Blob Indexer should continue indexing documents after encountering an error
This is related to...
...whilst a change has been added to easily configure the indexer to exclude unsupported document formats, this is inadequate. Bugs in the indexer and corrupt documents are going to ensure that there will always be a percentage of documents that can't be processed by the indexer. This shouldn't stop the indexer from processing data it can index.
When the indexer encounters an error processing a document it should log the error then proceed with indexing of the rest of the documents in storage.
A run shouldn't be halted until the indexer has attempted to index all the documents in the data source.
is this issue resolved? I am also facing the same issue. The indexers stops after it encounters a corrupted file and doesnt proceed further
michael lang commented
I have an indexer which is failing on a .docx document. I can download and open the document in word, it doesn't appear to be corrupt. When an indexing thousands of documents, my expectation is that a high percentage of the documents will be successfully indexed, I would never expect 100% of documents that should be index-able to be successfully indexed.