Range sizes definition#438
Merged
yangsw26 merged 1 commit intoMar 13, 2026
Merged
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
🚧 Files skipped from review as they are similar to previous changes (3)
🚥 Pre-merge checks | ✅ 1 | ❌ 2❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
8 tasks
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@tx_service/include/cc/template_cc_map.h`:
- Around line 11940-11953: The code path that transitions
std::get<0>(it->second) from RangeSizeStatus::kNotInitialized to kLoading
returns before applying the incoming delta, losing the first delta_size; modify
the branch handling (around the check of RangeSizeStatus and the call to
shard_->FetchTableRangeSize(table_name_, static_cast<int32_t>(partition_id),
cc_ng_id_, ng_term)) to record/apply the current delta_size into the stored
tuple (.second) before returning (e.g., update the size/delta field in
it->second so the initial in-memory change is preserved while the async load is
in progress).
- Around line 11969-11981: The current logic computes need_split and then
unconditionally writes it back into std::get<2>(it->second), which lets a
pre-set true flag be overwritten to false on a later update; change the update
to latch the split flag until an explicit reset clears it by reading the
existing flag first and only setting std::get<2>(it->second) = existing_flag ||
need_split (or equivalently compute need_split = existing_flag ||
new_condition), so the stored split state for the tuple containing
std::get<2>(it->second) is preserved across updates until the reset path clears
it; reference symbols: need_split, std::get<2>(it->second),
StoreRange::range_max_size, table_name_.StringView(), partition_id.
In `@tx_service/src/cc/cc_map.cpp`:
- Around line 481-486: After applying the persisted_size + delta and clearing
the delta in the async-load success path, run the same split-evaluation logic
used elsewhere so an oversized range doesn't remain unsplit: compute final_size,
compare it to the split threshold, set the per-range "split triggered" flag if
not already set, and either call the existing UpdateRangeSize (or the same
enqueue/signal used by writes to request a split) or surface a callback so the
caller can enqueue the split immediately; update the code around the block that
modifies std::get<0>(it->second) and std::get<1>(it->second) to perform that
check and signal.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: cf47539a-aaaf-4f4b-8531-5724b969884b
📒 Files selected for processing (4)
tx_service/include/cc/cc_map.htx_service/include/cc/template_cc_map.htx_service/include/type.htx_service/src/cc/cc_map.cpp
Init range sizes interface the interface of update range size fix rang size status when update add reset range status interface fix range size interface
yangsw26
force-pushed
the
range_size_definition
branch
from
7098f16 to
9a30770
Compare
yangsw26
changed the base branch from
main
to
shard_keys_one_core_for_rangepartition_base
liunyl
approved these changes
Mar 13, 2026
yangsw26
merged commit 46b9d87
into
shard_keys_one_core_for_rangepartition_base
4 checks passed
yangsw26
added a commit
that referenced
this pull request
Mar 16, 2026
* Range sizes definition (#438) In order to trigger range splits in a timely manner, the range size is tracked in memory. Init range sizes interface Update range size interface Reset range splitting status interface * Update store range size after datasync (#439) 1. Update storage range size After a successful flush, the store range size is persisted. 2. Load range size from storage When accessing the memory range size, if it has not yet been initialized, a fetch operation from storage is performed to retrieve the range size. * Maintaining range size on postwritecc (#440) 1. During a post-write operation, the range size information for the corresponding partition is maintained. 2. If a range is in the process of splitting, the range size is updated in the delta size information. 3. For a "double-write" operation, only the range size information for the newly split partition is updated. * Reset range size after range split (#441) 1. In the post-commit phase of a range split transaction, the range size of all related partitions is updated: base range size + delta size. 2. Reset the range splitting flag. 3. Update the kickoutcc process to accommodate the new key sharding logic. 4. Update the processing procedure for SampleSubRangeKeys to accommodate the new key sharding logic. * Update range size during create secondary index (#442) 1. Update range size during UploadBatchCc request for new index. 2. Update the UploadBatchCc process to accommodate the new key sharding logic. 3. Update the create secondary index process to accommodate the new key sharding logic. * Update range split replay log execution (#443) 1. Update the range size during data log replay. 2. For post commit range split log, update range size for each newly splitting ranges. 3. Update the data log replay process to accommodate the new key sharding logic. * Update scanslice request (#444) Update the structure definitions and related processing procedures of ScanSliceCc and RemoteScanSlice to adapt to the new key sharding logic. * Adapt DataSync with new key sharding (#445) 1. Update the structure definitions and related processing procedures of DataSyncScanCc and ScanSliceDeltaSizeCc, as well as the DataSync processing procedure, to adapt to the new key sharding logic. 2. Update key shard code for UpdateCkptTs request. * Adapt load slice with new key sharding (#446) Update the structure definition and related processing procedures of FillStoreSlice to adapt to the new key sharding logic. * Update process read/batchread operation (#447) Adapt read operation with new key sharding for range partition. * Create a high-priority DataSync task to trigger range split (#448) Update datasynctask constructor to check new and old range owner shard * Adapt cache sender with new key sharding (#449) To reduce cache hit rate, during range splitting, keys located on the new range that fall on other cores (local nodes or remote nodes) can be sent to the corresponding core. 1. Update the logic and related requests for sending range cache during range split 2. Update key shard for UploadBatchSlices rpc. * Update keycach to adapt to the new key sharding strategy. (#450) Including InitKeyCacheCc ， UpdateKeyCacheCc ，StoreSlice::cache_validity_ , StoreRange::key_cache_ * Reset range splitting status after datasync (#451) If a datasync task that is supposed to trigger a range split ends without actually triggering the split, the range splitting state needs to be reset. * Update comment (#455)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
In order to trigger range splits in a timely manner, the range size is tracked in memory.
Init range sizes interface
Update range size interface
Reset range splitting status interface
Summary by CodeRabbit
New Features
Documentation