-
Notifications
You must be signed in to change notification settings - Fork 0
Add observations of mid-deploy scale-out events #12
Conversation
699231f
to
d65cfb1
Compare
d65cfb1
to
811b9d7
Compare
In this test we deployed [application version](../dist) `ABC` to `500` in the stack | ||
[`ScalingAsgRollingUpdate`](../packages/cdk/lib/scaling-asg-rolling-update.ts) (CFN stack `playground-CODE-scaling-asg-rolling-update`). | ||
|
||
The aim of this test is to understand the behaviour in the ["Ophan scenario"](https://github.com/guardian/riff-raff/issues/1342). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Minor: I don't think this aim is applicable in this scenario?
The aim of this test is to understand the behaviour in the ["Ophan scenario"](https://github.com/guardian/riff-raff/issues/1342). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmm, not sure I agree. Any deployment can fail, and we can scale at any time. To me, "Ophan scenario" is "we need to scale during deployment", there's no restriction on if the artifact being deployed is good or bad.
TBH I was tempted to merge the two documents together, but it didn't read very clearly.
The aim of this test is to understand the behaviour in the ["Ophan scenario"](https://github.com/guardian/riff-raff/issues/1342). | ||
|
||
## Highlights | ||
The deployment leaves the service over-capacity. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I guess this is similar to the findings in https://github.com/guardian/testing-asg-rolling-update/blob/main/observations/healthy-to-unhealthy.md#highlights (but at least in this case we know it will eventually scale down automatically!)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These docs are very clear (even if the test results aren't quite what we were hoping for!) 👍
What does this change?
Document the behaviour of scale-out events mid-deployment, i.e. the Ophan scenario.