I am looking for a WebUI front-end to be able to schedule and start Gatling test scenario’s. The reason why I want to use a WebUI as opposed to the typical command line way of working, is that I want to offer Gatling load testing as a service in our Kubernetes development cluster (and to determine baseline/peak CPU/Memory statistics for our Cloud Native applications). The idea is that developers would be able to upload their Gatling test scenario’s into a front-end, start them there and get feedback on the results.
Anyone an idea if such a WebUI front-end is already available?
Kubernetes support is in our roadmap for FrontLine, expect it in a few months.
Thanks a lot Stephane for your reply.
I believe from a DEV perspective, developers have difficulty to baseline their apps wrt men/CPU request/limits on kubernetes or other container orchestration platforms. On the other hand, if OPS can provide this as a UI accessible webservice in a DEV cluster, developers can reuse there Gatling scenario's in an environment more close to ACC/PRO, establishing baselines together and make micro services a true cost saving exercise.
Really looking FW to this feature
Any update on frontline within docker containers on Kubernetes?
Planned for Q1 2018. Several of our customers are going with Openshift.
Any specifics about your env that might be interesting to share?
Do you want me to ping you once it’s available?
If you want to make your container OpenShift friendly from day one, make sure it can start as a random user. In OpenShift, the default behaviour to start docker based images as a container, is to completely ignore and “USER” entries in your Dockerfile and use a random user to start your docker process. The random user has a ID like eg 00000010000, and/but is always part of the root group. The trick to make make your container OpenShift friendly is to provide root group access towards the main process binary and its config files. The reason for all of this is security… OpenShift wants to prevent privilege escalation (the user id in the docker container does not exist on the nodes).
More information and reading on this: https://docs.openshift.com/container-platform/3.7/creating_images/guidelines.html#openshift-specific-guidelines
In practice, if you can run your docker image locally with "docker run -u 10000 -it gatling/frontline:latest" ", you are complying with OpenShift’s standard security constraints.
Note that most of docker images on docker-hub do not comply with this behaviour. One can overrule the security settings by adding anyuid SCC (eg: "oc adm policy add-scc-to-user anyuid -z gatling-frontline-user -n gatling-frontline"), but this should be avoided as much as possible.
Thanks a lot for the advice. I’ll let you know how things turn out!
Is there any place (website/github?) where I could follow up on this topic?
Well, our work on FrontLine is not public.