I'm using Azure Search with a Blob Storage indexer. I'm seeing failures in the execution history:-
[
{
"key": null,
"errorMessage": "Document 'https://mystorage.blob.core.windows.net/my-documents/Document/Repository/F/AD/LO/LO-min-0002-00.png' has unsupported content type 'image/png'"
}
]
Does this failure cause other documents (with supported content type) in the storage not to be indexed?
Yes, by default 1 failed document will stop indexing. You can increase that limit if you just have an occasional unsupported document. You can add AzureSearch_Skip metadata to the blobs you don't want to index as described here.
UPDATE You can also skip all blobs with certain file extensions from indexing - see here for details. This allows you to easily skip all .jpeg or .png files, for example.
Also, please vote for this UserVoice suggestion: Blob indexer should be able to skip unsupported content types instead of treating them as errors
