Tempdb use

Aleksandar

We've been monitoring how much tempdb space RGSSC uses over a couple of days and it looks like it may be an issue. Over the space of 6 hours, when we had one client open, it seemed to reach 1.4gb and when i closed SSMS it reset back down to 0mb.
I then left RGSSC running for 14 hours and it grew to 6.8gb in size.

Not sure if it makes a difference, but we are using a Shared Dev Model.
This is all done with 1 copy of RGSSC open AND no changes at all being made to any of the databases. When we left 2 copies open for a little while it seems to double, ei it has linear growth for each copy of RGSSC that is open.

So if we have 10 copies of RGSSC open for a 6 hour period this amounts to ~15gb in 6 hours...

Is this the normal behavior of RGSSC? Will throttling back the polling frequency deacrease the growth rate?

Thanks,
-Aleks

James B

Thanks for your post. The growth sounds quite a lot there, but it's going to vary to an extent depending on the specifics of the DB's (and number of DB's linked).
I'd certainly suggest changing the polling to begin with, and see if it helps. You can either increase the interval, or disable it completely to see if it helps.

Shared will cause more of a problem as each session is working against the same server (so the same TempDB will see the activity) but if you're experiencing issues with even one user connected then it sounds like changing the polling setting is a good idea.

kbrandin

I'm having the same issue. Using just 1 copy running is making too many calls for schema changes. Did you find a way to throttle back the polling frequency or turn it off completely and let developers manually get latest?

Aleksandar

yep, throttling it back is nice and easy, but turning off polling is even easier

see here:
http://www.red-gate.com/MessageBoard/vi ... hp?t=12837