smothiki

-1.compute.internal sh[2673]: 10.21.2.85 "POST /v1/auth/login/ HTTP/1.0" 200 52 "python-requests/2.5.1 CPython/2.7.6 Darwin/14.1.0"
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: ERROR Internal Server Error: /v1/auth/cancel
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: Traceback (most recent call last):
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: File "/usr/local/lib/python2.7/dist-packages/django/core/handlers/base.py", line 112, in get_response
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: response = wrapped_callback(request, *callback_args, **callback_kwargs)
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: File "/usr/local/lib/python2.7/dist-packages/django/views/decorators/csrf.py", line 57, in wrapped_view
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: return view_func(*args, **kwargs)
Mar 04 20:41:09 ip-10-21-1-139.us-west-1.compute.internal sh[2673]: File "/usr/local/lib/pyt

smothiki

ERROR k8stest: k8stest_v2.web.1 (create): Failed to retrieve unit: 422 status code 422 - {
Mar 06 03:56:42 sh[8612]: "kind": "Status",
Mar 06 03:56:42 sh[8612]: "creationTimestamp": null,
Mar 06 03:56:42 sh[8612]: "apiVersion": "v1beta1",
Mar 06 03:56:42 sh[8612]: "status": "Failure",
Mar 06 03:56:42 sh[8612]: "message": "Pod \"k8stest_v2.web.1\" is invalid: [metadata.name: invalid value 'k8stest_v2.web.1': name must be lowercase letters and numbers, with inline dashes or periods, spec.containers[0].name: invalid value 'k8stest_v2.web.1']",
Mar 06 03:56:42 sh[8612]: "reason": "Invalid",
Mar 06 03:56:42 sh[8612]: "details": {
Mar 06 03:56:42 sh[8612]: "id": "k8stest_v2.web.1",
Mar 06 03:56:42 sh[8612]: "kind": "Pod",

smothiki

page_title	page_description	page_keywords
Docker Swarm filters	Swarm filters	docker, swarm, clustering, filters

Filters

The Docker Swarm scheduler comes with multiple filters.

smothiki

etcdctl ls --recursive /deis
/deis/cache
/deis/cache/host
/deis/cache/port
/deis/controller
/deis/controller/protocol
/deis/controller/builderKey
/deis/controller/registrationEnabled
/deis/controller/schedulerModule
/deis/controller/image

smothiki

#cloud-config
---
ssh_authorized_keys:
  - <%= ssh_public_key %>
coreos:
  etcd:
    # generate a new token for each unique cluster from https://discovery.etcd.io/new
    # uncomment the following line and replace it with your discovery URL
    discovery: <%= discovery_url %> 
    addr: $private_ipv4:4001

smothiki

docker -H 172.17.8.103:2395 ps
CONTAINER ID        IMAGE               COMMAND                CREATED              STATUS              PORTS               NAMES
b7e510f3c9f9        redis:latest        "/entrypoint.sh redi   About a minute ago   Pending             6379/tcp            deis-03/some-redis3   
e8b0748017ca        redis:latest        "/entrypoint.sh redi   About a minute ago   Pending             6379/tcp            deis-03/some-redis2   
16ee2fe9f663        redis:latest        "/entrypoint.sh redi   16 hours ago         Up 16 hours         6379/tcp            deis-02/some-redis    

deis-02 ~ $ docker ps -a
CONTAINER ID        IMAGE               COMMAND                CREATED             STATUS              PORTS               NAMES

smothiki

docker swarm events

2015-04-22T20:02:30.000000000Z mysql:latest: (from  node:deis-03) pull
2015-04-22T20:02:30.000000000Z f95e051aef40f99ce9bff4e236db8e15b14789f3f079269c235e899b546c1991: (from mysql:5 node:deis-03) create
2015-04-22T20:02:30.000000000Z f95e051aef40f99ce9bff4e236db8e15b14789f3f079269c235e899b546c1991: (from mysql:5 node:deis-03) start
2015-04-22T20:03:55.000000000Z (from swarm node:deis-03) engine_disconnect
2015-04-22T20:04:59.000000000Z mysql:latest: (from  node:deis-02) pull
2015-04-22T20:04:59.000000000Z d90620613576f1cb3cb13bc633913ffeb0707cfaec98848fc98945c947a07e1e: (from mysql:5 node:deis-02) create

smothiki

docker swarm evetns

1. created mysql on deis-03
2. made deis-03 failvoer
3. re schedule mysql on deis-02
4. reconnect deis-03 and remove mysql as it has already been rescheduled

2015-04-22T20:23:35.000000000Z 8fbda0e636db3f8bf26042677a6f618efccfd26d124b91635ca6ba2f0d877eab: (from mysql:5 node:deis-03) create
2015-04-22T20:23:35.000000000Z 8fbda0e636db3f8bf26042677a6f618efccfd26d124b91635ca6ba2f0d877eab: (from mysql:5 node:deis-03) start
2015-04-22T20:24:00.000000000Z (from swarm node:deis-03) engine_disconnect

smothiki

User Forgot to set the config for schedulerModule

1. user ran deisctl install <scheduler> && deisctl start <scheduler> .
2. The scheduler will keep publishing the /deis/controller/schedulerTarget to scheduler endpoint. But still uses fleet with the endpoint of the new scheduler

User Forgot to stop the scheduler

1. if a user forgots to stop scheduler and set the config for schedulerModule to a different scheduler
2. Both the schedulers try to set deis/controller/schedulerTarget and controller keep updating the confd templates accordingly.

smothiki

Consideration

Docker engine should run on port 2375 in each node and manager should be able to access port 2395

etcd keys

1. Swarm uses deis/scheduler/swarm/node to identify other nodes in the cluster
2. Publishes swarm manager host IP to /deis/scheduler/swarm/host
3. set /deis/controller/schedulerModule to swarm to use swarm scheduler for deis

steps to install and start swarm

1. deisctl install swarm