Prevent scaling of cluster if count / resources exceed account resource limits #12167
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## 4.22 #12167 +/- ##
============================================
- Coverage 17.56% 17.56% -0.01%
- Complexity 15545 15546 +1
============================================
Files 5910 5910
Lines 529123 529164 +41
Branches 64627 64636 +9
============================================
+ Hits 92937 92940 +3
- Misses 425733 425771 +38
Partials 10453 10453
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
|
@Pearl1594 IMHO, resource limitation could be one of the reasons what cause CKS scaling to fail |
| } | ||
| } | ||
|
|
||
|
|
| @Inject | ||
| ResourceLimitService resourceLimitService; |
There was a problem hiding this comment.
the sonar warning here is new to me. If we need to deal with this that is going to be a huge effort as this pattern is all over the project.
I don't quite understand what you mean @weizhouapache - what issues and what change in process are we talking about. Could you please give some clarity on that. Thanks. |
I was thinking other possible edge cases, for example, the 2nd vm is not deployed due to the capacity of the system/storage. |
weizhouapache
left a comment
weizhouapache
left a comment
There was a problem hiding this comment.
code lgtm
not tested yet
3 participants
Description
This PR fixes: #12123
This pr prevents the scaling of the cluster by checking if the overall resources / vm count with the accounts resource limits to prevent cluster in an incorrect state.
Types of changes
Feature/Enhancement Scale or Bug Severity
Bug Severity
Screenshots (if appropriate):
How Has This Been Tested?
How did you try to break this feature and the system with this change?
Before FIx:
Set account's resource limit for VM count to 3
Deployed a CKS cluster with 1 worker node and then attempted to scale it to 3 and it fails
it scales by 1 node but increases the overall size of the cluster to 4 and puts the cluster in an incorrect state
After fix:
Performed the same test as above, here it prevents scaling altogether as it pre-emptively calculates if the overall resources exceeds what's set for the account.