issue: Milvus query error when fetching more than 16384 entries #5978

New Issue

GiteaMirror · 2025-11-11T16:41:16-06:00

GiteaMirror commented

2025-11-11 16:41:16 -06:00

Originally created by @julien-oss on GitHub (Aug 6, 2025).

Check Existing Issues

I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

0.6.18

Ollama Version (if applicable)

No response

Operating System

Ubuntu 24.04

Browser (if applicable)

No response

Confirmation

I have read and followed all instructions in README.md.
I am using the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have provided every relevant configuration, setting, and environment variable used in my setup.
I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
Start with the initial platform/version/OS and dependencies used,
Specify exact install/launch/configure commands,
List URLs visited, user input (incl. example values/emails/passwords if needed),
Describe all options and toggles enabled or changed,
Include any files or environmental changes,
Identify the expected and actual result at each stage,
Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

When querying more than 16384 entries, milvus.py should iterate queries and return top k entries

Actual Behavior

After the first iteration, offset is equal 16384 and limit 16384, resulting in error MilvusException: <MilvusException: (code=65535, message=invalid max query result window, (offset+limit) should be in range [1, 16384], but got 32 768)>

Steps to Reproduce

Have a collection > 16384 in milvus
RAG query on this collection

Logs & Screenshots

MilvusException: <MilvusException: (code=65535, message=invalid max query result window, (offset+limit) should be in range [1, 16384], but got 32 768)>

Additional Information

I edited milvus.py to use collection.query_iterator() instead of client.query()

Add in milvus.py :
from pymilvus import connections, Collection

And change query to :

def query(self, collection_name: str, filter: dict, limit: Optional[int] = None):
        connections.connect(uri=MILVUS_URI,token=MILVUS_TOKEN,db_name=MILVUS_DB)
        if limit is None:
            limit = ( 16384 * 10 )
        collection_name = collection_name.replace("-", "_")
        if not self.has_collection(collection_name):
            log.warning(
                f"Query attempted on non-existent collection: {self.collection_prefix}_{collection_name}"
            )
            return None

        filter_string = " && ".join(
            [
                f'metadata["{key}"] == {json.dumps(value)}'
                for key, value in filter.items()
            ]
        )
        collection=Collection(name=f"{self.collection_prefix}_{collection_name}")
        collection.load()
        all_results = []

        try:
            log.info(
                f"Querying collection {self.collection_prefix}_{collection_name} with filter: '{filter_string}', limit: {limit}"
            )

            # Use query_iterator to efficiently handle potentially large result sets
            iterator = collection.query_iterator(
                filter=filter_string,
                output_fields=[
                    "id",
                    "data",
                    "metadata",
                ],
                limit=limit, # Pass the limit directly; None means no limit.
            )
            # The iterator yields results in batches
            results = []
            while True:
                result = iterator.next()
                if not result:
                    iterator.close()
                    break
                all_results += result
            log.info(f"Total results from query: {len(all_results)}")
            return self._result_to_get_result([all_results])

        except Exception as e:
            log.exception(
                f"Error querying collection {self.collection_prefix}_{collection_name} with filter '{filter_string}' and limit {limit}: {e}"
            )
            return None

Now iteration works and we get the expected top k result

Originally created by @julien-oss on GitHub (Aug 6, 2025). ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version 0.6.18 ### Ollama Version (if applicable) _No response_ ### Operating System Ubuntu 24.04 ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior When querying more than 16384 entries, milvus.py should iterate queries and return top k entries ### Actual Behavior After the first iteration, offset is equal 16384 and limit 16384, resulting in error `MilvusException: <MilvusException: (code=65535, message=invalid max query result window, (offset+limit) should be in range [1, 16384], but got 32 768)> ` ### Steps to Reproduce Have a collection > 16384 in milvus RAG query on this collection ### Logs & Screenshots `MilvusException: <MilvusException: (code=65535, message=invalid max query result window, (offset+limit) should be in range [1, 16384], but got 32 768)>` ### Additional Information I edited milvus.py to use collection.query_iterator() instead of client.query() Add in milvus.py : `from pymilvus import connections, Collection` And change query to : ``` def query(self, collection_name: str, filter: dict, limit: Optional[int] = None): connections.connect(uri=MILVUS_URI,token=MILVUS_TOKEN,db_name=MILVUS_DB) if limit is None: limit = ( 16384 * 10 ) collection_name = collection_name.replace("-", "_") if not self.has_collection(collection_name): log.warning( f"Query attempted on non-existent collection: {self.collection_prefix}_{collection_name}" ) return None filter_string = " && ".join( [ f'metadata["{key}"] == {json.dumps(value)}' for key, value in filter.items() ] ) collection=Collection(name=f"{self.collection_prefix}_{collection_name}") collection.load() all_results = [] try: log.info( f"Querying collection {self.collection_prefix}_{collection_name} with filter: '{filter_string}', limit: {limit}" ) # Use query_iterator to efficiently handle potentially large result sets iterator = collection.query_iterator( filter=filter_string, output_fields=[ "id", "data", "metadata", ], limit=limit, # Pass the limit directly; None means no limit. ) # The iterator yields results in batches results = [] while True: result = iterator.next() if not result: iterator.close() break all_results += result log.info(f"Total results from query: {len(all_results)}") return self._result_to_get_result([all_results]) except Exception as e: log.exception( f"Error querying collection {self.collection_prefix}_{collection_name} with filter '{filter_string}' and limit {limit}: {e}" ) return None ``` Now iteration works and we get the expected top k result

GiteaMirror added the bug label 2025-11-11 16:41:16 -06:00

GiteaMirror closed this issue

2025-11-11 16:41:16 -06:00

GiteaMirror commented

2025-11-11 16:41:17 -06:00

@tjbck commented on GitHub (Aug 6, 2025):

PR welcome!

@tjbck commented on GitHub (Aug 6, 2025): PR welcome!

GiteaMirror commented

2025-11-11 16:41:17 -06:00

@rgaricano commented on GitHub (Aug 8, 2025):

for reference:
https://github.com/milvus-io/milvus/issues/36895
https://milvus.io/docs/with-iterators.md

@rgaricano commented on GitHub (Aug 8, 2025): for reference: https://github.com/milvus-io/milvus/issues/36895 https://milvus.io/docs/with-iterators.md

GiteaMirror commented

2025-11-11 16:41:18 -06:00

@tjbck commented on GitHub (Aug 13, 2025):

Addressed with ad98d4300b5e0dec0beb4db157873efde8577e1a!

@tjbck commented on GitHub (Aug 13, 2025): Addressed with ad98d4300b5e0dec0beb4db157873efde8577e1a!

GiteaMirror referenced this issue

2026-04-19 20:38:15 -05:00

[GH-ISSUE #5978] ERR_NETWORK_CHANGED in Microsoft Edge - no other browsers (new since >0.3.30) #14194

GiteaMirror referenced this issue

2026-04-25 04:09:11 -05:00

[GH-ISSUE #5978] ERR_NETWORK_CHANGED in Microsoft Edge - no other browsers (new since >0.3.30) #29722

GiteaMirror referenced this issue