
nimbus)
- nimbus machine correctly responds with checkStatus (it is logged in
nimbus log and I can see this in tcpdump).
And now I don't know what to try now.
Is there any nimbus option that forbids communication like this? Maybe
sth like this is on the supervisor side?
What can I do more to find out what is wrong in here?
Regards,
Pawel
