Commit Graph

91 Commits (cdc_test)

Author SHA1 Message Date
congqixia 99774548f2
enhance: [Cherry-pick] Add AllPartitionsID const to replace InvalidPartitionID (#31438) (#31515)
Cherry-pick from master
pr: #31438

"-1" as `InvalidPartitionID` previously used as All partition place
holder in delete cases. It's confusing and hard to maintain when a const
var has more than one meaning.

This PR add `AllPartitionsID` to replace these usages in delete
scenarios.

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-03-22 16:37:08 +08:00
chyezh 775129bd4e
enhance: add resource usage estimate for segment interface (#31309)
issue: #30931
pr: #31050

- move resource estimate function outside from segment loader.
- add load info and collection to base segment.
- add resource usage method for sealed segment.

Signed-off-by: chyezh <chyezh@outlook.com>
2024-03-21 09:27:07 +08:00
congqixia 4e83ca46bc
fix: [cherry-pick] Report offline info in `GetDataDistribition` (#31349)
Cherry pick from master
pr: #31347
See also #31345

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-03-18 17:05:05 +08:00
congqixia ba8197fafb
fix: Filter channel level zero segments when build level delete cache (#31129)
See also #31125

Delegator shall build level zero delete cache from l0 segments belongs
to it. Previously it build cache from all existing level zero segments
in the querynode which may lead to high memory usage and even panicking
when pk types are not matched

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-03-07 23:01:02 +08:00
smellthemoon a4f3e01a3a
fix: add the range search params check in proxy (#30423)
if check in Segcore, will not do the it when not insert data.
so, check "radius" and "range_filter" in proxy.
related with #30365

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-02-28 11:24:58 +08:00
xige-16 060c8603a3
fix: Support mvcc with hybrid serach (#30114)
issue: https://github.com/milvus-io/milvus/issues/29656
/kind bug

Signed-off-by: xige-16 <xi.ge@zilliz.com>

---------

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2024-02-01 16:03:03 +08:00
congqixia 7c086a4608
enhance: Set delete scope for LoadSegment streaming data (#30245)
See also #29474

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-01-30 11:13:02 +08:00
chyezh d300bc7bcb
fix: querynode num entity metric is broken by illegal label (#29948)
issue: #29766
also see pr: #29825

Signed-off-by: chyezh <ye.zhen@zilliz.com>
2024-01-14 10:23:00 +08:00
congqixia adf0c8885c
enhance: add trace span for wait tsafe (#29911)
Add tracing span for search/query operation waiting tsafe duration

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-01-12 14:03:04 +08:00
yah01 d357139064
fix: the entities num metric may be contributed more than once (#29767)
the growing segments contribute to this metric while inserting and
putting into the manager, but the current impl inserts data before
putting the segments into manager, which leads to double contributions

fix: #29766

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2024-01-10 10:00:51 +08:00
zhenshan.cao 60e88fb833
fix: Restore the MVCC functionality. (#29749)
When the TimeTravel functionality was previously removed, it
inadvertently affected the MVCC functionality within the system. This PR
aims to reintroduce the internal MVCC functionality as follows:

1. Add MvccTimestamp to the requests of Search/Query and the results of
Search internally.
2. When the delegator receives a Query/Search request and there is no
MVCC timestamp set in the request, set the delegator's current tsafe as
the MVCC timestamp of the request. If the request already has an MVCC
timestamp, do not modify it.
3. When the Proxy handles Search and triggers the second phase ReQuery,
divide the ReQuery into different shards and pass the MVCC timestamp to
the corresponding Query requests.

issue: #29656

Signed-off-by: zhenshan.cao <zhenshan.cao@zilliz.com>
2024-01-09 11:38:48 +08:00
congqixia fe47deebf3
fix: Set & Return correct SegmentLevel in querynode segment manager (#29740)
See also #27349

The segment level label in querynode used `Legacy` before segment level
was correctly passed in Load request. Now this attribute is still using
legacy so the metrics does not look right.

