The phrase "atomic test and set of disk block returned false for equality" typically points to a low-level synchronization failure within a filesystem or a storage area network (SAN). This error indicates that a system attempted to update a specific block of data but found that the block’s current state did not match the expected "baseline" state.
Re-mount Datastore: For persistent mount failures, some admins found success by removing and re-adding the datastore via the esxcli command line. The phrase "atomic test and set of disk
Environment: 10-node Ceph cluster, BlueStore backend, NVMe-over-Fabrics.
Error: OSD logs repeated: bluestore/StupidAllocator.cc: atomic test and set of disk block 0x4a20b returned false for equality.
Root cause: A network partition caused two OSDs to believe they held the same allocation bitmap lock. The storage array (NVMe target) correctly rejected the second OSD’s compare-and-write.
Fix: Reduced osd_heartbeat_grace from 20s to 5s, enabled faster fencing, and implemented retry logic with jitter. If false is returned consistently, the system enters
The Check: Before writing "State B," the system verifies that the block is still actually in "State A." Caution: Only clear reservations if you are certain
Iffalse is returned consistently, the system enters a spin-loop.
false wastes CPU cycles that could be used by the lock holder to release the lock.false generate high traffic on the memory bus (invalidating and re-reading the "lock" variable cache line), known as "cache thrashing."Caution: Only clear reservations if you are certain no active node holds them.