Remove invalid dspace healthcheck step from import jobs

[Kasinhou]

Problem

The scheduled Import Weekly run (Deploy DSpace → import-5 / import-8) fails even though the DB import itself succeeds. The post-import dspace healthcheck step aborts the job.

Tracking issue: dataquest-dev/dspace-customers#908
Failed run: https://github.com/dataquest-dev/dspace-angular/actions/runs/31497534447 (job import-8):

dspace healthcheck:
Command not found: healthcheck
Usage: dspace [command-name] {parameters}
##[error]Process completed with exit code 1.

Root cause

./dspace healthcheck is not a valid CLI command in the DSpace 7 backend (container image dataquest/dspace:dspace-7_x). The healthcheck launcher command (org.dspace.health.HealthReport) was a DSpace 5–era CLARIN feature and was never ported to the DSpace 7 fork — it is absent from dataquest-dev/DSpace dspace/config/launcher.xml on dtq-dev. The step is a leftover from DSpace 5. Since it runs under bash -e -o pipefail with no continue-on-error, the "command not found" (exit 1) fails the whole job.

Change set

Remove the dspace healthcheck step from the two jobs that had it, in .github/workflows/deploy.yml:

• import-5
• import-8

No replacement is added — there is no DSpace 7 equivalent, and the preceding dspace basic command step in each job already runs ./dspace version, ./dspace cleanup, ./dspace index-discovery, ./dspace oai import, and ./dspace checker -v -l, which cover the intended post-import verification. Minimal and surgical: 12 lines removed, nothing else touched.

Test evidence

# no healthcheck references remain
$ grep -ni healthcheck .github/workflows/deploy.yml
(no matches)

# workflow YAML still parses; all 8 jobs intact
$ python -c "import yaml; print(list(yaml.safe_load(open('.github/workflows/deploy.yml'))['jobs']))"
['deploy-5', 'deploy-8', 'import-5', 'import-8', 'playwright-after-deploy8', 'rest-tests-after-deploy8', 'playwright-after-import8', 'rest-tests-after-import8']

End-to-end verification is the next Import Weekly run — this workflow only executes on the self-hosted dspace-dep-1 runner, so it can't be exercised from a fork/local box.

Risk & rollback

Very low. The removed step never produced a successful result (the command does not exist), so no working behavior is lost. If a real health check is wanted later, it should be re-introduced as a valid DSpace 7 command in a separate change. Rollback: revert this commit.

Notes / assumptions

• Nothing consumes the step's output — it only echoed and invoked the missing command; no later step reads its result.
• Cross-repo: the fix lives in dspace-angular while the tracking issue is in dspace-customers, so a same-repo Fixes # keyword can't auto-close it; it is linked as a cross-reference instead.

🤖 Generated with Claude Code

[Copilot]

Pull request overview

Removes a failing post-import GitHub Actions step from the scheduled import jobs in deploy.yml, aligning the workflow with the actual DSpace 7 CLI (where ./dspace healthcheck is not available) so successful imports no longer get marked as failed.

Changes:

• Removed the dspace healthcheck step from import-5.
• Removed the dspace healthcheck step from import-8.
• Verified no remaining healthcheck references in the workflow file.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[milanmajchrak]

We want to run health report during every import, but at first we need to implement report diff enhancement to compare two health reports from file.