This PR add paramter for `NewSegment` and passes corrent values for each
invocation.

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-01-08 14:16:48 +08:00
yah01 a0cec4047a
fix: make the entity num metric accurate (#29643)
fix #29642

Signed-off-by: yah01 <yang.cen@zilliz.com>
2024-01-05 18:24:47 +08:00
congqixia da7c3cbd88
enhance: make delegator delete buffer holding all delete from cp (#29626)
See also #29625

This PR:
- Add a new implemention of `DeleteBuffer`: listDeleteBuffer
  - holds cacheBlock slice
  - `Put` method append new delete data into last block
  - when a block is full, append a new block into the list
- Add `TryDiscard` method for `DeleteBuffer` interface
  - For doubleCacheBuffer, do nothing
- For listDeleteBuffer, try to evict "old" blocks, which are blocks
before the first block whose start ts is behind provided ts
- Add checkpoint field for `UpdateVersion` sync action, which shall be
used to discard old cache delete block

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-01-04 17:02:46 +08:00
MrPresent-Han ed644983e2
enhance: add param for bloomfilter(#29388) (#29490)
related: #29388

Signed-off-by: MrPresent-Han <chun.han@zilliz.com>
2023-12-28 18:10:46 +08:00
congqixia b251c3a682
enhance: add ctx for HandleCStatus and callers (#29517)
See also #29516

Make `HandleCStatus` print trace id for better logging

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-12-27 16:10:47 +08:00
congqixia 02bc0d0dd5
fix: Add scope limit for querynode DeleteRequest (#29474)
See also #27515

When Delegator processes delete data, it forwards delete data with only
segment id specified. When two segments has same segment id but one is
growing and the other is sealed, the delete will be applied to both
segments which causes delete data out of order when concurrent load
segment occurs.

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-12-26 14:28:47 +08:00
congqixia ac95c52171
enhance: change protection to RLock for loadStreamDelete (#29450)
See also: #29332

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-12-25 23:27:02 +08:00
congqixia 1eacdc591b
fix: delegator may mark segment offline by mistake (#29343)
See also #29332

The segment may be released before or during the request when delegator
tries to forward delete request to yet. Currently, these two situation
returns different error code.

In this particular case, `ErrSegmentNotLoaded` and `ErrSegmentNotFound`
shall both be ignored preventing return search service unavailable by
mistake.

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-12-20 21:22:43 +08:00
yah01 9b3e06ae86
enhance: add more metrics for level zero segments (#29029)
- Add SegmentNum metric for level zero segments
- Add level zero segments size metirc

---------

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-12-07 14:34:35 +08:00
Gao 3e77365de5
fix: correct autoindex segment num (#28387)
Fix #28386 
Current code snippet 
```
// get delegator
sd, ok := node.delegators.Get(channel)
if !ok {
err := merr.WrapErrChannelNotFound(channel)
log.Warn("Query failed, failed to get shard delegator for search", zap.Error(err))
return nil, err
}
req, err = node.optimizeSearchParams(ctx, req, sd)
if err != nil {
log.Warn("failed to optimize search params", zap.Error(err))
return nil, err
}
// do search
results, err := sd.Search(searchCtx, req)
```

We could move these into `ShardDelegator`, and directly use sealed
segment num in `Search` methods, also segment num got outside could be
wrong when we specify partitions.

Signed-off-by: chasingegg <chao.gao@zilliz.com>
2023-11-22 11:12:22 +08:00
yah01 cc952e0486
enhance: optimize forwarding level0 deletions by respecting partition (#28456)
- Cache the level 0 deletions after loading level0 segments
- Divide the level 0 deletions by partition
related: #27349

---------

Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-21 18:24:22 +08:00
yah01 d20ea061d6
Fix panic while forwarding empty deletions to growing segment (#28213)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-08 16:42:21 +08:00
yah01 ece592a42f
Deliver L0 segments delete records (#27722)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-11-07 01:44:18 +08:00
wei liu ecec5dfcfd
fix retry on offline node (#28079)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-11-03 10:14:16 +08:00
congqixia e4fdf5e68e
Refine offline segments logic in shard delegator (#28073)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-11-01 20:04:24 +08:00
congqixia 13877a07ff
Add ctx parameter for tsafe pkg & NewDelegator method (#27877)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-10-26 19:14:10 +08:00
wei liu 8041fc3c75
avoid add empty growing segment to delegator distribution (#27930)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-10-26 10:10:10 +08:00
aoiasd 9ba79a97e4
Add some log for delete task when it need query (#27317)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2023-10-24 14:54:12 +08:00
yihao.dai d9431266bb
Fix panic at loadSegment during rolling upgrade (#27671)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2023-10-13 16:31:35 +08:00
yah01 be980fbc38
Refine state check (#27541)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-10-11 21:01:35 +08:00
wei liu 0695c88844
refine_log_of_update_target_version (#27572)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-10-11 06:51:33 +08:00
congqixia 258e1ccd66
Refine querynode scheduler lifetime (#26915)
This PR refines scheduler lifetime control:
- Move private tri-state into lifetime package
- Make scheduler block incoming "Add" task
- Make scheduler Stop wait until all previously accepted task done

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-28 10:21:26 +08:00
congqixia 8c59dba329
Refine queryHook mockery (#27394)
This PR move `QueryHook` interface to `optimizers` pkg
Update all mockery generated files to latest
Add makefile entry for `QueryHook`

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-28 10:01:26 +08:00
SimFG 26f06dd732
Format the code (#27275)
Signed-off-by: SimFG <bang.fu@zilliz.com>
2023-09-21 09:45:27 +08:00
congqixia cc9974979f
Add staticcheck linter and fix existing problems (#27174)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-19 10:05:22 +08:00
Xiaofan e51cb739eb
Remove some logs in querynode delegator (#27156)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2023-09-18 10:57:20 +08:00
aoiasd e107d0794c
support complex delete expression (#25752)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2023-09-12 10:19:17 +08:00
congqixia 2a5d574a0d
Fix querynodev2 concurrent load logic (#26959)
Fix logic error from #26926
function `waitSegmentLoadDone` shall return error when context is done

Make delegator control concurrency for each same segment
Related to #26908

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-10 07:41:18 +08:00
XuanYang-cn 7f1ae35e72
Add timeout in dispatcher, AsConsumer and Seek (#26686)
See also: #25309

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2023-09-08 09:51:17 +08:00
congqixia af5c01082b
Refine delegator lifetime control (#26881)
- Add SafeChan interface in lifetime package
- Embed SafeChan into  interface
- Replace private lifetime struct in delegator package with
- Refine delegator on-going task lifetime control and wait all accepted task done
- Fix potential goroutine leakage from  if delegator closed concurrently

/kind improvement

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-07 10:11:15 +08:00
MrPresent-Han 528948559f
fix false load failure for long unserviable period(#26813) (#26818)
Signed-off-by: MrPresent-Han <chun.han@zilliz.com>
2023-09-06 12:57:15 +08:00
congqixia 4b58c71908
Add ctx parameter for organizeTask and GetWorker method (#26835)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-05 10:05:48 +08:00
congqixia 1a8cf5c415
Organize all mockery generation commands in Makefile (#26826)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-09-04 21:19:48 +08:00
yah01 3349db4aa7
Refine errors to remove changes breaking design (#26521)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-09-04 09:57:09 +08:00
congqixia 89fc9aad82
Improve sync target version logic (#26630)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-08-29 23:12:27 +08:00
wei liu 7af0f7d90c
avoid concurrent sub/unsub on same channel (#26454)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-08-23 10:10:22 +08:00
congqixia 855a04539a
Set delegator lastUpdateTs to startTs (#26416)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-08-17 12:24:18 +08:00
wei liu 23baecd70f
set sealed segment to unreadable before sync target version (#26338)
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2023-08-15 17:27:35 +08:00
Enwei Jiao 7d61355ab0
Refactor log for Query (#26310)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-08-14 18:57:32 +08:00