Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Adrian Cole
    @adriancole
    as it listens on that :P
    web is on 8080 and has the dependency graph view etc
    Adrian Cockcroft
    @adrianco
    Ok. Will fix the 15 issue. Thanks!
    Adrian Cockcroft
    @adrianco
    much java is happening to my machine...., fans on...
    Adrian Cockcroft
    @adrianco
    Turns out i didnt have a java or docker dev env on my laptop. Working on it so i can test locally that i have the right fix, but still at conference...
    Adrian Cockcroft
    @adrianco
    I have zipkin running, and the post succeeds but the ui still shows the three original services, not the loaded set...
    $ curl -vs 192.168.99.100:8080/api/v1/spans -X POST --data @json_metrics/netflixoss_flow.json -H "Content-Type: application/json"
    *   Trying 192.168.99.100...
    * Connected to 192.168.99.100 (192.168.99.100) port 8080 (#0)
    > POST /api/v1/spans HTTP/1.1
    > Host: 192.168.99.100:8080
    > User-Agent: curl/7.43.0
    > Accept: */*
    > Content-Type: application/json
    > Content-Length: 730631
    > Expect: 100-continue
    > 
    < HTTP/1.1 100 Continue
    * We are completely uploaded and fine
    < HTTP/1.1 200 OK
    < Content-Type: text/html
    < Content-Length: 7518
    < 
    <!DOCTYPE html>
    <html lang='en'>
      <head>
        <meta charset='utf-8'>
        <meta http-equiv='X-UA-Compatible' content='IE=edge'>
    and so on...
    Adrian Cole
    @adriancole
    @adrianco smoking gun is 200
    http 200
    so looks like you posted to the UI endpoint
    it certainly shouldn't reply to a post
    well not with a 200 anyway
    try port 9411
    Adrian Cole
    @adriancole
    warning!
    I forgot that zipkin only looks back a day
    so if you use data who's timestamps are older than a day ( like the ones in git )
    well, they won't show up
    raising a bug as this is a UI problem
    Adrian Cockcroft
    @adrianco
    It works now, using port 9411 on fresh data
    thanks!
    Adrian Cole
    @adriancole
    fantastic
    Adrian Cockcroft
    @adrianco
    I'm going to talk about the zipkin work at RICON next week
    How can I clear out the display to load a fresh set of traces?
    docker-compose restart seems a bit heavy, and the built-in trace always appears
    Adrian Cockcroft
    @adrianco
    the POST should really clear everything before it loads the file
    Adrian Cole
    @adriancole
    @adrianco so you have an mysql server underneath
    oh
    so using latest dockery goodness
    removing the TRANSPORT_TYPE env from web and query will stop self-tracing
    ex. remove this - TRANSPORT_TYPE=http
    then after you are running docker-compose -f docker-compose-mysql.yml up
    you'll have a mysql node, which you can clear the data from
    I'll give you a one-liner in a few
    Adrian Cole
    @adriancole
    mysql -h 192.168.99.100 -u zipkin -pzipkin -D zipkin -e 'delete from zipkin_spans; delete from zipkin_annotations'
    any mysql client should work, ex brew install mysql
    Adrian Cockcroft
    @adrianco
    ETOOMANYTHINGS

    $ brew install mysql
    ==> Downloading http://cdn.mysql.com/Downloads/MySQL-5.6/mysql-5.6.20.tar.gz

    curl: (22) The requested URL returned error: 404 Not Found

    Adrian Cole
    @adriancole
    @adrianco try brew update?
    brew rusts :P
    Henri van den Bulk
    @hvandenb
    What's the best way of building the project with go when you have forked the project? It seems to have issues with find the packages.
    Henri van den Bulk
    @hvandenb
    My go build keeps failing with the packages local. I can use go get to get the packages
    @adrianco when I fork and make changes to some of the tooling it's not finding the packages. I end up using go get but that gets the packages from GitHub. How do you build?
    Adrian Cockcroft
    @adrianco
    Hi @hvandenb, sorry but I'm not sure what to do here. I use go get for all the other packages I need, and after forking them changed the import to point at the forked version until I could get a change committed.
    Henri van den Bulk
    @hvandenb
    @adrianco thanks I'll try that path, thanks for the tip. Want to make a number of enhancements in the neo4j and graph side
    Adrian Cockcroft
    @adrianco
    I appreciate the input, what are you planning to do?
    Henri van den Bulk
    @hvandenb
    Thinking about
    Bas van Beek
    @basvanbeek
    @hvandenb best way to work with forks is to do a regular go get on the master repository, go into the root dir of the repo and do a git remote add on the fork and pull in the changes.
    this way you do not need to rewrite import paths
    Bas van Beek
    @basvanbeek
    if wanting to work on the project itself you do the same. you can then push the changes you make to your fork and work with your fork until your merge request has been approved after which you pull from origin to get it up to date
    if working various angles you can split up in separate branches... all while keeping the repo in the correct gopath