10259 Commits

Author SHA1 Message Date
Cai Yudong
2bd2cca04a
enhance: Truly support multi vector data types in SearchBruteForce (#40499)
Issue: #38666

Signed-off-by: CaiYudong <yudong.cai@zilliz.com>
2025-03-10 18:36:03 +08:00
yihao.dai
2ca2e2dbc8
fix: Fix parsing import endTs (#40332)
Parsing import beginTs, endTs as a hybrid timestamp.

issue: https://github.com/milvus-io/milvus/issues/40326

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2025-03-10 17:38:04 +08:00
XuanYang-cn
e6c46a25ea
enhance: Use correct counter metrics for overall wa calculation (#40394)
- Use CounterVec to calculate sum of increase during a time period.
- Use entries number instead of binlog size

---------

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-03-10 16:34:06 +08:00
congqixia
391804c7fb
enhance: Add channel seal policy based on blocking l0 (#40505)
Related to #40502

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-10 16:28:04 +08:00
XuanYang-cn
6f70e6d1e1
enhance: Log start position of delete msgs (#40315)
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-03-10 14:58:05 +08:00
sre-ci-robot
a6d4121034
[automated] Update Knowhere Commit (#40486)
Update Knowhere Commit
Signed-off-by: sre-ci-robot sre-ci-robot@users.noreply.github.com

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2025-03-10 12:28:04 +08:00
XuanYang-cn
4bebca6416
enhance: Replace currRows with NumOfRows (#40074)
See also: #40068

---------

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-03-10 12:16:03 +08:00
cai.zhang
d6a650bd14
fix: Skip executing stats for zero segment (#40448)
issue: #40241

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-03-09 21:14:02 +08:00
smellthemoon
faae8ee518
fix: store wrong offset when build tantivy in nullable field (#40452)
#40454

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2025-03-09 09:34:04 +08:00
Bingyi Sun
37b118d55d
fix: Skip loading primary key if index has raw data (#39921)
issue: https://github.com/milvus-io/milvus/issues/39907

Signed-off-by: sunby <sunbingyi1992@gmail.com>
2025-03-06 17:46:02 +08:00
congqixia
7fbeb5624e
enhance: Avoid convert body byte slice to string in httpserver (#40405)
The convertion of byte slice to string may copy the underline data which
may cause extra memory and cpu time for httpserver

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-06 16:28:02 +08:00
congqixia
5c5273f95e
fix: Pass Knapsnak ptr to avoid compact multiple times (#40400)
Related to #40388

The small segments may be put into bucket twice due to value parameter
of Knapsnap.packWith

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-06 15:42:03 +08:00
sthuang
e0ec1aceeb
fix: skip storage v2 unstable ut for now (#40378)
related: #39173

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-03-05 20:30:39 +08:00
congqixia
fde80bc8b7
enhance: Remove debug log in rg handler v2 (#40376)
Remove debug log in resource group handler

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-05 19:18:00 +08:00
Spade A
3db56560fb
fix: fix concurrent issues in null offset (#40363)
issue: #40308
This issue fixes these two concurrent issues:
1. element in null_offset is used to set bitset where the size of bitset
is initialized by tantivy document count. However, there may still be
some documents that are not committed in tantivy but are null in
null_offset. So array out of range occurs.
2. null_offset can be read and write concurrently but there's no
synchronization protection.

---------

Signed-off-by: SpadeA <tangchenjie1210@gmail.com>
2025-03-05 17:48:00 +08:00
Ted Xu
878ce56079
fix: correct memory size estimation on arrays (#40312)
See: #40342

---------

Signed-off-by: Ted Xu <ted.xu@zilliz.com>
2025-03-05 16:54:09 +08:00
Zhen Ye
1637cf5664
enhance: better logging for grpc resolver (#40337)
issue: #40311

- better logging for grpc resolver
- remove the redundant streaming node manage client when streaming
service is disable

