I would like to verify something about a backlog I got yesterday in a big application.

All the hits are sent from the Web site to a SQL Server server. Both are two different VM servers.

We process at about 10 to 15 hits a second. We ended up yesterday at 550,000 hits. This is average hits during these days.

While many users were working ok on the site, two users encountered a situation that made them to not being able to work for about an hour. My analysis shows that it is not the hit itself that caused the situation. Something else triggered the situation. I am trying to know if SQL Server could be involved.

So, as mentioned, hundreds of users continued to do their work while two were not able to. All there hits were stuck on the backend and were all delivered one hour later with this message:

"Thread was being aborted."

My log error shows that all those hits, ended up in the error log all delivered in the same second. Those are hits that were sent during about an hour. One of the two users was in automation and was sending a very small hit also ended up in that situation. This is the smallest hit of the Web site being returned instantly. However, those hits, I can see them in the Hit log. They were in there for an hour and finally got released in the error log all at the same time.

This was caused by something external. I am trying to know if this could be related to SQL Server, holding off hits like that. I do not think it is IIS because in order for me to get the log, in the Error table, this is being written by SQL Server. So, if it would have been an IIS issue, The records in the Error table would indicate a progressive timestamp. In this case, they were stuck in SQL Server.

What could have cause this?