Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Jesse Vander Does
    @FreakTheMighty
    I think so
    It would, clone the repo and copy the configs, is that all?
    Traun Leyden
    @tleyden
    yeah, essentially just automate the manual labor you just went through
    Jesse Vander Does
    @FreakTheMighty
    Cool
    Do you keep the Vagrant file in your elastic-thought project directory, or do you keep it in the core-os repo?
    Traun Leyden
    @tleyden
    I keep mine in ~/Vagrant/core-os (and keep my ubuntu vagrant file under ~/Vagrant/ubuntu)
    Jesse Vander Does
    @FreakTheMighty
    k
    Jesse Vander Does
    @FreakTheMighty
    BTW I’ve played with pouchdb a bit.
    Traun Leyden
    @tleyden
    I've never actually played with it yet
    I lean towards mobile apps rather than web apps
    I mean native mobile apps
    but yeah, I think pouchdb is really cool for html5 dev
    Jesse Vander Does
    @FreakTheMighty
    I haven’t investigated much, but after setting up core os and sharing the elastic-thought source with the vagrant image, I ran ./elasticthought-cluster-init.sh -v 3.0.1 -n 3 -u user:passw0rd -p gpu
    After fetching the images I get Error: Cannot sync with the cluster using peers 127.0.0.1:4001 Kick off couchbase cluster 2015/03/23 03:55:59 Update-Wrapper: skipping update 2015/03/23 03:55:59 Connect to etcd on localhost 2015/03/23 03:55:59 verifyEnoughMachinesAvailable() 2015/03/23 03:55:59 getJsonData error: Get http://localhost:49153/fleet/v1/machines: dial tcp 127.0.0.1:49153: connection refused 2015/03/23 03:55:59 Failed: Get http://localhost:49153/fleet/v1/machines: dial tcp 127.0.0.1:49153: connection refused 2015/03/23 03:55:59 Error invoking target: exit status 1 Failed to kick off couchbase server
    Also, I’m curious, why use vagrant and docker in development? Seems like docker alone should suffice.
    I see now that error comes from starting the couchbase container, I’ll have to look into it more.
    Traun Leyden
    @tleyden

    Error: Cannot sync with the cluster using peers 127.0.0.1:4001

    that's a bad error -- 4001 is the etcd port, so it looks like etcd is not setup for some reasons (and etcd is a core component, without that nothing will work)

    I think before running the elasticthought-cluster-init.sh you should have verified the coreos setup with fleetctl list-machines. I'll add that to the readme
    Traun Leyden
    @tleyden
    try running fleetctl list-machines and paste the output
    Jesse Vander Does
    @FreakTheMighty
    From core-01 the output is core@core-01 ~ $ fleetctl list-machines 2015/03/24 05:20:15 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:15 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 100ms 2015/03/24 05:20:15 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:15 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 200ms 2015/03/24 05:20:15 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:15 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 400ms 2015/03/24 05:20:16 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:16 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 800ms 2015/03/24 05:20:16 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:16 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 1s 2015/03/24 05:20:17 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:17 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 1s 2015/03/24 05:20:18 ERROR fleetctl.go:212: error attempting to check latest fleet version in Registry: timeout reached 2015/03/24 05:20:18 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:18 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 100ms 2015/03/24 05:20:18 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:18 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 200ms 2015/03/24 05:20:18 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:18 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 400ms 2015/03/24 05:20:19 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:19 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 800ms 2015/03/24 05:20:19 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:19 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 1s 2015/03/24 05:20:20 INFO client.go:291: Failed getting response from http://127.0.0.1:4001/: dial tcp 127.0.0.1:4001: connection refused 2015/03/24 05:20:20 ERROR client.go:213: Unable to get result for {Get /_coreos.com/fleet/machines}, retrying in 1s Error retrieving list of active machines: timeout reached
    Jesse Vander Does
    @FreakTheMighty
    Are you saying etcd isn’t setup on my computer, or on the VMs?
    Traun Leyden
    @tleyden
    On the vms, in coreos.
    I think you might have a bad etcd discovery token. In the config.rb there are instructions to uncomment some stuff which will pull a fresh token each time coreos is started.
    (On my phone, will give more details later)
    Jesse Vander Does
    @FreakTheMighty
    Cool, I'll investigate.
    Jesse Vander Does
    @FreakTheMighty
    I've been poking around the repo, how does your go code bind to caffe? Are you shelling out, or are there go cpp bindings?
    Traun Leyden
    @tleyden
    Just shelling out for now
    I was considering rewriting the whole worker in python at some point, since there are nice python bindings
    j0n3
    @j0n3
    Hi there! I want to use this to send an image from an app and get the tags to make my app select a "category" from recogniced image. Is elasticthought suitable for this? Thanks!
    I have a coreOS vm and cloned elasticthought repo from github, but don't know how to get it all up and running
    j0n3
    @j0n3
    My steps are
    1. sudo su
    2. etcd & fleetd &
    3. cd elasticthought
    4. ./elasticthought-cluster-init.sh -v 3.0.1 -n 3 -u user:passw0rd -p cpu
      and this is the output:
      ```core-01 elastic-thought # ./elasticthought-cluster-init.sh -v 3.0.1 -n 3 -u user:passw0rd -p cpu
      true
      Kick off couchbase cluster
      2015/03/26 10:59:53 Update-Wrapper: updating to latest code
      github.com/tleyden/couchbase-cluster-go (download)
      github.com/coreos/fleet (download)
      github.com/coreos/go-systemd (download)
      github.com/tleyden/go-etcd (download)
      github.com/docopt/docopt-go (download)
      github.com/tleyden/couchbase-cluster-go
      github.com/tleyden/couchbase-cluster-go/cmd/couchbase-cluster
      github.com/tleyden/couchbase-cluster-go/cmd/couchbase-fleet
      github.com/tleyden/couchbase-cluster-go/cmd/sync-gw-cluster
      github.com/tleyden/couchbase-cluster-go/cmd/sync-gw-config
      2015/03/26 11:00:17 Connect to etcd on localhost
      2015/03/26 11:00:17 verifyEnoughMachinesAvailable()
      2015/03/26 11:00:17 getJsonData error: Get http://localhost:49153/fleet/v1/machines: dial tcp 127.0.0.1:49153: connection refused
      2015/03/26 11:00:17 Failed: Get http://localhost:49153/fleet/v1/machines: dial tcp 127.0.0.1:49153: connection refused
      2015/03/26 11:00:17 Error invoking target: exit status 1
      Failed to kick off couchbase server
      core-01 elastic-thought #
    ERROR event.go:107: etcd watcher {Watch /_coreos.com/fleet/job} returned error: unexpected end of JSON input
    INFO client.go:291: Failed getting response from http://localhost:4001/: unexpected EOF
    ERROR client.go:213: Unable to get result for {Watch /_coreos.com/fleet/job}, retrying in 100ms
    Traun Leyden
    @tleyden
    @j0n3 you need to enable the fleet api in your cluster config. Not sure how well its documented, but if you check earlier conversation w @FreakTheMighty you might be able to figure it out. I will check the docs, and look into ways to make this easier.
    Traun Leyden
    @tleyden

    @j0n3 does your user-data file look like this? https://github.com/tleyden/elastic-thought#update-cloud-config

    if it does but its still not working, you may need to run vagrant reload --provision

    j0n3
    @j0n3
    2015/03/26 15:04:23 Connect to etcd on localhost
    2015/03/26 15:04:23 Failed to get admin credentials from etc: Error getting key. Err: 100: Key not found (/couchbase.com/couchbase-node-state) [1054]
    2015/03/26 15:04:23 Error invoking target: exit status 1
    Couchbase cluster node:
    Failed to find ip of couchbase node

    I've uncommented this:$new_discovery_url='https://discovery.etcd.io/new'

    To automatically replace the discovery token on 'vagrant up', uncomment

    the lines below:

    #
    #
    if File.exists?('user-data') && ARGV[0].eql?('up')
    require 'open-uri'
    require 'yaml'

    token = open($new_discovery_url).read

    data = YAML.load(IO.readlines('user-data')[1..-1].join)
    data['coreos']['etcd']['discovery'] = token

    yaml = YAML.dump(data)
    File.open('user-data', 'w') { |file| file.write("#cloud-config\n\n#{yaml}") }
    end

    omg text formatting :(
    first of all, thank you for your time :)
    Traun Leyden
    @tleyden

    @j0n3 you have to get past this error:

    Get http://localhost:49153/fleet/v1/machines: dial tcp 127.0.0.1:49153: connection refused

    that means it's trying to connect to the coreos fleet api on port 49153, but fleet isn't listening on that port. I'm guessing because your user-data file is wrong or you need to run vagrant reload --provision

    yeah np!
    btw I'm planning to fork the coreos vagrant repo to make this easier
    j0n3
    @j0n3
    this would be great!
    Traun Leyden
    @tleyden
    yes, you should uncomment that stuff regarding the discovery token
    j0n3
    @j0n3
    I'm new to docker and this deep learning stuff is out of my knowledge...
    my goals are:
    1. getting this up and running.
    2. Learn how this work in order to feed own model database
    3. how to query the API for an image and get its tags
    it is a very interesting project so thank you for sharing!
    Traun Leyden
    @tleyden
    For 2., do you mean you don't want to train your own model database, you just want to stick in a pre-trained model database?
    j0n3
    @j0n3
    the more model database, the less work, isn't it?
    we are running a classified ads site
    Traun Leyden
    @tleyden
    yes it's definitely less work. you'll need to do a little hacking because the elastic-thought api doesn't support that use case yet, but it's possible.