Signed-off-by: chyezh <chyezh@outlook.com>
2025-03-05 15:12:01 +08:00
Ted Xu
96952ad3c5
fix: compaction task cannot be genereted if size greater than max size (#40348)
See: #40343

---------

Signed-off-by: Ted Xu <ted.xu@zilliz.com>
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Co-authored-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-05 14:40:01 +08:00
yihao.dai
004a1875dc
enhance: Introduce batch subscription in msgdispatcher (#39863)
Introduce a batch subscription mechanism in msgdispatcher: the
msgdispatcher now includes a vchannel watch task queue, where all
vchannels in the queue will subscribe to the MQ only once and pull
messages from the oldest vchannel checkpoint to the latest.

issue: https://github.com/milvus-io/milvus/issues/39862

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2025-03-05 14:38:02 +08:00
Bingyi Sun
be4d09561b
fix: Fix missing null or non-exist key in json index (#40336)
issue: #35528

---------

Signed-off-by: sunby <sunbingyi1992@gmail.com>
2025-03-05 11:48:02 +08:00
sthuang
63a7c4570e
feat: storage v2 sync (#39663)
related: #39173

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-03-05 11:22:15 +08:00
SimFG
a3755cf409
fix: improve error handling and unit tests for InitMetaCache function (#40322)
- issue: #40320

Signed-off-by: SimFG <bang.fu@zilliz.com>
2025-03-05 11:08:13 +08:00
junjiejiangjjj
b2e630b1a1
feat: Support TEI serving and support int8 embedding (#40199)
#35856

Signed-off-by: junjie.jiang <junjie.jiang@zilliz.com>
2025-03-05 10:18:00 +08:00
Zhen Ye
9ca5088f62
fix: duplicate consuming from stream for invisble segment (#40316)
issue: #40207

Signed-off-by: chyezh <chyezh@outlook.com>
2025-03-04 15:54:00 +08:00
sthuang
d77756cf2d
fix: fix storage v2 cgo mem leak (#40305)
related: #39173

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-03-04 14:16:00 +08:00
XuanYang-cn
837ac295fa
enhance: Remove iterators in datanode (#40301)
Iterators are long deprecated, but sort are still using it. This PR
unifies stats task with the latest compaction common functions and
remove the usage of iterators.

1. Rename `datanode/compaction` to `datanode/compactor`
2. Add `internal/compaction` and move some compaction commons into it.
3. Replace `DeltalogIterators` with `ComposeDeleteFromDeltalogs`
4. Remove `datanode/iterators`

See also: #39242

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-03-04 12:14:00 +08:00
Bingyi Sun
7040ba1c12
enhance: make json path index support term filter (#40140)
issue: #35528

---------

Signed-off-by: sunby <sunbingyi1992@gmail.com>
2025-03-04 11:56:02 +08:00
congqixia
f2ea4d6370
fix: Add duplication param check for create index (#40254)
Related to #40156

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-03-04 11:42:02 +08:00
Chun Han
16aa123185
fix: restful drop db properties failure(#39953) (#40257)
related: #39953

Signed-off-by: MrPresent-Han <chun.han@gmail.com>
Co-authored-by: MrPresent-Han <chun.han@gmail.com>
2025-03-03 15:39:59 +08:00
Zhen Ye
f47ab31f23
enhance: remove redundant resource key watch operation, just keep consistency of wal (#40235)
issue: #38399
related PR: #39522

- Just implement exclusive broadcaster between broadcast message with
same resource key to keep same order in different wal.
- After simplify the broadcast model, original watch-based broadcast is
too complicated and redundant, remove it.

---------

Signed-off-by: chyezh <chyezh@outlook.com>
2025-03-03 14:40:05 +08:00
sthuang
de02a3ebcc
feat: Storage v2 binlog packed record reader and writer (#40221)
related: #39173

---------

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-03-03 10:24:02 +08:00
Zhen Ye
8eb662b4dc
enhance: add more metrics for async cgo component (#40136)
issue: #40014

Signed-off-by: chyezh <chyezh@outlook.com>
2025-03-03 09:56:03 +08:00
sre-ci-robot
6a57a1973f
[automated] Update Knowhere Commit (#40283)
Update Knowhere Commit
Signed-off-by: sre-ci-robot sre-ci-robot@users.noreply.github.com

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2025-03-03 01:11:58 +08:00
cai.zhang
5a810400b5
enhance: Optimize Task Scheduling to Enable Concurrent Execution (#40251)
issue: #39101 

2.5 pr: #40104

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-03-02 18:38:00 +08:00
zhagnlu
7a17fb68ec
enhance: add monitor metric for retrieve raw data (#40141)
#40078

Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2025-03-02 18:30:01 +08:00
zhagnlu
8c19e5c4a7
enhance: decrease delete record dump snapshot limit (#40101)
#40100

Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2025-03-02 17:55:59 +08:00
yihao.dai
c368113233
fix: Fix task delta cache data race (#40259)
issue: https://github.com/milvus-io/milvus/issues/40258

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2025-03-02 16:52:09 +08:00
Zhen Ye
2ff657f2d9
fix: wal may panics when context canceled (#40265)
issue: #40264

- wal may panics when context canceled
- scanner may data race when closing

Signed-off-by: chyezh <chyezh@outlook.com>
2025-02-28 17:41:58 +08:00
Zhen Ye
bc8e02df44
enhance: always enable streaming service (#40253)
issue: #38399

Signed-off-by: chyezh <chyezh@outlook.com>
2025-02-28 15:38:01 +08:00
cai.zhang
dc46b08bdf
fix: Fix bug where prefix matching fails when wildcards are in prefix (#40020)
issue: #40019

---------

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-28 10:29:59 +08:00
cai.zhang
762a644d76
enhance: Limit the speed of the generating stats task (#39644)
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-28 10:27:59 +08:00
wei liu
b0806bb900
fix: task delta cache leak due to duplicate task id (#40183)
issue: #40052

task delta cache rely on the taskID is unique, so it incDeltaCache at
AddTask, and decDeltaCache at RemoveTask, but the taskID allocator is
not atomic, which cause two task with same taskID, in such case, it will
call incDeltaCache twice, but call decDeltaCacheOnce, which cause delta
cache leak.

---------

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2025-02-28 10:22:08 +08:00
cai.zhang
13aff35a83
enhance: Add metrics for parse expression (#39654)
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-28 10:07:58 +08:00
Micka
5cc104b412
fix: Change CMake variable for switch to knowhere-cuvs (#40105)
issue: https://github.com/milvus-io/milvus/issues/39883

Signed-off-by: Mickael Ide <mide@nvidia.com>
2025-02-27 22:05:58 +08:00
Chun Han
259f9106ad
enhance: refine variable-length-type memory usage(#38736) (#39578)
related: #38736

Signed-off-by: MrPresent-Han <chun.han@gmail.com>
Co-authored-by: MrPresent-Han <chun.han@gmail.com>
2025-02-27 21:13:58 +08:00
wei liu
94f55df7fb
enhance: clean shard location cache after collection released (#40088)
issue: #40077

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2025-02-27 19:42:05 +08:00
cai.zhang
fc31308c5c
fix: Skip text index creation when segment is zero after sorting (#39962)
issue: #39961

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-27 17:55:58 +08:00
cai.zhang
a74580c1ca
fix: Set task version for stats task (#40035)
issue: #40034

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-27 17:49:59 +08:00
cai.zhang
eb04686348
enhance: Add metrics for proxy queue (#40070)
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-02-27 17:30:01 +08:00
yihao.dai
a148a78629
fix: Fix collections with duplicate names can be created (#40143)
This PR introduces two restrictions:
1. Before dropping a collection, all aliases associated with that
collection must be dropped.
2. When creating a collection, if the collection name duplicates any
alias, the collection creation will fail.

issue: https://github.com/milvus-io/milvus/issues/40142

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2025-02-27 17:27:57 +08:00