New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Jun 2023 #10

Closed

vadasambar opened this issue Jun 1, 2023 · 22 comments

Labels

Contributor

vadasambar commented Jun 1, 2023

Last month's thread:
May 2023: #9

What

This is an issue to update daily status on what I am doing as a member of vadafoss community.

How

Follow this format

## <3-letter-month> <2-letter-day> <4-letter-year>
### やったこと
### Problem
### Try
### TODO
### WIP

やったこと: yatta-koto lit. 'things I did' in Japanese (I just find it sounds closer to what I want to say)
Problem (optional): problems I faced
Try (optional): what I am trying
TODO (optional): things to do
WIP (optional): work in progress

The text was updated successfully, but these errors were encountered:

vadasambar added the daily-updates label

Contributor Author

vadasambar commented Jun 1, 2023 •

edited

Loading

June 1 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820

TODO

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 based on Daniel's comment here
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 2, 2023 •

edited

Loading

June 2 2023

やったこと

tweeted about Karpenter applying to be a part of sig-autoscaling: https://twitter.com/_vadasambar/status/1664498144447700993
add comment voicinding support for Karpenter to be a part of sig-autoscaling + concerns
- REQUEST: Migrate github.com/aws/karpenter-core kubernetes/org#4258 (comment)
planning for blogposts
reply to thread I started around adding support for kwok provider in CAPI https://kubernetes.slack.com/archives/C8TSNPY4T/p1685648610609449
- https://kubernetes.slack.com/archives/C8TSNPY4T/p1685682763191499?thread_ts=1685648610.609449&cid=C8TSNPY4T

WIP

TODO

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 based on Daniel's comment here
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 5, 2023 •

edited

Loading

June 5 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - fix: invalid leading UTF-8 octet error during template parsing

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 based on Daniel's comment here

TODO

come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 6, 2023 •

edited

Loading

June 6 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
- fix: use lister to get and list
- fix: did not find nodegroup annotation error
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)

TODO

come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 7, 2023 •

edited

Loading

June 7 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
- fix: CA node recognizing fake nodegroups
- fix: node not getting deleted
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)

TODO

come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 8, 2023 •

edited

Loading

June 8 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - PoC is working

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - try running scale test on local kind cluster
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)

TODO

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - update ticket with screenshots of successful run
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 9, 2023 •

edited

Loading

June 9 2023

やったこと

updated my github profile README
- add contributions from cluster-autoscaler + some restructuring
- update link to my pull requests

WIP

writing a blogpost

TODO

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - try running scale test on local kind cluster
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - update ticket with screenshots of successful run
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 12, 2023 •

edited

Loading

June 12 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - update ticket with screenshots of successful run
    - done: https://github.com/vadafoss/vadafoss/issues/165#issuecomment-1586627541

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - try running scale test on local kind cluster
work on a new kubernetes issue

TODO

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 13, 2023 •

edited

Loading

June 13 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - ran tests on my local kind cluster (I can CA taking around 10m to fully scale-up the cluster to accomodate 5-6k pods)

WIP

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - update CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 with the results of scale test

TODO

work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 14, 2023 •

edited

Loading

June 14 2023

やったこと

look for possible ways to confirm the issue in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769
- implementing kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
  - perform scale tests again to check p99 for scale up (time between CA seeing unschedulable pod and sending a request to the public cloud provider to scale up)
  - update CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 with the results of scale test
    - updated: CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- refactor: add observer interface

WIP

TODO

write first draft of proposal for kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 15, 2023 •

edited

Loading

June 15 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- feat: use scaleStateNotifier in place of clusterstate

WIP

write first draft of proposal for kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820

TODO

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 16, 2023 •

edited

Loading

June 16 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- test: fix syntax errors
tweeted about issue with provisioning 1.26.1 kind clusters: https://twitter.com/_vadasambar/status/1669578962643881984
- shared the PSA on kubernetes slack #cluster-api as well: https://kubernetes.slack.com/archives/C8TSNPY4T/p1686893951962439

WIP

working on a blogpost
reading about wasm

TODO

write first draft of proposal for kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 19, 2023 •

edited

Loading

June 19 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- feat: wip track scale down failures along with scale up failures
write first draft of proposal for kwok provider: feat: implement kwok cloud provider kubernetes/autoscaler#5820
- [AEP] docs: add proposal for kwok provider kubernetes/autoscaler#5869

WIP

TODO

working on a blogpost
reading about wasm
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 20, 2023 •

edited

Loading

June 20 2023

やったこと

reached out to kwok community around the kwok proposal: https://kubernetes.slack.com/archives/C04QPDWGW0M/p1687237015984919
reached out to the sig-scalability community around the kwok proposal: https://kubernetes.slack.com/archives/C09QZTRH7/p1687237346496719
small fixes in kwok proposal
- docs: add better link for PoC implementation
- docs: fix wording as of now -> as of writing this

clean-up kwok proposal and put it in ready for review state
- added comment: [AEP] docs: add proposal for kwok provider kubernetes/autoscaler#5869 (comment)

try reproducing The pod is successfully removed but the configmap volume remains on the node kubernetes/kubernetes#113987
- wasn't able to reproduce: The pod is successfully removed but the configmap volume remains on the node kubernetes/kubernetes#113987 (comment)

