Commit Graph

18517 Commits (ad37b98cdab681a846085edab58e72af699c8349)

Author SHA1 Message Date
smellthemoon 288844e3cf
enhance: Reduce the goroutine in flowgraph to 2 (#28233) (#28545)
each node in flow graph alloc a goroutine, but it is actually executed
sequentially and can be placed in one goroutine. InputNode will consume
msg form msgstream, alloc one goroutine.
issue: #24826 
pr: #28233

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-11-24 15:00:26 +08:00
chyezh 1b86efbafb
fix: [cherry-pick] error root path for catalog when using tikv (#28463)
Related Issue: #28162 
Related pr: #28163

Signed-off-by: chyezh <ye.zhen@zilliz.com>
2023-11-24 14:28:24 +08:00
yah01 c6ba4fa102
enhance: enhance the accuracy of memory usage (#28554) (#28559)
before this, Milvus use container/system's memory info to get the memory
usage, which could be inaccurate.

we allocates the memory by private anon mmap,
then rss - shared would be the accurate memory usage

resolve https://github.com/milvus-io/milvus/issues/28553
pr: #28554

---------

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-24 14:12:25 +08:00
cai.zhang 968d6fdcde
fix: Fix bug for parse expression that include quotes (#28418)
master pr: #28416 
issue: #28365
Fix bug for parsing error when a string enclosed in single quotes in an
expression contains multiple double quotes.
such as:
```
expr = "tag == '\"blue\"'"
```

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2023-11-24 11:02:24 +08:00
yah01 5ca7851f4a
enhance: refine error messages (#28424) (#28614)
- Split the simple reason and full detail
- Refine existing error messages related: #28422
related: https://github.com/milvus-io/milvus/issues/28422
pr: #28424

---------

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-24 10:04:24 +08:00
zhagnlu 0bda17e97b
enhance: Add precheck when chunk manager init (#28593)
#28329 
pr: https://github.com/milvus-io/milvus/pull/28330

Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-11-23 20:34:25 +08:00
zhagnlu 41667f3227
enhance: Disable stdout buffer (#28597)
#28390 
pr: https://github.com/milvus-io/milvus/pull/28391

Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-11-23 20:20:24 +08:00
zhagnlu e13b3aa9b1
fix: panic if minio port is empty (#28595)
#28460 
pr: https://github.com/milvus-io/milvus/pull/28461

Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-11-23 20:18:33 +08:00
Enwei Jiao c73bb26782
fix: disable reset kafka connection timeout (#28681)
pr: https://github.com/milvus-io/milvus/pull/28642
issue https://github.com/milvus-io/milvus/issues/28588

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-11-23 19:42:30 +08:00
wei liu 33bbdf6c88
fix: Fix channel manager bg checker exit when disable auto balance (#28459) (#28606)
issue: #28454
pr: #28459

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-23 19:18:29 +08:00
wayblink a04b528ce7
fix: [cherry-pick] failed to release collection with more than 128 partitions (#28567)
issue: #28343
pr: #28446

Signed-off-by: wayblink <anyang.wang@zilliz.com>
2023-11-22 14:00:23 +08:00
smellthemoon d2ebbe2317
enhance: create goroutine only once(#28594) (#28609)
create goroutine only once when getOrCreateMergedTimeTickerSender
pr: #28594

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-11-22 10:26:28 +08:00
congqixia 0fbd610e89
fix: [cherry-pick] querynodev2 local worker failed to maintain collection ref (#28631)
Cherry-pick from master
pr: #28590 #28598
See also #28589 #28596 
Increase ref for collection during load and unref after load completed.
Use the same logic protection from services.go `LoadSegments`
Perform `Unref` after release sealed segments

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-22 10:04:23 +08:00
zhuwenxing 558e301236
test:skip a bulk insert testcase temporarily (#28621)
skip a bulk insert test case temporarily.
It is a known issue but needs more time to solve. skip the test case is
for not blocking other PR

Signed-off-by: zhuwenxing <wenxing.zhu@zilliz.com>
2023-11-21 19:12:22 +08:00
wei liu c7ec882033
enhance: Remove rpc during querycoord start (#28396) (#28604)
issue: #28332
pr: #28396

during querycoord's recover, it try to call `DescribeCollection` and
`ShowPartitions` to root coord, to checker whether collection or
partition has been released in rootcoord. but if rootcoord isn't not
ready yet, the rpc will fail, the querycoord panic.

to fix this, we remove rpc call during querycoord's start

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-21 18:08:29 +08:00
smellthemoon d724b07037
enhance: Use single instance for mergedTimeTickerSender (#27730) (#28546)
use single instance for mergedTimeTickerSender
issue: https://github.com/milvus-io/milvus/issues/24826
pr: https://github.com/milvus-io/milvus/pull/27730

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-11-21 16:04:23 +08:00
PowderLi 92e5f96064
fix: symbol 'GetStorageMetrics' and 'enableDynamicField' (#28581)
issue: #28579 #28504
pr: #28580

1. replace enableDynamic with enableDynamicField
2. cgo directly link to milvus_storage

Signed-off-by: PowderLi <min.li@zilliz.com>
2023-11-21 15:30:23 +08:00
PowderLi 1383680b3a
add internal storage metrics (#28279)
/kind improvement
issue: #28277

pr: #28278

Signed-off-by: PowderLi <min.li@zilliz.com>
2023-11-19 17:22:25 +08:00
PowderLi 4a46ae7a8a
Improve HTTP server include metrics and RESTful API (#28210)
/kind improvement
issue: #27653

pr: #28045

Signed-off-by: PowderLi <min.li@zilliz.com>
2023-11-19 17:16:21 +08:00
congqixia 484131a550
fix: [cherry-pick] protect EventDispatcher map with mutex (#28540) (#28555)
Add mutex protection for `EventDispatcher.registry` map 
Fix #28538
pr: #28540 #28552
Also add unit test case for all methods for EventDispatcher

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-19 12:56:20 +08:00
yah01 2073a0a76a
enhance: reduce 1x copy for variable length field while retrieving (#28345) (#28531)
- Reduce 1x copy for varchar/string/JSON/array types while retrieving
- Reduce 1x copy for int8/int16 while retrieving
fix #28530
pr: #28345

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-17 20:48:21 +08:00
yah01 1743f581e6
enhance: reduce 1x copy while retrieving data from growing segment (#28323) (#28475)
We first fill the data into a vector and then copy it into the proto,
for some types (exclude variable-length types and int8, int16),
data could be directly copied into the proto.

Sealed segment has been optimized in
https://github.com/milvus-io/milvus/pull/28106.

pr: #28323

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-17 14:40:26 +08:00
congqixia d0f94f3d17
fix: make qcv2 observer dispatcher execute exactly once (#28472) (#28477)
Cherry-pick from master
pr: #28472
See also #28466

In `taskDispatcher.schedule`, same task may be resubmitted if the
previous round did not finish
In this case, TaskObserver.check may set current target by mistake,
which may cause the random search/query failure

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-17 01:34:21 +08:00
Enwei Jiao 9997eeb4bb
fix: Fix rocksdb compile failed (#28486)
pr: #28484
/kind improvement

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-11-16 16:38:20 +08:00
yah01 21d995b1d3
enhance: improve the compile speed (#28444) (#28464)
improve the CI compile speed with mold
pr: #28444

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-15 23:32:21 +08:00
KumaJie cbd3e100b8
[cherry-pick]Add CSV file import function (#28161)
issue: #27148

from pr: #27149

Signed-off-by: kuma <675613722@qq.com>
Co-authored-by: kuma <675613722@qq.com>
2023-11-15 18:32:20 +08:00
yah01 6820124366
enhance: remove legacy error related code (#28385) (#28415)
/kind improvement
pr: #28385

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-15 17:24:21 +08:00
SimFG cb8e10b48b
[2.3]: Hide the password info when failing to authorize (#28437)
/kind improvement
pr: #28428

Signed-off-by: SimFG <bang.fu@zilliz.com>
2023-11-15 14:40:26 +08:00
Bennu 600461bef6
[skip e2e] update etcd health check for 2.3 branch (#28443)
pr: #27980

Signed-off-by: Bennu-Li <yunmei.li@zilliz.com>
2023-11-15 11:54:18 +08:00
yah01 c92e6b654f
Reduce 1x memory copy for retrieving data (#28106) (#28217)
/kind improvement
Before this, while retrieving data (query/search), we first copy the
data into a fixed vector, and then copy data from this into the proto
field.
Now we can directly copy the data into the proto field.

