[lweberru]

Dev release 39558649713 built and published its exact-source artifacts successfully, then failed staging lzc-plan-template: Cloud Foundry reported gzip: stdout: No space left on device, with staging_disk_in_mb=4096. The new application was not deployed. Earlier package, image and native runtime checks passed; the failure is the additional droplet-compression staging footprint.

Set the zero-instance Runner template's disk quota to 6144 MiB, matching the existing Web/API staging allowance. Keep validation tasks, cold jobs and warm Runner processes explicitly at 1024 MiB memory / 4096 MiB runtime disk. The template has zero permanent instances. Preserve image verification, package identity, isolated destinations and all release gates. Update the English/German Runner and CF runbooks.

Validation:

• Release, image-staging, buildpack and Runner guards: 44 passed under dev-skip-assistant-qualification; manifest contract verifies 6144 MiB and zero template instances.
• Warm process / pool tests: 76 passed, including explicit 4096-MiB runtime disk.
• CF platform/Application job request tests: 6 passed, now explicitly asserting 1024 MiB memory / 4096 MiB task disk for both brokers.
• Relevant formatting/lint and git diff --check pass.

Hosted acceptance of the larger staging allowance remains to be verified after merge. Start a fresh Dev release with runner_pool_size=1, release_mode=dev-skip-assistant-qualification, build_only=false and empty release_run_id. This Dev mode currently disallows artifact promotion; do not re-run the failed attempt or bypass its gates. No production or customer infrastructure Apply.