inttests: kill test runner containers faster

[chmeliik]

Previously, each test scenario would wait 10 seconds to remove the test container. This is because the test containers use 'sleep' as the entrypoint. Sleep doesn't have SIGTERM handling, so when it runs as PID 1, SIGTERM doesn't do anything and the container engine resorts to SIGKILL after 10 seconds.

Kill test runner containers immediately (without the 10 second wait) if they use --entrypoint sleep. This greatly speeds up integration tests.

[chmeliik]

When running

go test ./integration_tests/build_test.go -v

On my machine it takes 105 seconds in main vs 15 seconds with this change

[qodo-code-review]

Code Review by Qodo

🐞 Bugs (0) 📘 Rule violations (1) 📎 Requirement gaps (0)

1. --time breaks docker rm 📘 Rule violation ⛯ Reliability

Description

Delete() appends the --time flag whenever killTimeout is set, but containerTool may be
docker where rm does not support --time. This can cause container deletion failures and
flaky/slow integration tests due to leftover containers.

Code

integration_tests/framework/runner_container.go[R271-276]

func (c *TestRunnerContainer) Delete() error {
-	stdout, stderr, _, err := c.executor.Execute(containerTool, "rm", "-f", c.name)
+	args := []string{"rm", "-f", c.name}
+	if c.killTimeout != nil {
+		args = append(args, "--time", strconv.Itoa(*c.killTimeout))
+	}
+	stdout, stderr, _, err := c.executor.Execute(containerTool, args...)

Evidence

Compliance requires handling edge cases and external dependency differences; the code now
conditionally adds a tool-specific flag without checking which container engine is in use, and the
framework explicitly supports both podman and docker.

Rule 3: Generic: Robust Error Handling and Edge Case Management
integration_tests/framework/runner_container.go[271-276]
integration_tests/framework/runner_container.go[235-240]
integration_tests/framework/common.go[52-59]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
`Delete()` appends `--time` to `rm` when `killTimeout` is set. The framework supports both `podman` and `docker`, but `docker rm` does not support `--time`, so deletions may fail.

## Issue Context
`Start()` sets `killTimeout` to `0` when `ReplaceEntrypoint` is enabled, meaning the new `--time` behavior is exercised frequently.

## Fix Focus Areas
- integration_tests/framework/runner_container.go[235-240]
- integration_tests/framework/runner_container.go[271-276]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

ⓘ The new review experience is currently in Beta. Learn more

[snyk-io]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scanner	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues
✅	Licenses	0	0	0	0	0 issues
✅	Code Security	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.