Thanks for info. Agree with you, these are essential features to both clients and platform teams. Looking forward for them …
On Dec 6, 2016, at 3:13 PM, Srinivas Naga Kotaru (skotaru) <skotaru cisco com> wrote:
we are continuously hearing 2 complaints from our users
not much verbose info to troubleshott/narrow down 2 commonf failures
1. Pod unable to come up. Why it failed, what caused?
2. Deployment failure. Why it failed ? what is the reason?
Most clints using console, so would be nice if we add as much info as possible to both console and oc clients.
Current console info/events not sufficient for them. One example I can quickly quote was, we restricted quota and foret to put scopes. Since scopes missing, quota is applied to deploy pods also, since quota is insufficient, deployment pods unable to schedule. We adding scopes to nonterminated pods so that quota not applicable short lived deploy pods. Verbose info from console/oc unable to narrow issue due to quota. They simply failing without throwiing any meaningful info
1.5 will contain deployment and replication controller conditions available from the API that indicate the reason why pods couldn't be created, or if rollout is blocked for some reason. This *should* be something that when made visible in the UI and CLI makes it clear to users what the problem is. Copying Jessica for that. It may need some info copied into the deployment logs (which are not very verbose today)
We'd also like to do more to capture failure scenarios from containers into the container status - that's not currently lined up but is desired.
dev mailing list
dev lists openshift redhat com