Enable custom kubernetes scheduler for a namespace

Question

I have a k8 job that brings up multiple pods. This job is used for load testing so all the pods need to come up at the same time. Job shouldn't be started until nodes are available for all pods to be scheduled. I came across kube-batch https://github.com/kubernetes-sigs/kube-batch to do this scheduling. I have couple of questions:

1. How to enable kube-batch for only one namespace in a cluster?

2. Installed kube-batch by following the tutorial. But pods are failing on startup with below error. How to resolve this error?

I1204 20:07:55.911393       1 allocate.go:96] Queue <default> is overused, ignore it.

I1204 20:07:55.911399       1 allocate.go:194] Leaving Allocate ...

I1204 20:07:55.911407       1 backfill.go:41] Enter Backfill ...

I1204 20:07:55.911413       1 backfill.go:71] Leaving Backfill ...

E1204 20:07:55.911521       1 runtime.go:69] Observed a panic: "invalid memory address or nil pointer dereference" (runtime error: invalid memory address or nil pointer dereference)

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/runtime/runtime.go:76

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/runtime/runtime.go:65

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/runtime/runtime.go:51

/usr/local/go/src/runtime/asm_amd64.s:522

/usr/local/go/src/runtime/panic.go:513

/usr/local/go/src/runtime/panic.go:82

/usr/local/go/src/runtime/signal_unix.go:390

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/pkg/scheduler/framework/session.go:368

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/pkg/scheduler/plugins/gang/gang.go:154

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/pkg/scheduler/framework/framework.go:58

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/pkg/scheduler/scheduler.go:102

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/pkg/scheduler/scheduler.go:85

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:133

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:134

/home/root1/servicecomb/go/src/github.com/kubernetes-sigs/kube-batch/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:88

/usr/local/go/src/runtime/asm_amd64.s:1333

panic: runtime error: invalid memory address or nil pointer dereference [recovered]
        panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x148 pc=0x10ab979]

Answer 1

Not sure what you are trying to achive is doable. In my opinion what you can do is to modify the pods dockerfile to include Supervisord . Then in supervisord specify the commands you want to run when the pods come in running state using priority for supervisord. Example

[program:api]
directory=/usr/local
command=go main.go
priority=100
autostart=true
autorestart=true
stderr_logfile=/var/log
stdout_logfile=/var/log