Container shutdown when 1 instance available about caprover HOT 9 CLOSED

Alright. I'll look into this.

from caprover.

Make sure to deploy your image after you have set the docker registery. Just re-deploy the app and see if it solves it.

The reason is that when you deploy the app before connecting to registry, the image is not replicated in the registry. Therefore only manager has access to it.

from caprover.

I can make sure that the registry is working as I configure it as self registry hosting and the worker node is joined.
I re-deploy the sample node app, replicate it to 10 instance, but only 5 instance up and running and only run on the manager.

from caprover.

Bizarre. I'll be able to look into this in a few hours. Stay tuned.

from caprover.

By the way, do you have any firewall enabled on your manager node. Particularly port 996? You may want to disable your fireball temporarily and retry.

from caprover.

Ok, I'll try to disable ufw for a while.
But do I need to open port 996 in my worker node too?
It seems both of nodes are closed now.

from caprover.

Here is the result after I disable the firewall and try to create mongo-express one click. Neither works.

from caprover.

@nampdn

I just retested the entire thing, and it just works for me. This is my setup/process:

• Get a 1gb server on Digital Ocean
• Setup DNS and do captainduckduck serversetup and successfully accessing captain at captain.do5.mydomain.com
• Get another 1gb server on Digital Ocean
• run ufw disable on both servers
• Attach the second server via Captain dashboard "Nodes" section
• Create a selfhosted docker registry via Captain dashboard
• Wait for 1 minute. Refresh the page and login to Captain to see if correctly see Current Docker Registry: Local at registry.do5.captainduckduck.com:996
• Create a new app.
• IMPORTANT: at this point, if I scale the app, it's gonna fail to create on other nodes.
• I deployed the sample nodejs app to the test app.
• Check if it's working by visiting the link
• Changed the app instance to 4.
• Run docker service ps srv-captain--test and I see it correctly spins up instances on worker node:

root@docker-1gb-nyc3-01:~# docker service ps srv-captain--test 
ID                  NAME                      IMAGE                                                      NODE                        DESIRED STATE       CURRENT STATE             ERROR                              PORTS
iwny8s8co7kx        srv-captain--test.1       registry.do5.captainduckduck.com:996/img-captain--test:1   docker-1gb-nyc3-01          Running             Running 15 seconds ago                                       
tpv7zdzhcabm         \_ srv-captain--test.1   registry.do5.captainduckduck.com:996/img-captain--test:1   docker-1gb-nyc3-01-WORKER   Shutdown            Shutdown 16 seconds ago                                      
vezk6ayqu7p1         \_ srv-captain--test.1   registry.do5.captainduckduck.com:996/img-captain--test     docker-1gb-nyc3-01          Shutdown            Shutdown 52 seconds ago                                      
rtuthnujbybc         \_ srv-captain--test.1   registry.do5.captainduckduck.com:996/img-captain--test     docker-1gb-nyc3-01-WORKER   Shutdown            Rejected 2 minutes ago    "No such image: registry.do5.c…"   
ow1i0yd6prqu         \_ srv-captain--test.1   registry.do5.captainduckduck.com:996/img-captain--test     docker-1gb-nyc3-01-WORKER   Shutdown            Rejected 3 minutes ago    "No such image: registry.do5.c…"   
q1uynplz2bv3        srv-captain--test.2       registry.do5.captainduckduck.com:996/img-captain--test:1   docker-1gb-nyc3-01-WORKER   Running             Running 18 seconds ago                                       
fwju7owg3aiw        srv-captain--test.3       registry.do5.captainduckduck.com:996/img-captain--test:1   docker-1gb-nyc3-01          Running             Running 19 seconds ago                                       
1xd733af3pgj        srv-captain--test.4       registry.do5.captainduckduck.com:996/img-captain--test:1   docker-1gb-nyc3-01-WORKER   Running             Running 18 seconds ago                                       

As you can see above, there are 4 instances running, 2 on main node, 2 on worker node. Can you please re-test?

from caprover.

I tried with different containers and cannot reproduce the issue. Please upgrade to latest version (v0.3.1) and re-open if the issue still exists.

You can update your instance by going to web dashboard > settings > install updates.

from caprover.