Is your feature request related to a problem? Please describe.
I have created a kubernetes cluster with 2 nodes, each with 2vCPU + 2GB memory.
the cluster is able to debug, and I can access the dashboard.
However when I try to build a service, it got hanged at the PENDING state forever.
After investigation, the workflow pods cannot be started and remains at unscheduled state.
It remains the same after I added another 2GB node to the cluster, but got resolved after I add a new 4GB node to the node-pool.
Suspect that the build workflow request more than 2GB of memory.
Describe the solution you'd like
able to build with less memory, especially with smaller node type.
Would be nice if the memory required for the build workflow is configurable.
Is your feature request related to a problem? Please describe. I have created a kubernetes cluster with 2 nodes, each with 2vCPU + 2GB memory. the cluster is able to debug, and I can access the dashboard. However when I try to build a service, it got hanged at the
PENDING
state forever. After investigation, the workflow pods cannot be started and remains atunscheduled
state. It remains the same after I added another 2GB node to the cluster, but got resolved after I add a new 4GB node to the node-pool.Suspect that the build workflow request more than 2GB of memory.
Describe the solution you'd like able to build with less memory, especially with smaller node type. Would be nice if the memory required for the build workflow is configurable.