https://flyte.org
Join Slack
Hi, everyone! Any tips to help debugging Helm errors when trying to start the new sandbox lite? I ca...
# announcements
s
Hi, everyone! Any tips to help debugging Helm errors when trying to start the new sandbox lite? I can only see
Copy code
Starting Docker daemon...
Done.
Starting k3s cluster...
Done.
Deploying Flyte...
"flyteorg" has been added to your repositories
Release "flyte" does not exist. Installing it now.
Error: timed out waiting for the condition
g
which sandbox are you using ? 
sandbox or demo
s
I'm using demo, and starting it using Docker, not flytectl
Here's my Docker command:
Copy code
docker run -d --name flyte-sandbox --privileged --mount type=bind,src=.,dst=/root -p 30080-30082:30080-30082 -p 30084:30084 -p 30086:30086 -p 30088:30088 -p 30089:30089 <http://cr.flyte.org/flyteorg/flyte-sandbox-lite:sha-5570eff6bd636e07e40b22c79319e46f927519a3|cr.flyte.org/flyteorg/flyte-sandbox-lite:sha-5570eff6bd636e07e40b22c79319e46f927519a3>
g
Can you give me your k8s state while running demo sandbox ?
Copy code
docker exec -it flyte-sandbox /bin/sh
k3s kubectl get pods -n flyte
s
one sec, will do!
I don't know if this is relevant, but I'm trying to run the sandbox on Windows. The old image, 
flyte-sandbox
, started normally, but I'm having trouble with sandbox-lite
g
Not sure but we can check. Can you post the log here ?
s
sorry for taking so long
message has been deleted
g
Can you check why are you getting ImagePullBackOff in k8s-dashboard ?
s
logs.txt
logs.txt
i think its just taking a bizarre amount of time
4min to pull the postgres image, while on a linux machine it takes 1min30
so a timeout happens
g
cc: @high-park-82026
s
hey, just an update: it really was a timeout. we "fixed" it by injecting the --timeout flag at the entrypoint of the sandbox:
Copy code
sed -i "/^helm upgrade/ s/$/ --timeout 20m0s/" /flyteorg/bin/flyte-entrypoint.sh && tini flyte-entrypoint.sh
g
That’s Great but it’s strange that image pull is taking a lot of time on Windows
s
Docker on Windows is very unstable... The other day, we were unable to start a task because there was no resources available. After some debugging, we found out that there's a bug on WSL2 that makes it so that it doesn't free resources; WSL2 was using 12Gb of RAM and not freeing up any of it...
👍 1
164 Views
Open in Slack
PreviousNext
Powered by

Flyte

Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Powered by