This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

So does the blob recall job commit in batches along the way or does it treat the whole recall operation as a single transaction?

We plan to recall about 150 GB of data using the StoragePoint BLOB Recall

We have 200 GB free on the sql data disk partition and 50 GB free on the transaction log partition . Our DBA says the transaction

 log will fill up until you commit and then it will flush. So does this blob recall job commit in batches 

 along the way or does it treat the whole 150 GB operation as a single transaction?  

If the latter is 50 GB free logs enough to let the job run to completion or will we have a problem? Thanks, -Matt

 

  • Hi Matt,

     

    While the job does collect the items in batches, the batches themselves are not created as transactions. The good news is that the recall of each individual BLOB itself is handled as a transaction, so the job will issue a commit after every record update of content back to the SharePoint database table. This is a bigger deal for pre-SharePoint 2013 farms where the varbinary string being written back to the table is as large as it's respective item. With the introduction of shredding in 2013, at the default shred-size allocation it might seem like overkill to commit after every BLOB but it does leave it open for those who elect to specify a larger chunk size for their shreds. I hope this helps!