Fork me on GitHub
#onyx
<
2018-11-15
>
lucasbradstreet01:11:06

@lutz.buech might be worth checking the code to see if it’s setting the region even if the endpoint is set. I know someone had it working with minio in the past

Lutz13:11:59

It turns out it doesn't have to do with regions or bucket names. Changing from localhost to 127.0.0.1 got rid of the misleading error message. Kudos to @dave.dixon!

lucasbradstreet01:11:27

@lellis which computing tree are you referring to?

lellis01:11:20

@lucasbradstreet all tasks of workflow. [in….out]

lellis03:11:30

when i mean tree of computation i mean all job done.

lellis03:11:30

Reading the doc about ABS i understand that i can't control time out of a restart of an input, but this happens in my code, so how can i treat this? Or at least capture when its happens this.

manuel10:11:50

hi everybody. I am trying to run this locally: https://github.com/onyx-platform/onyx-examples/tree/0.14.x/lifecycles I have Zookeeper running via docker with: docker run --name some-zookeeper --restart always -d zookeeper -p 2181:2181 But I get a ConnectionLoss error:

org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /onyx
    code: -4
    path: "/onyx"
Is there something I am missing?

manuel10:11:43

This is the Zookeeper I am using, if it helps: https://hub.docker.com/_/zookeeper/

manuel11:11:50

If I leave out -p 2181:2181, I see:

❯ docker ps
CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS                          NAMES
30e63d362ba9        zookeeper           "/docker-entrypoint.…"   2 seconds ago       Up 1 second         2181/tcp, 2888/tcp, 3888/tcp   some-zookeeper
Running the example I now get:
❯ clj -A:run
[]

manuel11:11:14

these are the relevant changes I made in core.clj:

(def env-config
  {:zookeeper/address "127.0.0.1:2181"
   :zookeeper/server? true
   :zookeeper.server/port 2181
   :onyx/tenancy-id id})

(def peer-config
  {:zookeeper/address "127.0.0.1:2181"
   :onyx/tenancy-id id
   :onyx.peer/job-scheduler :onyx.job-scheduler/balanced
   :onyx.messaging/impl :aeron
   :onyx.messaging/peer-port 40200
   :onyx.messaging/bind-addr "localhost"})

Lutz13:11:59

It turns out it doesn't have to do with regions or bucket names. Changing from localhost to 127.0.0.1 got rid of the misleading error message. Kudos to @dave.dixon!