Improper handling on CPU request/limit in dashboard

[amisevsk]

Describe the bug

When using devfiles that specify certain CPU limits/requests, Che returns invalid values for the CPU resource after adding the limit/request for the injected editor. Starting a workspace with a memory request of 4000m, for example, results in the injected devfile requesting 4030.0000000000005m instead of 4030m. This occurs because the injected CPU request is 30m, which internally gets parsed to

(4.0 + 0.03) * 1000

The CPU virtual resource in Kubernetes uses m as the smallest unit, so fractional m values are invalid.

Che version

7.56@latest

Steps to reproduce

1. Start workspace from devfile https://gist.githubusercontent.com/l0rd/a200cc8a036a3564278e9bdb3e44cfe1/raw/2a9a7d1fb8275ce0c72407f3eed73d3fe4b03f2e/big-workspace.yaml
2. After workspace is applied to the cluster, check the devfile in the dashboard for an invalid CPU request value.

Expected behavior

CPU request should be floored before returning to avoid fractional parts.

Runtime

other (please specify in additional context)

Screenshots

No response

Installation method

chectl/latest

Environment

Linux

Eclipse Che Logs

No response

Additional context

No response

[l0rd]

I think this was reported downstream and originally planned to be addressed with the introduction of container contributions and deprecation/removal of the handler (but that's a good news that we already have a fix for it since contributions are not here yet).

[che-bot]

Issues go stale after 180 days of inactivity. lifecycle/stale issues rot after an additional 7 days of inactivity and eventually close.

Mark the issue as fresh with /remove-lifecycle stale in a new comment.

If this issue is safe to close now please do so.

Moderators: Add lifecycle/frozen label to avoid stale mode.

[l0rd]

Already fixed