This optimization can't be done with int8, int16 due to the proto
doesn't provide the two types, we store them in int32s

Also, this can't be done with variable length field like string, JSON,
see https://github.com/protocolbuffers/protobuf/issues/10866. I tried
but it seems proto doesn't guarantee the memory layout as we expected,
it crashed
pr: #28106

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-15 10:50:20 +08:00
yah01 e36976c474
enhance: modify log to avoid ambiguity and improve readability (#28331) (#28414)
Remove the "failCount" log field, which is ambiguous
replace the status (int32) with string, to improve the readability for
log of task removed
pr: #28331

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-15 10:26:19 +08:00
congqixia e28efc2d34
[2.3] Fix MQ Factory timeout unit test (#28395) (#28431)
Cherry-pick from master
pr: #29395
This PR change timeout case to a longer duration in case of slow runner 
Also change timeout case behavior of using timeout deadline instead of
waiting
See also #28394

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-15 10:10:19 +08:00
congqixia fce0284881
[2.3] Refine datanode Timetick Sender (#28393) (#28430)
cherry pick from master
pr: #28393
- Use explicit lifetime control methods: `Start` and `Stop`
- Allow control retry option
- Make sure tt sender worker exit after `Stop` return

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-15 10:08:26 +08:00
groot 9732ae18be
Update bitset lib to v1.10.0 (#28412)
issue: #27939 
pr: #27984

Signed-off-by: yhmo <yihua.mo@zilliz.com>
2023-11-14 17:40:18 +08:00
ThreadDao d582193772
Add test cases for L0 delete (#28382)
- test delete records in delta logs, WAL, L0 segment

Signed-off-by: ThreadDao <yufen.zong@zilliz.com>
2023-11-14 10:26:24 +08:00
aoiasd 2d4659fe4a
Add write cache for access log (#27792) (#28142)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2023-11-13 15:08:18 +08:00
zhuwenxing c7fe443b93
[test]Remove xfail for a bulk insert testcase (#28283)
Signed-off-by: zhuwenxing <wenxing.zhu@zilliz.com>
2023-11-13 11:48:18 +08:00
nico 9451720a31
update test cases for 2.3 branch (#28335)
Signed-off-by: nico <cheng.yuan@zilliz.com>
2023-11-13 09:34:20 +08:00
sre-ci-robot 2ae60d72f8
Update Pytest image changes (#28321)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2023-11-13 09:32:30 +08:00
sre-ci-robot f414b84ad2
Bump milvus version to v2.3.3 (#28352)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2023-11-10 20:26:21 +08:00
wei liu f2e972abb3
Fix external grpc server doesn't graceful stop (#28310)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-09 22:12:22 +08:00
cai.zhang c8cab4dfc3
No hit when the index exceeds the array length (#28302)
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2023-11-09 22:10:32 +08:00
congqixia 83b7f05fe1
Sync proto generated files to avoid --dirty tag (#28316)
The proto generated files is out of sync for image build env
This will cause --dirty="-dev" tag in Milvus build version
Sync changed files to avoid this case

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-09 21:22:21 +08:00
yah01 0497bbf05d
Handle exception while loading (#28306)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-09 19:28:35 +08:00
nico 4d95a7368d
update pymilvus version (#28299)
Signed-off-by: nico <cheng.yuan@zilliz.com>
2023-11-09 19:04:22 +08:00
congqixia 44063be0a6
Fix buffer FieldData has no `ElementType` and array logsize always zero (#28296)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-09 14:22:36 +08:00
wei liu fe7c87592c
Fix auto balance block channel reassign after datanode restart (#28276)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-09 13:52:19 +08:00
wei liu 16dc26833b
Fix retry when proxy stopped (#28263)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-09 10:10:19 +08:00
PowderLi 39c24fe07b
print azure sdk log (#28241)
Signed-off-by: PowderLi <min.li@zilliz.com>
2023-11-09 10:08:20 +08:00
wei liu d3f149c403
fix unstable auto balance config ut (#28289)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-09 10:02:19 +08:00