We recently moved our staging and preview environments to Scaleway. We noticed that now we see increased latency to our application if any environment on the cluster is deploying. We are using dedicated CPU instances. We did not notice this behavior when we were deployed on AWS.
Hi @prki ,
What Qovery install on cloud provider regarding the elements that could make a difference in term of performance are exactly the same. So I don’t think it could come from this point.
However, compared to AWS, Scaleway does not manage the autoscaling of the load balancer. By default, we set the smallest one, the most cost-effective one.
I think you can give it a try by changing the Load Balancer size in the cluster’s advanced settings (Cluster Advanced Settings | Docs | Qovery). You’ll find this load_balancer.size option, and you’ll have to update to a bigger one proposed by Scaleway:
Please let us know if it helps.
Thanks
The environment is not getting a lot of traffic so the load balancer solution seems like a long shot but we will try it. I can’t produce any charts at the moment because we were unable to install Datadog on that cluster (issue reported here).
We upgraded the load balancer but the issue persists. Can’t produce charts at the moment, but it happens frequently so it’s pretty easy to reproduce on a page like this. It may not be related to deployments after all but happens intermittently. The issue was not visible on a previous cluster on AWS.
We triggered node replacement in the cluster and the issue seems to have been resolved. We will track if it comes back over time.