November 26, 2019 at 1:46 am
Has anyone any idea why we might experience connection timeouts to a SQL 2016 box when a SQL log backup is running. The SQL box has plenty of resources and uses an EMC backup product. Its an application that uses filestream pretty heavily. The IO susbsystem response times are OK. Memory, network & CPU arent stressed.
Any ideas on how to start looking at this?
November 26, 2019 at 12:52 pm
A timeout is caused when resources are unavailable, causing excessive waits. So, something is going south there. You've narrowed it down to the EMC backup solution running. What do the wait statistics look like during this process? Further, you could use extended events to find out exactly what waits the backup command itself is experiencing. This will help to pretty rapidly narrow down what resource is running in short supply. Probably, it's I/O, disk most likely, but possibly memory, or both. However, go gather the knowledge. That'll tell you where to focus your efforts.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
November 26, 2019 at 3:16 pm
If it's a SAN then you might have the same problem that we had with Amazon
depends on your configuration
our connectivity was shared over a network connection which was also shared with our disks, when the disks got busy then we lost our network connection
MVDBA
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply