Account/domain limits stay cached forever

[hrak]

/kind bug

What steps did you take and what happened:

The CloudStack client in pkg/cloud is using https://github.com/jellydator/ttlcache/ to cache certain data fetched from the CS API. When the client connects for the first time, it fetches account/domain limits and stores them in the TTL cache.

The ttlcache package is configured with SkipTtlExtensionOnHit(false), meaning that as long as the cached account/domain limits are being hit by the client, their TTL never expires.

This is causing problems when f.e. a domain is out of a certain resource (CPU cores/memory f.e.) on initial hit & cache of the account/domain limits, and the operator frees some resources in the domain. CAPC will not pick up the newly freed/available resources, causing VMs to never start up.

This could easily be solved by setting SkipTtlExtensionOnHit to true.

What did you expect to happen:

CAPC should cache account/domain limits only for a certain time, not extend the TTL upon cache hit.

Anything else you would like to add:

I will address this with a PR shortly.

Environment:

• Cluster-api-provider-cloudstack version:
• Kubernetes version: (use kubectl version):
• OS (e.g. from /etc/os-release):

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale