-
Notifications
You must be signed in to change notification settings - Fork 3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug]: Failed to import data using RemoteBulkWriter #39348
Comments
/assign @xiaocai2333 |
The import operation will wait for the index to be built. The current progress shows 80%, which indicates that it is waiting for the index build. Based on the logs you provided, the system is indeed in the index-building phase. |
/assign @lihuiping1994 |
The waiting time is very long, usually tens or even tens of hours, and the progress still hasn't changed. According to what you said, how long should we wait for 300w data to complete building the index |
/assign @xiaocai2333 |
During the waiting period for the index to be established, after a long time, when using the get_iint_deprogress function to check the task status, there may be an inability to connect to the database and a crash phenomenon |
Can you provide the indexnode log during importing? @lihuiping1994 |
milvus-standalone | [2025/01/17 08:24:05.895 +00:00] [WARN] [datacoord/task_scheduler.go:335] ["task queueing time is too long"] [taskID=455369798530384487] ["queueing time(ms)"=3228983] |
/assign @xiaocai2333 |
@lihuiping1994 for 3 million vectors, it should't be too slow, unless milvus flush many smaller files when importing. due to lack of memories. Suggestions:
There seems to be a lot of smaller segments, I guess it is due to lack of memory and force flush |
also for full log |
Windows 16GB memory, 12 core CPU, running V2.5.3 version on Docker desktop without any parameter restrictions on the container. I tried several times afterwards, with a cycle of about 20 hours, but the progress still did not change. We have now deleted the Docker mapping directory and cannot provide milvus.log |
/assign @xiaocai2333 |
16G Might be not enough for 3million data, it might be ok with 768dim ,but definitely not 1024 or 1536 dim. please offer more logs and we can do nothing without all the detailed logs |
if you are using standalone, I would recommend you to start from at least 12core 32GB memory. |
I mainly want to test the use of RemoteBulkWriter to import data. The native Docker standalone version did not import successfully, so I did not run the import program on a server with better performance, as production environment resources are limited |
the remote bulk writer only prepare the file and write it to S3. |
From the portion of logs you provided, it is evident that there was an issue during the index-building phase, but the exact cause cannot be determined. A complete log is needed for further investigation. Could you provide the full log following this approach? @lihuiping1994 |
When I import 300w simulated data again using the local Docker environment, I will provide complete logs for easy troubleshooting |
Is there an existing issue for this?
Environment
Current Behavior
When I imported 300w data using bulkwritten and checked the status of the collection task using the list_iimport_jobs function, I saw that progress was stuck at 80 in the returned response, "state": "Importing", But the task details are all "state": "Completed" (when the imported data volume is only 1000, there is no situation of being uploaded)
{
"code": 0,
"data": {
"collectionName": "clue_data",
"completeTime": "",
"details": [
{
"completeTime": "2025-01-16T08:42:09Z",
"fileName": "[5.parquet]",
"fileSize": 6776153,
"importedRows": 520161,
"progress": 100,
"state": "Completed",
"totalRows": 520161
},
{
"completeTime": "2025-01-16T08:42:09Z",
"fileName": "[6.parquet]",
"fileSize": 5201131,
"importedRows": 399060,
"progress": 100,
"state": "Completed",
"totalRows": 399060
},
{
"completeTime": "2025-01-16T08:42:09Z",
"fileName": "[2.parquet]",
"fileSize": 6774659,
"importedRows": 520219,
"progress": 100,
"state": "Completed",
"totalRows": 520219
},
{
"completeTime": "2025-01-16T08:42:09Z",
"fileName": "[1.parquet]",
"fileSize": 6773335,
"importedRows": 520238,
"progress": 100,
"state": "Completed",
"totalRows": 520238
},
{
"completeTime": "2025-01-16T08:42:09Z",
"fileName": "[3.parquet]",
"fileSize": 6776574,
"importedRows": 520161,
"progress": 100,
"state": "Completed",
"totalRows": 520161
},
{
"completeTime": "2025-01-16T08:41:55Z",
"fileName": "[4.parquet]",
"fileSize": 6778112,
"importedRows": 520161,
"progress": 100,
"state": "Completed",
"totalRows": 520161
}
],
"fileSize": 39079964,
"importedRows": 3000000,
"jobId": "455347834366209865",
"progress": 80,
"state": "Importing",
"totalRows": 3000000
}
}
{
"code": 0,
"data": {
