The thread in kill_sessions_common.cpp acquires the client lock and reads the logical session id on opCtx, while the thread in sharding_ddl_util.cpp sets a new logical session id without the client lock.

Consequently, we could end up with a data race at the logical session id from the opCtx.

A possible solution could be to protect this write with the client lock.