issue: #33103
when try to do stopping balance for stopping query node, balancer will
try to get node list from replica.GetNodes, then check whether node is
stopping, if so, stopping balance will be triggered for this replica.
after the replica refactor, replica.GetNodes only return rwNodes, and
the stopping node maintains in roNodes, so balancer couldn't find
replica which contains stopping node, and stopping balance for replica
won't be triggered, then query node will stuck forever due to
segment/channel doesn't move out.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #32910
* split replica's node list to channels when create replicas
* balance nodes among channels when node change happens
* implement channel level balance, let balance happens in channel level
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #32466
this PR enhance that when shard location changed, update proxy's shard
leader cache. in case of query node failover case, proxy can find
replica recover
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
related to #32165
1. for all the manager, support collection level index
2. remove collection level filter to avoid extra cpu usage when
collection number increases
Signed-off-by: xiaofanluan <xiaofan.luan@zilliz.com>
See also #32440
- Add loadTask in collection observer
- For load collection/partitions, load task shall timeout as a whole
- Change related constructor to load jobs
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #30647
- Add declarative resource group api
- Add config for resource group management
- Resource group recovery enhancement
---------
Signed-off-by: chyezh <chyezh@outlook.com>
issue: #31091
This PR add GetByFilter interface in leader view manager, instead of all
kind of get func
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #30647
- ReplicaManager manage read only node now, and always do persistent of
node distribution of replica.
- All segment/channel checker using ReplicaManager to get read-only node
or read-write node, but not ResourceManager.
- ReplicaManager promise that only apply unique querynode to one replica
in same collection now (replicas in same collection never hold same
querynode at same time).
- ReplicaManager promise that fairly node count assignment policy if
multi replicas of collection is assigned to one resource group.
- Move some parameters check into ReplicaManager to avoid data race.
- Allow transfer replica to resource group that already load replica of
same collection
- Allow transfer node between resource groups that load replica of same
collection
---------
Signed-off-by: chyezh <chyezh@outlook.com>
issue: #31091
This PR add GetByFilter interface in channel dist manager, instead of
all kind of get func
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
See also #31103
Since querycoord need index meta information from datacoord only, broker
shall use `ListIndexes` to skip segment index building check logic in
datacoord
This PR is also related to #30538, in which DescribeIndex caused lots of
memory usage and lead to OOM eventually
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
This PR mainly improve two items:
1. Target observer should refresh loading status during init time. An
uninitialized loading status blocks search/query. Currently, the target
observer refreshes every 10 seconds, i.e. we'd need to wait for 10s for
no reason. That's also the reason why we constantly see false log
"collection unloaded" upon mixcoord restarts.
2. Delete session when service is stopped. So that the new service
doesn't need to wait for the previous session to expire (~10s).
Item 1 is the major improvement of this PR, which should speed up init
time by 10s.
Item 2 is not a big concern in most cases as coordinators usually shut
down after stop(). In those cases, coordinator restart triggers serverID
change which further triggers an existing logic that deletes expired
session. This PR only fixes rare cases where serverID doesn't change.
integration test:
`go test -tags dynamic -v -coverprofile=profile.out -covermode=atomic
tests/integration/coordrecovery/coord_recovery_test.go -timeout=20m`
Performance after the change:
Average init time of coordinators: 10s
Hardware: M2 Pro
Test setup: 1000 collections with 1000 rows (dim=128) per collection.
issue: #29409
Signed-off-by: yiwangdr <yiwangdr@gmail.com>
issue: #29453
sync distribution by rpc will also call loadSegment/releaseSegment,
which may cause all kinds of concurrent case on same segment, such as
concurrent load and release on one segment.
This PR add leader_checker which generate load/release task to correct
the leader view, instead of calling sync distribution by rpc
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
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>
Related to #29575
Add `getCollectionTarget` method which is atomic when scope is
`CurrentTargetFirst` or `NextTargetFirst`
Also return error when executor finds no channel in target manager
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
we found the load got stuck probably, and reviewed the logs.
the target observer seems not working, the reason is the taskDispatcher
removes the task in a goroutine, and modifies the task status after
committing the task into the goroutine pool, but this may happen after
the task removed, which leads to the task will never be removed
related #29086
Signed-off-by: yah01 <yang.cen@zilliz.com>
we consume the delta data from the lastest channel checkpoint while
loading segment,
this works well without level 0 segments, but now it may lead to miss
some delta data,
so we have to consume from the current target's channel checkpoint
related: #27349
---------
Signed-off-by: yah01 <yah2er0ne@outlook.com>
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>
- Add `taskDispatcher` to submit and run task async safely
- Change `LeaderObeserver` and `TargetObserver` schedule and manual check action to submitting task into dispatcher
- Fix logic problem in collection observer when manual check return false
See also #27494
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
SimFG
wei liu
chyezh
Xiaofan
congqixia
yiwangdr
yah01
yah01
aoiasd