> We waited for our weekend backup/restore job to run (so as not to
> disturb any users abnormally) and
> as for today, blobs (containing our monolithic .exe) are
> uploading again correctly ... Seems like,
> historically, something happens to our prometheus_update table
> when we occasionally delete gobs of
> blobs (past versions of our software) to make room for new ones.
> (Not room per se -- just ... no
> reason to keep 'em around.) Maybe eventually we'll figure out
> what's corrupting that table, but for
> now we'll just run on the assumption that it happens, and a
> backup/restore fixes it ... And no, it
> wasn't IBX's fault. Any other ideas? I kinda wish a more obvious
> error had been thrown, indicating
> something was actually wrong with the database, not just ...
> random lack of uploading.
>