Coordinate Attrition targets within a test phase

[tclinkenbeard-oai]

Summary

Coordinate machine Attrition targets within a test phase so concurrent Attrition workloads share the strongest machinesToLeave floor and cannot claim the same zone. Recheck immediately before killing, while allowing a replacement workload to revisit its own zone, and add a focused lifecycle/floor/replacement regression.

Failure evidence

Managed correctness job failed tests/slow/SwizzledDdBalance.toml once in 190,859 runs with RandomSeed 1023295228, JoshuaSeed 978392406014025441, buggify and fault injection enabled. Immutable artifact:

The exact artifact replay reproduced the timeout and 1,000,001-line trace overflow. Three explicit Attrition workloads independently targeted all five zones despite two declaring machinesToLeave=3, leaving recovery unavailable while ChangeConfig attempted a coordinator move. This is a harness availability failure, distinct from the earlier old-TLog QuietDatabase assertion family.

Validation

Managed clang -Werror compile/link and the focused /fdbserver/workloads/MachineAttrition/phaseCoordination regression passed on both integrated revision e86c1a7e64346cd7aa4d9e445189cac79b7ccad3 and upstream base 1ae8e7f2383bbd79aa860940507db6ec0754055d.

Exact-seed patched replays passed on both binaries. Trace audits show only two unique killed zones (preserving the three-zone floor), successful ChangeConfig completion, and zero TestFailure, TestUnexpectedlyNotFinished, or TracedTooManyLines events. Additional managed smoke runs passed:

• slow/SwizzledDdBalance.toml --seed 1023295229 --buggify on -fi on
• fast/StatusDuringOutage.toml --seed 2456305 --buggify on -fi on
• fast/CycleTest.toml --seed 4101 --buggify on -fi on

The managed build wrapper reported a post-link nonzero status because the worker image lacks sccache; compilation/link, binary execution, source-hash identity, and focused tests were verified explicitly.

[foundationdb-ci]

Result of foundationdb-pr-clang-ide on Linux RHEL 9

• Commit ID: 348b22e
• Duration 0:33:02
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr-clang-arm on Linux RHEL 9

• Commit ID: 348b22e
• Duration 0:45:50
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr-macos-m1 on macOS 14.x

• Commit ID: 348b22e
• Duration 0:58:01
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr-clang on Linux RHEL 9

• Commit ID: 348b22e
• Duration 1:02:41
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr on Linux RHEL 9

• Commit ID: 348b22e
• Duration 1:06:43
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr-macos on macOS 14.x

• Commit ID: 348b22e
• Duration 1:28:18
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of foundationdb-pr-cluster-tests on Linux RHEL 9

• Commit ID: 348b22e
• Duration 1:32:28
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)
• Cluster Test Logs zip file of the test logs (available for 30 days)