Problem with scavenging and calling incorrect port ?

[iangregsondev]

Hi,

great chart! I have everything working on the UI but I notice scavenging maybe hitting the wrong port - not sure.

I am getting an error on the scheduled job so I decided to kubectl exec manually into the pod .. and I notice the port it is trying to use - it is not listening

root@eventstore-0:/# lsof -i
COMMAND   PID USER   FD   TYPE    DEVICE SIZE/OFF NODE NAME
eventstor   1 root   17u  IPv4 128987236      0t0  TCP eventstore-0.eventstore.eventstore.svc.cluster.local:1113 (LISTEN)
eventstor   1 root   20u  IPv4 128986353      0t0  TCP *:2113 (LISTEN)
eventstor   1 root   21u  IPv4 129025536      0t0  TCP eventstore-0.eventstore.eventstore.svc.cluster.local:2113->10-42-1-237.eventstore-admin.eventstore.svc.cluster.local:52412 (CLOSE_WAIT)

2113 is there but not 2112 which is what the scavenge job tried to use..

If I use the CURL command manually inside the pod and call it using 2113 - this works.

This is my chart-values

scavenging:
  enabled: true
  image: lachlanevenson/k8s-kubectl
  imageTag: latest
  schedule: 0 2 * * *

I haven't changed any other settings - so I assume its using these

## Internal HTTP 
port.intHttpPort: 2112
#### External HTTP 
port.extHttpPort: 2113

By default.

Any idea what I should be doing ?

should I create an intHttpPort and apply it as 2113.

Isn't intHttpPort to do with the gossip between clusters ?

I do notice the Config Map is using the intHttpPort

        kubectl exec ${podname##pod/} -- \
          curl \
            -si \
            -d {} \
            -X POST \
            -u 'admin:{{ .Values.admin.password | default "changeit" }}' \
            "http://localhost:{{ .Values.intHttpPort }}/admin/scavenge"
            

Any ideas ? - Thanks in advance.

[andymeierdev]

Hey sorry, just noticed the issues on this repo. Will try to take a look this weekend.