Export - CSV (All fields)
Export - CSV (Current fields)
SERVER-16569
do not update the replication progress map when receiving a heartbeat from primary
db.currentOp reports operations with incorrect namespace values
Help message for the --filePermissions option got converted from octal to decimal.
enable transaction log compression for WiredTiger
db.currentOp(true) missing BackgroundJob stats for 2.8
Disable "index_bigkeys_nofail.js" for sharding passthrough on slow2 suite
writeConflictException inefficiencySERVER-16518
listCollections and listIndexes return results in cursor object format
remove rbid from ismaster() resultsSERVER-16515
MMAPv1 should have a singleton RecordAccessTracker, rather than creating one per database
improve stability of multiVersion/w_majority_change.js
remove slow2/replset_prefetch_stress.js
Add stress test for mixed version and mixed storage engine replica sets, and regression test for SERVER-16189
initial sync needs to lock properly before truncating oplog
replication should obtain DB X lock when applying an index build op
Uncaught WriteConflictException in TTLMonitor
Reduce contention on ReplicationCoordinatorImpl::_mutex
sh.status shows incorrect owner and timing of balancer lock acquisition
Index entry too long breaks foreground index builds, even with FailIndexKeyTooLong set to false
Map Reduce Output to Collection Occasionally Kills the Server
OpContext should reuse Lockers
do not touch sync source connection outside of bgsync thread
Failed index builds can assert on progressMeter
Replace internalShardingMaxSplitPointsPerOperation
Remove unused List1 class
Attempting to create index on local database on replica set member in state REMOVED causes invariant failure
findAndModify with sort exceeds internal sort memory limit, silently fails
Log exception information in terminate handler where possible
[Create|Update]User command should not allow the "mechanism" field
Lock manager perf tests
prohibit simultaneous heartbeat reconfig and candidacy role state
index_restart_secondary.js
checkpointing with many collections/indexes is slow and contention prone
Queued updates are showing up as queued reads in mongostat
test upgrade/downgrade for master/slave
test difference in meaning of w: majority in mixed mode replicasets
syncSourceFeedback can spin, with network traffic, during reconfig
replica set monitor can display negative ping times
Log get flooded with remap
mongos_no_replica_set_refresh.js
Don't use stale connections for heartbeats
rollback_auth.js improvement
Concurrent mapReduces with same replacement collection fail with WT
WT session mutex is highly contented
replsets_prefetch_stress.js failed under ASAN slow2_WT test
db.getReplicationInfo shell helper relies on system.namespaces
Reduce temporary NamespaceString copies
minvalid2.js
sharding_rs2.js should clean up data on all replicas
More tests need to be agnostic to primary election