WIP

TODO

working on a blogpost
reading about wasm
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 21, 2023 •

edited

Loading

June 21 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- feat: register failed scale down with scale state notifier

WIP

read about cluster-api

TODO

fix CA logs flood AWS provider warning Failed to check cloud provider has instance for fargate is flooding the log kubernetes/autoscaler#5842 (comment)
working on a blogpost
reading about wasm
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 22, 2023 •

edited

Loading

June 22 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- test: wip tests for ScaleDownCandidatesDelayProcessor

WIP

read about cluster-api

TODO

fix CA logs flood AWS provider warning Failed to check cloud provider has instance for fargate is flooding the log kubernetes/autoscaler#5842 (comment)
working on a blogpost
reading about wasm
address comments in [AEP] docs: add proposal for kwok provider kubernetes/autoscaler#5869
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 23, 2023 •

edited

Loading

June 23 2023

やったこと

tweeted about using CAPI with CA: https://twitter.com/_vadasambar/status/1672126279297150977
fix CA logs flood AWS provider warning Failed to check cloud provider has instance for fargate is flooding the log kubernetes/autoscaler#5842 (comment)
- PR: fix: CA on fargate causing log flood kubernetes/autoscaler#5887

WIP

work on a blogpost

TODO

read about cluster-api
reading about wasm
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 26, 2023 •

edited

Loading

June 26 2023

やったこと

CI failing because of flaky tests fix: CA on fargate causing log flood kubernetes/autoscaler#5887
- fixed
reminded sig-acalability about kwok proposal: https://kubernetes.slack.com/archives/C09QZTRH7/p1687803036468249?thread_ts=1687237346.496719&cid=C09QZTRH7

WIP

read about cluster-api

TODO

reading about wasm
work on a blogpost
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 27, 2023 •

edited

Loading

June 27 2023

やったこと

change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
- test: add unit tests for ScaleDownCandidatesDelayProcessor
read about cluster-api
reminded Richard about creating issues for kwok provider in CAPI: https://kubernetes.slack.com/archives/C8TSNPY4T/p1687848500383719?thread_ts=1685648610.609449&cid=C8TSNPY4T

WIP

TODO

reading about wasm
work on a blogpost
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 28, 2023 •

edited

Loading

June 28 2023

やったこと

WIP

create sample cluster-api provider
respond to new review comments in feat: support custom scheduler config (without extenders) kubernetes/autoscaler#5708
review cluster-autoscaler has a panic in NodeDeletionBatcher.AddNode kubernetes/autoscaler#5891

TODO

reading about wasm
work on a blogpost
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 29, 2023 •

edited

Loading

June 29 2023

やったこと

trying out https://github.com/capi-samples/cluster-api-provider-kwok on my local machine
- created PR to fix go build failing: https://github.com/capi-samples/cluster-api-provider-kwok/pull/2/files
create sample cluster-api provider
respond to new review comments in feat: support custom scheduler config (without extenders) kubernetes/autoscaler#5708
- responded and addressed comments:
  - refactor: rename GetSchedulerConfig -> ConfigFromPath

WIP

trying out https://github.com/capi-samples/cluster-api-provider-kwok on my local machine

TODO

review cluster-autoscaler has a panic in NodeDeletionBatcher.AddNode kubernetes/autoscaler#5891
reading about wasm
work on a blogpost
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

Contributor Author

vadasambar commented Jun 30, 2023 •

edited

Loading

June 30 2023

やったこと

reading about wasm

WIP

TODO

try out https://github.com/capi-samples/cluster-api-provider-kwok on my local machine
review cluster-autoscaler has a panic in NodeDeletionBatcher.AddNode kubernetes/autoscaler#5891
reading about wasm
work on a blogpost
run scale kwok scale test again based on the new information in CA scale-up delays on clusters with heavy scaling activity kubernetes/autoscaler#5769 (comment)
work on a new kubernetes issue
change approach for feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729 (use observer based approach)
come up with a way to test feat: support --scale-down-delay-after-* per nodegroup kubernetes/autoscaler#5729
- manually
- in code
create a follow-up ticket to enhance ignore-vol annotation in CA: Ability to ignore some containers (sidecars) when evaluating if a pod is evictable or not kubernetes/autoscaler#3947 (comment)
respond to new comment in cluster autoscaler deleted non-empty node as ScaleDownEmpty kubernetes/autoscaler#5790
check out new comments in Cluster Autoscaler resets unneeded since time to 0s kubernetes/autoscaler#5618
respond to new comment in Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
Cluster autoscaler ignore node taints? kubernetes/autoscaler#4231
- go through the code to understand what might be happening
check Add info about mirror pods in FAQ kubernetes/autoscaler#5566 (comment)
check comment support overprovsioning without pending pods kubernetes/autoscaler#5377 (comment)
check the Kanban

Problem

check the Kanban

vadasambar closed this as completed

vadasambar mentioned this issue

Closed

vadasambar changed the title ~~June 2023~~ Jun 2023

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment