When there are long-running transactions the server can get stuck with the cache full of dirty data because the thread that is responsible for aborting timed-out transactions may become stuck trying to evict data that is pinned in cache by the long-running transactions that it's tasked with aborting.
- depends on
-
WT-6602 Allow operation timeout ms to be passed to commit and rollback
-
- Closed
-
- duplicates
-
SERVER-51512 cache evict problem
-
- Closed
-
- is duplicated by
-
SERVER-50348 Take advantage of WT transaction rollback timeout in multi-doc txn expirer
-
- Closed
-
- is related to
-
SERVER-51281 mongod live locked
-
- Closed
-
-
WT-6507 Exit cache eviction worker after our operation has timed out
-
- Closed
-
- mentioned in
-
Page Loading...