My Setup: I have a single machine docker swarm "cluster". Simply said, I have a stack deployment that is composed of two services A and B. Both services are connected (through an external overlay network) to another stack running a traefik proxy to expose those services to the public. I can reach both services via their traefik routing from my browser.
What doesn't work though: I can not reach service A from within service B using A's public domain (via its traefik routing). I always get a connection timeout when attempting a HTTP call.
I this some regular expected behavior that can be fixed with some option or is my setup somehow broken? I read that endpoint_mode: dnsrr might help in some situations of this kind but it really didn't make a difference for me. I tried it on both services as well as on the traefik service.
I don't want to overwhelm you with all the configuration details of my machine and swarm deployments right here as that might be overkill if I just made a configuration mistake that's obvious from my problem description.
For the ambitious reader, here are some further details:
$ docker info
Client:
Debug Mode: false
Server:
Containers: 125
Running: 59
Paused: 0
Stopped: 66
Images: 328
Server Version: 19.03.8
Storage Driver: overlay2
Backing Filesystem: <unknown>
Supports d_type: true
Native Overlay Diff: true
Logging Driver: loki
Cgroup Driver: cgroupfs
Plugins:
Volume: local
Network: bridge host ipvlan macvlan null overlay
Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
Swarm: active
NodeID: juwtdqk60ufj3rityctog7ev0
Is Manager: true
ClusterID: 3mte2zcw4nfc1jq17dzwvtoi3
Managers: 1
Nodes: 1
Default Address Pool: 10.0.0.0/8
SubnetSize: 24
Data Path Port: 4789
Orchestration:
Task History Retention Limit: 5
Raft:
Snapshot Interval: 10000
Number of Old Snapshots to Retain: 0
Heartbeat Tick: 1
Election Tick: 10
Dispatcher:
Heartbeat Period: 5 seconds
CA Configuration:
Expiry Duration: 3 months
Force Rotate: 0
Autolock Managers: false
Root Rotation In Progress: false
Node Address: 178.254.21.80
Manager Addresses:
178.254.21.80:2377
Runtimes: runc
Default Runtime: runc
Init Binary: docker-init
containerd version: 7ad184331fa3e55e52b890ea95e65ba581ae3429
runc version: dc9208a3303feef5b3839f4323d9beb36df0a9dd
init version: fec3683
Security Options:
apparmor
seccomp
Profile: default
Kernel Version: 4.15.0-96-generic
Operating System: Ubuntu 18.04.4 LTS
OSType: linux
Architecture: x86_64
CPUs: 8
Total Memory: 31.41GiB
Name: rv1324
ID: L7JD:POVR:EPMY:JGMF:3BFX:DQJA:B5KK:O3PG:YH44:TNLJ:YD3I:GROZ
Docker Root Dir: /var/lib/docker
Debug Mode: false
Registry: https://index.docker.io/v1/
Labels:
Experimental: false
Insecure Registries:
127.0.0.0/8
Live Restore Enabled: false
The swarm stack I'm trying to deploy can be found here
https://github.com/skuzzle/cmp/blob/0f8004b41f1a486fee7b6705c4bcbc39a2414412/swarm-stack/feature.yml
The connectivity problem is between the cmpauthorization and the cpmfrontend service. In order to finish an OAuth2 authorization process, the latter service needs to send a POST request to the authorization service's public domain.
