Reboot required to revitalize server?

I'm having a problem with my database server whereby performance decreases
after 12-18 hours in production and is only restored after a reboot. [almost
like hitting a wall]
Of course, the memory usage increases throughout the day and eventually
settles with a few hundred MB of free memory. SQL Server seems to top off at
1.75GB memory usage. [3GB total] Processor usage never seems to be a
problem, and there are no problematic table locks at the time a reboot is
necessary.
Can anyone point me in the right direction to begin tracing the problem. I
realize there are a million possibilities as to the cause of the problem,
but an "I'd start here..." suggestion would be helpful.
Thanks in advance.Hi
You don't say what your system is doing or how much load there is on it. If
there are a significant number of inserts/updates your indexes could be
fragmented and/or statistics out of date. Check out DBCC SHOWCONTIG and
sp_updatestats in books online.
John
"Chris" wrote:

> I'm having a problem with my database server whereby performance decreases
> after 12-18 hours in production and is only restored after a reboot. [almo
st
> like hitting a wall]
> Of course, the memory usage increases throughout the day and eventually
> settles with a few hundred MB of free memory. SQL Server seems to top off
at
> 1.75GB memory usage. [3GB total] Processor usage never seems to be a
> problem, and there are no problematic table locks at the time a reboot is
> necessary.
> Can anyone point me in the right direction to begin tracing the problem.
I
> realize there are a million possibilities as to the cause of the problem,
> but an "I'd start here..." suggestion would be helpful.
> Thanks in advance.
>
>|||Thanks, I'll begin my research.
"John Bell" <jbellnewsposts@.hotmail.com> wrote in message
news:401CC40F-0E95-4B2B-B18D-4089A531942F@.microsoft.com...
> Hi
> You don't say what your system is doing or how much load there is on it.
> If
> there are a significant number of inserts/updates your indexes could be
> fragmented and/or statistics out of date. Check out DBCC SHOWCONTIG and
> sp_updatestats in books online.
> John
> "Chris" wrote:
>|||Hi
I have been thinking a bit more about this! This sounds more like a lack of
resources. Does stopping/starting SQL Server have the same effect as a
re-boot? This would narrow it down to the process itself!
You may also want to rule out open transactions (DBCC OPENTRAN)
John
"Chris" wrote:

> Thanks, I'll begin my research.
> "John Bell" <jbellnewsposts@.hotmail.com> wrote in message
> news:401CC40F-0E95-4B2B-B18D-4089A531942F@.microsoft.com...
>
>|||I'll try that when/if it happens again. I found that there was some
fragmentation, since my nightly index rebuilding sproc was disabled. Thanks
for the help.
"John Bell" <jbellnewsposts@.hotmail.com> wrote in message
news:E40523F0-980B-4D0F-BBCE-74D564E2DCE4@.microsoft.com...
> Hi
> I have been thinking a bit more about this! This sounds more like a lack
> of
> resources. Does stopping/starting SQL Server have the same effect as a
> re-boot? This would narrow it down to the process itself!
> You may also want to rule out open transactions (DBCC OPENTRAN)
> John
> "Chris" wrote:
>