I have a process querying this table every 20 seconds
select table_id, item_guid, oplsn_fseqno, oplsn_bOffset, oplsn_slotid
from [MyDatabaseName].[sys].[filetable_updates_1991678143] with (readpast) order by table_id
I only noticed it by chance in SQL Profiler when checking something else.. I have 20+ databases, named TEMP_xxx, which I had restored earlier and the query was being run against all of them, every 20 seconds. It was only running against those TEMP_xxx database and one other (real) database (plenty more database on this server/instance)
I had intended to set all the TEMP_xxx databases to READ_ONLY so decided to do that and (might be coincidence) having set them to READ_ONLY the query stopped - but it is still running against the one, real, database (i.e. the one still set to READ_WRITE).
The table [sys].[filetable_updates_1991678143] does not exist in [MyDatabaseName] (as in: I get an error if I attempt a SELECT * FROM).
The phantom query is always run by LoginName=sa, ApplicationName="", SessionLoginName="" Diffrent SID each time.
Nothing useful came up on Google - or I'm not asking the right question!
I don't know what it is either, but Google brought me to here and here which then points to couple of other articles and so on. I assume you have researched those, but the constant theme seems to be "reboot fixes it"
I'd seen the first one - he says it came back after a couple of weeks after reboot he associated it with AV Groups but we don't have those so might be coincidence for him too.
Second link was new to me, thanks. He has neither Filestream nor AV and makes the comment (which I share ) "I don't think they have much of an impact but they are irritating while looking through sql profiler events" - because the APP etc. columns in SQL Profiler are blank I can't filter them out easily
That thread (several people mentioned they had the same problem) seemed to peter out Oct-2014 which suggest they found a fix. I'm on later SPacks than the @@version posted in that thread (everything bad a more recent, apparently unrelated, HotFix)
This is a brand new server, databases all restored on it over this weekend (and an additional "before upgrade" copy restored as TEMP_xxx). No one using the servers, except me (and an indexing process), but the whole lot is going live at 08:00 tomorrow morning when the users arrive for work so it had better not be a real problem or I will be on Bread & Water for the rest of my life!
he associated it with AV Groups but we don't have those so might be coincidence for him too.
) "I don't think they have much of an impact but they are irritating while looking through sql profiler events" - because the APP etc. columns in SQL Profiler are blank I can't filter them out easily