Skip to content
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

Kafka balance fixes to 24 1 #3537

Merged
merged 3 commits into from
Apr 8, 2024
Merged

Kafka balance fixes to 24 1 #3537

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6f49701
Fix verify
niksaveliev Apr 6, 2024
5ae8fd6
Fix
niksaveliev Apr 6, 2024
f0f0e94
fix
niksaveliev Apr 7, 2024
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
25 changes: 13 additions & 12 deletions ydb/core/kafka_proxy/actors/kafka_read_session_actor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,11 +36,10 @@ void TKafkaReadSessionActor::HandleWakeup(TEvKafka::TEvWakeup::TPtr, const TActo
return;
}

for (auto& topicToPartitions: NewPartitionsToLockOnTime) {
auto& partitions = topicToPartitions.second;
for (auto& [topicName, partitions]: NewPartitionsToLockOnTime) {
for (auto partitionsIt = partitions.begin(); partitionsIt != partitions.end(); ) {
if (partitionsIt->LockOn <= ctx.Now()) {
TopicPartitions[topicToPartitions.first].ToLock.emplace(partitionsIt->PartitionId);
TopicPartitions[topicName].ToLock.emplace(partitionsIt->PartitionId);
NeedRebalance = true;
partitionsIt = partitions.erase(partitionsIt);
} else {
Expand Down Expand Up @@ -408,6 +407,8 @@ void TKafkaReadSessionActor::HandlePipeDestroyed(TEvTabletPipe::TEvClientDestroy
}

void TKafkaReadSessionActor::ProcessBalancerDead(ui64 tabletId, const TActorContext& ctx) {
NewPartitionsToLockOnTime.clear();

for (auto& [topicName, topicInfo] : TopicsInfo) {
if (topicInfo.TabletID == tabletId) {
auto partitionsIt = TopicPartitions.find(topicName);
Expand Down Expand Up @@ -579,8 +580,7 @@ void TKafkaReadSessionActor::HandleReleasePartition(TEvPersQueue::TEvReleasePart
auto newPartitionsToLockCount = newPartitionsToLockIt == NewPartitionsToLockOnTime.end() ? 0 : newPartitionsToLockIt->second.size();

auto topicPartitionsIt = TopicPartitions.find(pathIt->second->GetInternalName());
Y_ABORT_UNLESS(topicPartitionsIt != TopicPartitions.end());
Y_ABORT_UNLESS(record.GetCount() <= topicPartitionsIt->second.ToLock.size() + topicPartitionsIt->second.ReadingNow.size() + newPartitionsToLockCount);
Y_ABORT_UNLESS(record.GetCount() <= (topicPartitionsIt.IsEnd() ? 0 : topicPartitionsIt->second.ToLock.size() + topicPartitionsIt->second.ReadingNow.size()) + newPartitionsToLockCount);

for (ui32 c = 0; c < record.GetCount(); ++c) {
// if some partition not locked yet, then release it without rebalance
Expand All @@ -599,18 +599,19 @@ void TKafkaReadSessionActor::HandleReleasePartition(TEvPersQueue::TEvReleasePart
}

NeedRebalance = true;
size_t partitionToReleaseIndex = 0;
size_t i = 0;
ui32 partitionToRelease = 0;
ui32 i = 0;

for (size_t partIndex = 0; partIndex < topicPartitionsIt->second.ReadingNow.size(); partIndex++) {
if (!topicPartitionsIt->second.ToRelease.contains(partIndex) && (group == 0 || partIndex + 1 == group)) {
for (auto curPartition : topicPartitionsIt->second.ReadingNow) {
if (!topicPartitionsIt->second.ToRelease.contains(curPartition) && (group == 0 || curPartition + 1 == group)) {
++i;
if (rand() % i == 0) { // will lead to 1/n probability for each of n partitions
partitionToReleaseIndex = partIndex;
if (rand() % i == 0) {
partitionToRelease = curPartition;
}
}
}
topicPartitionsIt->second.ToRelease.emplace(partitionToReleaseIndex);

topicPartitionsIt->second.ToRelease.emplace(partitionToRelease);
}
}

Expand Down
Loading