Hi, I'm having issues deploying Airbyte on a windo...
# ask-community-for-troubleshooting
a
Hi, I'm having issues deploying Airbyte on a windows 10 machine in docker desktop. The volumes appear to create properly but it goes into an endless failure loop on startup which appears to be due to the SchedulerApp waiting on the server and timing out.
Copy code
airbyte-server      |   at io.airbyte.server.ServerApp.getServer(ServerApp.java:185) [io.airbyte-airbyte-server-0.35.49-alpha.jar:?]
airbyte-server      |   at io.airbyte.server.ServerApp.main(ServerApp.java:236) [io.airbyte-airbyte-server-0.35.49-alpha.jar:?]
airbyte-scheduler   | 2022-03-11 15:47:10 INFO i.a.s.a.SchedulerApp(waitForServer):220 - Waiting for server to become available...
airbyte-worker      | 2022-03-11 15:47:11 WARN i.t.i.r.GrpcSyncRetryer(retry):56 - Retrying after failure
airbyte-worker      | io.grpc.StatusRuntimeException: DEADLINE_EXCEEDED: Deadline exceeded after 4.998877400s.
airbyte-worker      |   at io.grpc.stub.ClientCalls.toStatusRuntimeException(ClientCalls.java:262) ~[grpc-stub-1.44.1.jar:1.44.1]
👀 1
m
This error happened in Temporal Container. How much resources are you allowing Docker to use? CPU/memory/disk? Are you using WSL2?
a
I am using WSL2. I'm not sure what resources have been allocated - I don't have a .wslconfig file configured so I assume that windows is allocating whatever the default values are but I am having difficulty determining what these values would be. What resources does airbyte require? EDIT: according to the wsl2 docs the memory should be either 8GB or 80% of 16GB depending upon what I assume is the windows build number. So at least 8GB. CPU should be 8 logical and I have 219GB of disk space free.
I have exported the logs for temporal and server. I'm not seeing any errors in the temporal logs except for the initial registration of the default namespace which is then successfully created and registered. Is there a way to attach the logs other than by pasting them into a message?
c
Did this ever get figured out? I'm running into the same issue
a
I wound up getting our sysadmin to deploy into our kubernetes cluster.