Stop container if DMS is killed
Estimation
Story points : 5 SP
Estimated Duration: 6 Days
Description
Click to expand
Who
- @sntshk -- developer;
- @avimanyu786 @dagiopia -- reviewer;
- @kabir.kbr @janaina.senna -- advice and help related to the architecture, reviewers.
What
- If DMS is killed, container shouldn't keep running.
How
- The default behaviour for docker is to detach from the host program. This is also true on CLI. When you start a container from a bash window, exit that bash window, that container would still run.
- Do some R&D on the topic to realize how to stop the container if DMS is killed.
Why
- We need to implement the sequence diagram for running a ML on a GPU and compensate the computer provider described here.
When
- No technical dependencies.
Acceptance Criteria
Click to expand
- Code review
- Merge request
- Functional tests showing the container is stopped when the DMS is killed.
Work Breakdown Structure (WBS)
Task | Description | Duration | Status | Start Date | End Date | Comment |
---|---|---|---|---|---|---|
A | Modify DMS Go app to append container IDs to YAML file | 2 Hrs | Done | |||
B | Update DMS Go app to check if sidecar container is running | 4 Hrs | Done | |||
C | Start sidecar container with YAML file mounted if not running | 3 Hrs | Done | |||
D | Create Python script to monitor DMS process and read container IDs | 2 Hrs | Done | |||
E | Update Python script to stop containers if DMS is not running | 1 Hrs | Done | |||
F | Build Docker image for sidecar container | 2 Hr | Done | |||
G | Modify DMS deployment process to include sidecar container | 3 Hrs | Done | |||
H | Testing and editing if needed | 3 Hrs | Done | |||
I | container names system | 6 Hrs | Done | 04.07.2023 | 04.07.2023 | |
J | try a watchdog proccess | 8 Hrs | Done | 05.07.2023 | 06.07.2023 |
Edited by Ahmed Melouk