Disable core dumps for expected death tests in CI (#5163) * Disable core dumps for expected death tests in CI Run death-test suites separately with core dumps disabled when CI enables dump collection. Keep core dumps for other tests and preserve failures from both phases. This prevents successful death tests from filling the runner disk before the rest of the suite starts. * Make the death-test core dump opt-out explicit and guard it Replace the `unset COREDUMP_DIR` subshell with a NANO_DISABLE_CORE_DUMPS flag handled by run-tests.sh. The previous form relied on the reader knowing that run-tests.sh re-raises the limit whenever COREDUMP_DIR is set, so unsetting it was load bearing rather than incidental. Setting the limit with `ulimit -c 0` covers the hard limit as well. Fail the run if the death-test filter stops matching anything. The split rests entirely on the `DeathTest` suite name suffix; if that convention drifts, the death tests silently rejoin the dump-collecting pass and fill the runner's disk again with nothing reporting it. Move the test executable path into a common.sh helper, now that both run-tests.sh and run-core-tests.sh need it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
__________
Type: commit
Hash: f1244ee46278b378faee3e4e64f88dcfeab5e2b5
Created: yesterday
__________
gr0v1ty/nano-node:f1244ee46278b378faee3e4e64f88dcfeab5e2b5
| Test Case | Status | Duration | Median Deviation | Log | Graph |
|---|