User should be warned in OSIO UI when OpenShift limits are reached

[joshuawilson]

From @ldimaggi on July 10, 2017 2:30

In experimenting with a new (and empty) user account:

• I am seeing a predictable pattern in that a user can create 2 projects - depending on the characteristics of each project - for example - if a user creates a Vert.X project and a Spring Boot Healthcheck project. the stage and run resources look like this:

oc describe quota compute-resources -n osiotest3142-stage
Name:		compute-resources
Namespace:	osiotest3142-stage
Scopes:		NotTerminating
 * Matches all pods that do not have an active deadline.
Resource	Used	Hard
--------	----	----
limits.cpu	2	2
limits.memory	1Gi	1Gi

• If that user creates ONE Che workspace, however, the Che related workspaces reach the quota:

oc describe quota compute-resources -n osiotest3142-che
Name:		compute-resources
Namespace:	osiotest3142-che
Scopes:		NotTerminating
 * Matches all pods that do not have an active deadline.
Resource	Used	Hard
--------	----	----
limits.cpu	5859m	6
limits.memory	3000Mi	3Gi

• Warnings are made visible in OpenShift console - but not in the OSIO UI.

Copied from original issue: fabric8-ui/fabric8-ui#1681

[joshuawilson]

From @ldimaggi on July 10, 2017 2:30

Related to:
#355

[joshuawilson]

From @ldimaggi on July 10, 2017 2:32

Also note - if a user attempts to create a 2nd Che workspace, this error is logged in Che:

[ERROR] Pod with deployment name che-ws-q2jpva15vr6sho4t not found
[ERROR] Pod with deployment name che-ws-q2jpva15vr6sho4t not found

But - no error is displayed in the OSIO UI.

[joshuawilson]

From @ldimaggi on July 19, 2017 18:13

Related to:

fabric8-ui/fabric8-ui#1214

[qodfathr]

@joshuawilson is this different than #874? Can we close one as a dupe of the other. As mentioned in #287, we need an entire epic devoted to quota.

[ldimaggi]

Closing as duplicate of #874