"records": [
{
"collectionName": "clue_data",
"jobId": "455347834366209865",
"progress": 80,
"state": "Importing"
}
]
}
}
Expected Behavior
After importing 300w of data using bulkwritten, use the list_iimporter function to check the status of the collection task. The progress should be 100
Steps To Reproduce
Milvus Log
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210528]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210529]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210530]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210546]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210547]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210548]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210549]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210550]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210556]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210557]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210558]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210559]
milvus-standalone | [2025/01/16 09:16:29.213 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210560]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210516]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210517]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210518]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210519]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210520]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210507]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210508]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210509]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210510]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210536]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210537]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210538]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/task_index.go:316] ["query task index info successfully"] [taskID=455347834369816525] ["result state"=InProgress] [failReason=]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210539]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/task_scheduler.go:279] ["task is processing"] [taskID=455347834369836027] ["task type"=JobTypeIndexJob] [state=JobStateInit]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_engine_version_manager.go:84] ["Merged current version"] [current=6]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/task_index.go:261] ["index task pre check successfully"] [taskID=455347834369836027] [segID=455347834366210507]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/index_service.go:153] ["receive new flushed segment"] [segmentID=455347834366210540]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [indexnode/indexnode_service.go:210] ["Get Index Job Stats"] [traceID=b34136e54011f7c4396c5e1629385366] [unissued=0] [active=1] [slot=0]
milvus-standalone | [2025/01/16 09:16:29.214 +00:00] [INFO] [datacoord/task_scheduler.go:258] ["there is no idle indexing node, waiting for retry..."]
milvus-standalone | [2025/01/16 09:16:30.213 +00:00] [INFO] [datacoord/task_scheduler.go:248] ["task scheduler"] ["task num"=108]
milvus-standalone | [2025/01/16 09:16:30.213 +00:00] [INFO] [datacoord/task_scheduler.go:279] ["task is processing"] [taskID=455347834369816525] ["task type"=JobTypeIndexJob] [state=JobStateInProgress]
milvus-standalone | [2025/01/16 09:16:30.214 +00:00] [INFO] [datacoord/task_index.go:316] ["query task index info successfully"] [taskID=455347834369816525] ["result state"=InProgress] [failReason=]
milvus-standalone | [2025/01/16 09:16:30.215 +00:00] [INFO] [datacoord/task_scheduler.go:279] ["task is processing"] [taskID=455347834369836027] ["task type"=JobTypeIndexJob] [state=JobStateInit]
milvus-standalone | [2025/01/16 09:16:30.216 +00:00] [INFO] [datacoord/index_engine_version_manager.go:84] ["Merged current version"] [current=6]
milvus-standalone | [2025/01/16 09:16:30.216 +00:00] [INFO] [datacoord/task_index.go:261] ["index task pre check successfully"] [taskID=455347834369836027] [segID=455347834366210507]
milvus-standalone | [2025/01/16 09:16:30.216 +00:00] [INFO] [indexnode/indexnode_service.go:210] ["Get Index Job Stats"] [traceID=86d2c09ad877eda7b158e6569efc5965] [unissued=0] [active=1] [slot=0]
milvus-standalone | [2025/01/16 09:16:30.216 +00:00] [INFO] [datacoord/task_scheduler.go:258] ["there is no idle indexing node, waiting for retry..."]
Anything else?
No response
The text was updated successfully, but these errors were encountered: