Use unified nginx helm chart #636
Conversation
lianhao
force-pushed
the
unified_nginx
branch
2 times, most recently
from
030fe5b to
4d71f08
Compare
|
The faqgen failure is due to the known issue of opea-project/GenAIComps#969 |
lianhao
force-pushed
the
unified_nginx
branch
2 times, most recently
from
002d680 to
6139cdb
Compare
lianhao
force-pushed
the
unified_nginx
branch
from
6139cdb to
1d16ef4
Compare
|
we need to wait until langserve 0.3.1 is available in pypi to unblock the faqgen CI failure |
- Add helm chart support for OPEA nginx - Use unified nginx chart in E2E Signed-off-by: Lianhao Lu <[email protected]>
lianhao
force-pushed
the
unified_nginx
branch
from
1d16ef4 to
900798e
Compare
|
gaudi CI environment issue: |
| targetCPUUtilizationPercentage: 80 | ||
| # targetMemoryUtilizationPercentage: 80 |
There was a problem hiding this comment.
IMHO these variable names could be a bit shorter.
| targetCPUUtilizationPercentage: 80 | |
| # targetMemoryUtilizationPercentage: 80 | |
| targetCPUPercentage: 80 | |
| # targetMemoryPercentage: 80 |
| # We usually recommend not to specify default resources and to leave this as a conscious | ||
| # choice for the user. If you do want to specify resources, uncomment the following | ||
| # lines, adjust them as necessary, and remove the curly braces after 'resources:'. | ||
| # limits: | ||
| # cpu: 100m | ||
| # memory: 128Mi | ||
| # requests: | ||
| # cpu: 100m | ||
| # memory: 128Mi |
There was a problem hiding this comment.
I've never used cpu/mem metrics for scaling (only application specific custom metrics), but according to docs: https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/#how-does-a-horizontalpodautoscaler-work
CPU/mem resource requests need to be specified for HPA to work on relative CPU/mem targets:
For per-pod resource metrics (like CPU), the controller fetches the metrics from the resource metrics API for each Pod targeted by the HorizontalPodAutoscaler. Then, if a target utilization value is set, the controller calculates the utilization value as a percentage of the equivalent resource request on the containers in each Pod.
...
Please note that if some of the Pod's containers do not have the relevant resource request set, CPU utilization for the Pod will not be defined and the autoscaler will not take any action for that metric.
There was a problem hiding this comment.
So I think that:
requestspart should be enabled, after verifying that current values match "normal" resource usage for specified nginx version- Comment for
requestsvalues should state that "normal" nginx resource usage should be checked and values updated accordingly, before enabling autoscaling - Comment for limits needs to state that they should be enabled only after checking how much nginx needs when it's stressed, and using those values + some extra for growth (increased buffers, possible leakage etc)
There was a problem hiding this comment.
Can we do this in another PR along with issue #643 ?
Description
Add helm chart support for OPEA nginx
Use unified nginx chart in E2E.
Issues
Fixes #635.
Type of change
List the type of change like below. Please delete options that are not relevant.
Dependencies
List the newly introduced 3rd party dependency if exists.
Tests
Describe the tests that you ran to verify your changes.