Helm Chart 0.10.1 not working, no good logs
tirelibirefe opened this issue · 5 comments
tirelibirefe commented
Bug Description
fides Helm Chart 0.10.1 doesn't work.
Steps to Reproduce
- Install Fides as described on the procedure.
2.Use existing postgres
3.Use existing Redis
4.Not enable SSL on ingress; terminate SSL on ALB.
5.Use ClusterIP instead of NodePort
6.Install 1 or 2 worker nodes.
7.Use ACK for S3
Expected behavior
Expecting to see a working Fides.
Screenshots
Environment
- Version: kubernetes 1.24
Additional context
Logs:
k logs -n fides worker-fides-757c9cbcb7-h527v
> Loaded config from: /etc/fides/config/fides.toml
2023-04-11 11:52:12.252 | INFO | fides.api.ops.tasks:_create_celery:47 - Creating Celery app...
2023-04-11 11:52:12.255 | INFO | fides.api.ops.tasks:_create_celery:65 - Autodiscovering tasks...
/root/.local/lib/python3.10/site-packages/snowflake/connector/options.py:96: UserWarning: You have an incompatible version of 'pyarrow' installed (6.0.1), please install a version that adheres to: 'pyarrow<8.1.0,>=8.0.0; extra == "pandas"'
warn_incompatible_dep(
k logs -n fides fides-7744b7f854-djbx6 ─╯
> Loaded config from: /etc/fides/config/fides.toml
2023-04-11 11:51:10.325 | INFO | fides.api.ops.tasks:_create_celery:47 - Creating Celery app...
2023-04-11 11:51:10.433 | INFO | fides.api.ops.tasks:_create_celery:65 - Autodiscovering tasks...
tirelibirefe commented
I tried following combinations:
deployPostgres: false # (used aws-rds)
deployRedis:false # (used external redis)
no luck.
deployPostgres: true
deployRedis:false # (used external redis)
no luck.
deployPostgres: false # (used aws-rds)
deployRedis:true
no luck.
deployPostgres: true
deployRedis: true
no luck.
All I get is that:
Loaded config from: /etc/fides/config/fides.toml
2023-04-11 19:41:48.756 | INFO | fides.api.ops.tasks:_create_celery:47 - Creating Celery app...
2023-04-11 19:41:48.771 | INFO | fides.api.ops.tasks:_create_celery:65 - Autodiscovering tasks...
/root/.local/lib/python3.10/site-packages/snowflake/connector/options.py:96: UserWarning: You have an incompatible version of 'pyarrow' installed (6.0.1), please install a version that adheres to: 'pyarrow<8.1.0,>=8.0.0; extra == "pandas"'
tirelibirefe commented
Also tried
workers:
count: 0
the log is that:
Loaded config from: /etc/fides/config/fides.toml
2023-04-11 19:49:54.523 | INFO | fides.api.ops.tasks:_create_celery:47 - Creating Celery app...
2023-04-11 19:49:54.623 | INFO | fides.api.ops.tasks:_create_celery:65 - Autodiscovering tasks...
no luck.
tirelibirefe commented
also tried "dev" and "2.9.2" image tags, same log and same result.
RobertKeyser commented
Hello! Thanks for opening this issue. Are you able to share your values.yaml or the command used to install the Helm chart?
I'm also seeing several OOMKilled
pods, which indicates that they have run out of available memory. Which Kubernetes platform are you running it on (e.g. EKS, AKS, MiniKube, etc)?
tirelibirefe commented
Hello @RobertKeyser
Thanks for your reply.
Good catch! I disabled resource limits and Fides works now.
Thank you