Description of problem:
When using a round number, pod quota is shortened (i.e. 1000 -> 1k), but this is consequently treated as "1", chopping off the k.

Version-Release number of selected component (if applicable):
3.3, 3.4. Unconfirmed in 3.5. Fixed by 3.6

How reproducible:
Confirmed

Steps to Reproduce:
1. Create a quota:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: compute-resources
spec:
  hard:
    pods: "1000" 

2. Scale a deployment from 1 to 2


Actual results:
"Quota limit reached"

Expected results:
I expect about 999 more pods to be schedulable

Additional info:
I thought a bug was opened on this and fixed for 3.6 but I cannot locate it using search terms "openshift pod quota"