ppd-dashboard

CMS PPD Physics Run Monitoring

Main Endpoint: http://ceyhun-k8s-lbva4duqns2g-node-0:32000 // accessible only in CERN network

How frontend works

• You have to provide root file name and object directory of that root file in the input bars:
• Example:
  • Go to: http://ceyhun-k8s-lbva4duqns2g-node-0:32000
  • ROOT File EOS Path : /eos/cms/store/group/comm_dqm/DQMGUI_data/Run2023/SpecialHLTPhysics15/0003667xx/DQM_V0001_R000366713__SpecialHLTPhysics15__Run2023B-PromptReco-v1__DQMIO.root
  • TDirectory Path Inside ROOT File :DQMData/Run 366713/EcalPreshower/Run summary/ESRecoSummary
  • Hit Show button and see 5 histograms which are directly read from EOS and their JSONs are rendered using JSROOT.

How backend works

• Pretty basic FastAPI, PyROOT and tests. Reads root file and returns JSONs (bit complicated than that 😊 ).
• We use mounted /eos in k8s so we can access CERN wide accessible EOS directories with proper kerberos authentication.
• ROOT files are read directly from EOS and histograms return with their JSON representation to the frontend.
• All the heavy work is carried on backend side and only JSON results send as responses.

Repository structure

Each folder includes its README.md file to understand details such as how to run server, how to build and push docker image, how to create new GH tag, etc.

• backend: FastAPI, PyROOT, tests, its Dockerfile
• frontend: Vue.js, Vite(just for vue deployment and build), tailwind CSS, JSROOT, its Dockerfile
• kubernetes: ppd-dashboard.yaml Kubernetes manifest file for both Frontend and Backend deployment
• .github/workflows: CI, GitHub actions that automatically builds docker images of both frontend and backend, and pushes to docker registry:
  • https://hub.docker.com/repository/docker/mrceyhun/ppd-dashboard-front
  • https://hub.docker.com/repository/docker/mrceyhun/ppd-dashboard-back

Kubernetes

Backend container details

• We could not solve kinit -k -t keytab $principle, preauthentication fails but kinit $principle works. Still debugging. Probably it's lack of a library, but I could not find it in ubuntu.
• We provide config.yaml in configMap which includes backend server details.
• Cron is to update kerberos ticket automatically however keytab problem prevent to use it.
• Currently, it's running with manual kinit and background job python backend/main.py >>/proc/1/fd/1 2>&1 &.
• SwaggerUI can be accessible in http://ceyhun-k8s-lbva4duqns2g-node-0:32001/docs

Frontend container details

• Normally, it is not advised to run both front/back ends in the same pod. However, this is just for test.
• One workaround is to provide backend API base url to the frontend service in run time.
  • We provide backend base url to axios so we don't define it in each axios request.
  • It is provided in frontend/src/main.js
  • We cannot use vite env variable feature of https://vitejs.dev/guide/env-and-mode.html because we build our service and run nginx.
  • So, we use a custom solution after build to replace env variable name with sed in substitute_environment_variables.sh before running nginx.

Internal documentations

• kubernetes/README.md
• frontend/README.md
• .github/workflows/README.md