Cello cannot add working host. about cello-archived

[yeasy]

1. Cello will add a host, and treat these host as docker driver.
2. Cello will create the fabric cluster automatically.
   Hence, your 2nd step is meaningless. Just setup another vm as compute node (docker installed with fabric images inside), then add its ip:2375 through cello's dashboard.

[yeasy]

I updated the deployment doc and clarify the architect, hope it can help you.

https://github.com/yeasy/cello/blob/master/docs/deployment.md

Thanks!

[WhiteCrow]

Hi，Yeasy

Thank you very much for your update !

I have another question for you.

1st, I setup another VM (Ubuntu 16.04) that use VirtualBox and bridge pattern network.
2nd, I installed docker with fabric images.
3th, I add
DOCKER_OPTS="$DOCKER_OPTS -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock --api-cors-header='*' --default-ulimit=nofile=8192:16384 --default-ulimit=nproc=8192:16384"
in /etc/default/docker and setup Firewall like deployment document in VM.

Cello logs error yet like this.

dashbard      | [2016-11-28 11:27:00,637] DEBUG [resources.host] [host.py:57          host_create()] - name=Test, daemon_url=10.0.1.12:2375, capacity=1fillup=false, schedulable=false, log=local/
dashbard      | [2016-11-28 11:27:00,637] DEBUG [modules.host] [host.py:67               create()] - Create host: name=Test, daemon_url=10.0.1.12:2375, capacity=1, log=local/, autofill=false, schedulable=false
dashbard      | [2016-11-28 11:27:00,643] ERROR [agent.docker_swarm] [docker_swarm.py:132         check_daemon()] - Exception in check_daemon Error while fetching server API version: HTTPConnectionPool(host='10.0.1.12', port=2375): Max retries exceeded with url: /version (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7fc489b82860>: Failed to establish a new connection: [Errno 111] Connection refused',))
dashbard      | [2016-11-28 11:27:00,643] WARNING [modules.host] [host.py:83               create()] - The daemon_url is inactive:tcp://10.0.1.12:2375
dashbard      | [2016-11-28 11:27:00,645] ERROR [agent.docker_swarm] [docker_swarm.py:159   detect_daemon_type()] - Error while fetching server API version: HTTPConnectionPool(host='10.0.1.12', port=2375): Max retries exceeded with url: /version (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7fc489ea8f28>: Failed to establish a new connection: [Errno 111] Connection refused',))
dashbard      | [2016-11-28 11:27:00,645] ERROR [agent.docker_swarm] [docker_swarm.py:238 setup_container_host()] - Invalid host_type=None
dashbard      | [2016-11-28 11:27:00,646] WARNING [modules.host] [host.py:89               create()] - Test cannot be setup
dashbard      | [2016-11-28 11:27:00,646] WARNING [resources.host] [host.py:75          host_create()] - Failed to create host Test
dashbard      | 172.19.0.6 - - [28/Nov/2016 11:27:00] "POST /api/host HTTP/1.0" 400 -
dashbard      | [2016-11-28 11:27:02,691] DEBUG [resources.index] [utils.py:71        request_debug()] - path=/, method=GET
dashbard      | [2016-11-28 11:27:02,691] DEBUG [resources.index] [utils.py:72        request_debug()] - request args:
dashbard      | [2016-11-28 11:27:02,691] DEBUG [resources.index] [utils.py:75        request_debug()] - request form:
dashbard      | [2016-11-28 11:27:02,691] DEBUG [resources.index] [utils.py:78        request_debug()] - request raw body data:
dashbard      | [2016-11-28 11:27:02,692] DEBUG [resources.index] [utils.py:79        request_debug()] - b''
dashbard      | [2016-11-28 11:27:02,692] DEBUG [resources.index] [utils.py:80        request_debug()] - None
dashbard      | [2016-11-28 11:27:02,693] DEBUG [modules.cluster] [cluster.py:46                 list()] - List all active clusters
dashbard      | [2016-11-28 11:27:02,694] DEBUG [modules.cluster] [cluster.py:50                 list()] - List all released clusters
dashbard      | [2016-11-28 11:27:02,696] DEBUG [modules.cluster] [cluster.py:46                 list()] - List all active clusters
dashbard      | [2016-11-28 11:27:02,697] DEBUG [modules.cluster] [cluster.py:46                 list()] - List all active cluster

Is it a permission problem or network problem ?

[yeasy]

aha, have u restart the docker service after 3rd step?

please first make sure the manager can access worker nodes through port 2375.

[kid143]

@WhiteCrow I think your problem is that your swarm might not be properly configured. 2375 is the default port swarm manager listening. Make sure the host ip is correct and the swarm manager is running.

[WhiteCrow]

Thank you ! @yeasy and @kid143

@kid143 help me resolve the problem.

In ubuntu 16.04, before edit config file /etc/default/docker, we need create file /etc/systemd/system/docker.service.d/override.conf and write env variables like that:

[Service]
EnvironmentFile=-/etc/default/docker
ExecStart=
ExecStart=/usr/bin/dockerd -H fd:// $DOCKER_OPTS

After updated /etc/systemd/system/docker.service.d/override.conf and /etc/default/docker, run
sudo systemctl daemon-reload
and
sudo systemctl restart docker.service
It fixed the problem.

Thank you again! yeasy.

[kid143]

Add some info on this issue.

From Ubuntu 16.04, systemd is used to manage all daemon service including docker engine.
But the original docker engine service unit do not define enough start options(listening port etc).

So you have to override the original service unit by adding /etc/systemd/system/docker.service.d/override.conf.

The DOCKER_OPTS environment variable must contain the options you use.

[yeasy]

@WhiteCrow
I'm collecting the adoption cases of Cello, and plan to add into the documentation.
Could you help provide some information of your deployment, like:

• Number of nodes
• Number of chains
• User scenario