Reputation: 11
Hello Azure Search Team, We have been experiencing an issue in our facade using .Net SDK to Azure Search service that search result pagination includes duplicate documents.
Everything was working fine when we have 1 instance of AzureSearch service, index database without partitioning and replication up until yesterday. But, we just changed the replication to 3 and partitioning to 2 for performance improvement, we started seeing such behavior.
Our query is simple such as search=*&$skip=0&$top=50 (Page1), search=*&$skip=50&$top=50 (Page2)
, now page 2 has some of the content from page1.
Can someone suggest what could be the cause of this behavior? Our expectation was irrespective of number of results if we use correct skip and top we should be able to browse through content.
NOTE:
Steps we have already verified that index data doesn't have duplicate entries.
It's working as expected when service instance (Azure Search) is 1 and no replication/partitioning.
We also eliminated facade from the picture and directly using portal to query Azure Search service.
Upvotes: 1
Views: 607