- Daily health check results are visible for all devices.
- Device health is automatically checked before bringing a device back online.
- Devices failing health checks are marked offline and notifications are sent to Linaro Validation Team members to address the problem.
- 99% health check jobs succeed.
- 90% scheduler availability for processing jobs.
- 99% scheduler availability for accepting jobs.
- Allow views to be restricted by groups. A user need to be logged in and belong to that group to see them.
- Allow views to be marked as public. Approved views can be viewed by anyone.
- Ensure raw logs in the scheduler follow the same permissions and visibility as those of the job and bundle stream it will be submitted to.
- Ensure access through commandline tools and APIs is consistent with permissions from the web user interface.
- LAVA can deploy a new kernel image and rootfs to run in a fast model.
- Tests can be executed and results collected in LAVA on fast model instances.
- LAVA should be able to deploy Linaro Android image in fast models and run tests in it. This criteria depends on Linaro Android image availability.
- Investigate monitoring solutions.
- Create a mailing list for critical errors to go to LAVA administrators.
- Application errors from LAVA should be easily visible to site administrators.
- Critical OS resources should be monitored.
- Notification for critical events should be sent to LAVA administrators: site isn't available to users, site down, disk space alert.
Platform/Validation/Roadmap/2012Q1 (last modified 2012-04-04 11:02:57)