Skip to content

[release-7.3] Add team shard count to DD team info log and improve logging in KVS. #12125

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: release-7.3
Choose a base branch
from
Open

[release-7.3] Add team shard count to DD team info log and improve logging in KVS. #12125

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions fdbclient/ServerKnobs.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -629,6 +629,7 @@ void ServerKnobs::initialize(Randomize randomize, ClientKnobs* clientKnobs, IsSi
init( SHARDED_ROCKSDB_HISTOGRAMS_SAMPLE_RATE, 0.001 ); if( isSimulated ) SHARDED_ROCKSDB_HISTOGRAMS_SAMPLE_RATE = deterministicRandom()->random01();
init( SHARDED_ROCKSDB_USE_DIRECT_IO, false ); if (isSimulated) SHARDED_ROCKSDB_USE_DIRECT_IO = deterministicRandom()->coinflip();
init( SHARDED_ROCKSDB_FLUSH_PERIOD, 0.0 ); if (isSimulated) SHARDED_ROCKSDB_FLUSH_PERIOD = deterministicRandom()->randomInt(100, 1000);
init( SHARDED_ROCKSDB_DETAILED_STATS, false ); if (isSimulated) SHARDED_ROCKSDB_DETAILED_STATS = deterministicRandom()->coinflip();


// Leader election
Expand Down
1 change: 1 addition & 0 deletions fdbclient/include/fdbclient/ServerKnobs.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -597,6 +597,7 @@ class ServerKnobs : public KnobsImpl<ServerKnobs> {
double SHARDED_ROCKSDB_HISTOGRAMS_SAMPLE_RATE;
bool SHARDED_ROCKSDB_USE_DIRECT_IO;
double SHARDED_ROCKSDB_FLUSH_PERIOD;
bool SHARDED_ROCKSDB_DETAILED_STATS;

// Leader election
int MAX_NOTIFICATIONS;
Expand Down
3 changes: 3 additions & 0 deletions fdbserver/DDTeamCollection.actor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3639,6 +3639,8 @@ class DDTeamCollectionImpl {
}
}

self->shardsAffectedByTeamFailure->traceTeamShardMapping();

// TODO: re-enable the following logging or remove them.
// TraceEvent("LocalityRecordKeyName", self->getDistributorId())
// .detail("Size", internedLocalityRecordKeyNameStrings.size())
Expand Down Expand Up @@ -4942,6 +4944,7 @@ void DDTeamCollection::traceAllInfo(bool shouldPrint) const {
traceMachineTeamInfo();
traceLocalityArrayIndexName();
traceMachineLocalityMap();
shardsAffectedByTeamFailure->traceTeamShardMapping();
}

void DDTeamCollection::rebuildMachineLocalityMap() {
Expand Down
18 changes: 12 additions & 6 deletions fdbserver/KeyValueStoreShardedRocksDB.actor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1298,12 +1298,18 @@ class ShardManager {
int numLevels = 0;
for (auto it = cfMetadata.levels.begin(); it != cfMetadata.levels.end(); ++it) {
std::string propValue = "";
ASSERT(shard->db->GetProperty(shard->cf,
rocksdb::DB::Properties::kCompressionRatioAtLevelPrefix +
std::to_string(it->level),
&propValue));
e.detail("Level" + std::to_string(it->level),
std::to_string(it->size) + " " + propValue + " " + std::to_string(it->files.size()));
if (SERVER_KNOBS->SHARDED_ROCKSDB_DETAILED_STATS) {
ASSERT(shard->db->GetProperty(shard->cf,
rocksdb::DB::Properties::kCompressionRatioAtLevelPrefix +
std::to_string(it->level),
&propValue));
e.detail("Level" + std::to_string(it->level),
std::to_string(it->size) + " " + propValue + " " +
std::to_string(it->files.size()));
}
if (it->level == 0) {
e.detail("Level0Files", it->files.size());
}
if (it->size > 0) {
++numLevels;
}
Expand Down
28 changes: 28 additions & 0 deletions fdbserver/ShardsAffectedByTeamFailure.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -196,6 +196,34 @@ void ShardsAffectedByTeamFailure::setCheckMode(CheckMode mode) {
checkMode = mode;
}

void ShardsAffectedByTeamFailure::traceTeamShardMapping() const {
Team prevTeam;
int count = 0;
int teamCount;
for (auto it = team_shards.begin(); it != team_shards.end(); ++it) {
if (it->first != prevTeam) {
if (count > 0) {
TraceEvent("DDTeamShardCount")
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How many events will be emitted?

.detail("IsPrimary", prevTeam.primary)
.detail("Team", prevTeam.toString())
.detail("Shards", count);
}
count = 1;
prevTeam = it->first;
++teamCount;
} else {
++count;
}
}
if (count > 0) {
TraceEvent("DDTeamShardCount")
.detail("IsPrimary", prevTeam.primary)
.detail("Team", prevTeam.toString())
.detail("Shards", count);
}
TraceEvent("DDTeamShardStats").detail("TotalShards", team_shards.size()).detail("TotalTeams", teamCount);
}

void ShardsAffectedByTeamFailure::check() const {
if (checkMode == CheckMode::ForceNoCheck)
return;
Expand Down
2 changes: 2 additions & 0 deletions fdbserver/include/fdbserver/ShardsAffectedByTeamFailure.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,6 +106,8 @@ class ShardsAffectedByTeamFailure : public ReferenceCounted<ShardsAffectedByTeam
void assignRangeToTeams(KeyRangeRef keys, const std::vector<Team>& destinationTeam);
void check() const;
void setCheckMode(CheckMode);
// Prints the team shard mapping as trace events.
void traceTeamShardMapping() const;

PromiseStream<KeyRange> restartShardTracker;

Expand Down