Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Chanwit Kaewkasi
    @chanwit
    Active mode: then put agent there
    Then Pause mode: agent still running
    Then Down to upgrade, agent will be gone
    Then we need to set managers to be Active and start the agent again
    Thank you @megastef it should be working now
    @megastef would you like me to hammer again with 60,000 containers?
    Stefan Thies
    @megastef
    Sure, nodes counter shows 533 again :)
    Could you explain to me what "active" mode means? Why was agent removed from masters ..
    Chanwit Kaewkasi
    @chanwit
    sure
    to prevent managers to go berserk, the document recommends that we should put managers into the Drain mode.
    The drain mode is "docker node update --availability drain mg0"
    with this all containers will be gradually remove from the mode until the node "drained"
    This "Drain" mode won't work for our global-service agent.
    That's what we have to set the manager to be "Active" first, which is the default mode.
    Stefan Thies
    @megastef
    Thx. Got it.
    Chanwit Kaewkasi
    @chanwit
    After that to prevent all managers to go berserk, they should be put into the "Pause" mode rather than "Drain".
    The problem is that when the manager goes down, the agent is gone. And because the manager is "Paused", the agent will so disappear.
    This is our discovery today :-)
    TomWillFixIT
    @tomwillfixit_twitter
    :+1:
    Chanwit Kaewkasi
    @chanwit
    it's amazing that we've got 60,000 tasks with just 530 nodes from @tomwillfixit_twitter and @toughIQ_twitter
    TomWillFixIT
    @tomwillfixit_twitter
    Happy to help
    I can't take too much credit @demonware have been great in supporting docker projects over the past 3 years
    any tweets, mentions are appreciated ;)
    Christian Tawfik
    @toughIQ_twitter
    always a pleasure to help :-)
    Chanwit Kaewkasi
    @chanwit
    tweeted
    Stefan Thies
    @megastef
    14.8k container create events in 30 seconds :) Nice rate.
    TomWillFixIT
    @tomwillfixit_twitter
    thanks
    Stefan Thies
    @megastef
    @chanwit are you going to run another workoad?
    Chanwit Kaewkasi
    @chanwit
    yes, sure
    can we @tomwillfixit_twitter
    ?
    Chanwit Kaewkasi
    @chanwit
    $ docker network inspect mynet
    [
        {
            "Name": "mynet",
            "Id": "94gorr7osj6rmqperkkhod5hm",
            "Scope": "swarm",
            "Driver": "overlay",
            "EnableIPv6": false,
            "IPAM": {
                "Driver": "default",
                "Options": null,
                "Config": [
                    {
                        "Subnet": "10.3.3.0/20",
                        "Gateway": "10.3.0.1"
                    }
                ]
            },
            "Internal": false,
            "Containers": null,
            "Options": {
                "com.docker.network.driver.overlay.vxlanid_list": "257"
            },
            "Labels": null
        }
    ]
    here we go
    TomWillFixIT
    @tomwillfixit_twitter
    yeah work away :)
    I'd like to shut the nodes down around 5pm if we are done for the day?
    Chanwit Kaewkasi
    @chanwit
    how many hours left for me?
    $ docker service ps mysql-galera
    ID                         NAME            IMAGE                                  NODE              DESIRED STATE  CURRENT STATE                 ERROR
    71q4xp3dfu6lbn2ieew5b77sj  mysql-galera.1  perconalab/percona-xtradb-cluster:5.6  ip-172-31-21-240  Running        Running 26 seconds ago
    47f15zracfbf75p0f6t0cm82x  mysql-galera.2  perconalab/percona-xtradb-cluster:5.6  ip-172-31-21-207  Running        Preparing about a minute ago
    3ynpzxp8ewgkvez3brngt4q83  mysql-galera.3  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-156  Running        Running 59 seconds ago
    cggu3s3uwf2wh4gxried793lw  mysql-galera.4  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-229  Running        Running 56 seconds ago
    7pss1mgib1scp7cb2u43eoh00  mysql-galera.5  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-213  Running        Running 57 seconds ago
    1dnz9vloylqanhstzf8j6e0d8  mysql-galera.6  perconalab/percona-xtradb-cluster:5.6  ip-172-31-22-6    Running        Running 18 seconds ago
    2nbd2jjnr3efa7a98b9hfxg72  mysql-galera.7  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-65   Running        Running 32 seconds ago
    239vr9kmslhgidozh8tjv4ezt  mysql-galera.8  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-119  Running        Running about a minute ago
    9av9ah8psc4oqu5qiflxrzcfp  mysql-galera.9  perconalab/percona-xtradb-cluster:5.6  ip-172-31-20-154  Running        Running about a minute ago
    TomWillFixIT
    @tomwillfixit_twitter
    oh that's GMT so another hour
    you need longer?
    Chanwit Kaewkasi
    @chanwit
    no no
    it's going to be done by 15 minutes
    MySQL setup
    TomWillFixIT
    @tomwillfixit_twitter
    cool
    Chanwit Kaewkasi
    @chanwit
    and WordPress is starting to scale
    TomWillFixIT
    @tomwillfixit_twitter
    we have any erlang based services we can try scaling?
    Chanwit Kaewkasi
    @chanwit
    sure !
    have you got it on docker hub?
    TomWillFixIT
    @tomwillfixit_twitter
    i will by next week :)
    Chanwit Kaewkasi
    @chanwit
    I see :-)
    Chanwit Kaewkasi
    @chanwit
    @tomwillfixit_twitter during the real run, could you use "dw-" as your machine prefixes ?