Hi,
I've a strange behavior on my server.
The response time is good after a reboot of the computer, but, after a long
work on it (half day or a day)
The performance slow down, so I reboot the server to recover my performance.
Why?
How to diagnostic this problem?
Thanks.
Jerome.
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.595 / Virus Database: 378 - Release Date: 2004-02-25Have a look at perfmon
If you have to do this every day/after using it a while then a resource is
being depleted seriously and a reboot frees it up again.
Look at Memory, Processor to start with.
What version and SP of SQL Server ?
--
Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP)
www.SQLDTS.com - The site for all your DTS needs.
I support PASS - the definitive, global community
for SQL Server professionals - http://www.sqlpass.org
"Jj" <willgart@._A_hAotmail_A_.com> wrote in message
news:e6Ke4vrCEHA.2908@.TK2MSFTNGP09.phx.gbl...
> Hi,
> I've a strange behavior on my server.
> The response time is good after a reboot of the computer, but, after a
long
> work on it (half day or a day)
> The performance slow down, so I reboot the server to recover my
performance.
> Why?
> How to diagnostic this problem?
> Thanks.
> Jerome.
>
> --
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.595 / Virus Database: 378 - Release Date: 2004-02-25
>|||Fairly Normal with Windows
"Jj" <willgart@._A_hAotmail_A_.com> wrote in message
news:e6Ke4vrCEHA.2908@.TK2MSFTNGP09.phx.gbl...
> Hi,
> I've a strange behavior on my server.
> The response time is good after a reboot of the computer, but, after a
long
> work on it (half day or a day)
> The performance slow down, so I reboot the server to recover my
performance.
> Why?
> How to diagnostic this problem?
> Thanks.
> Jerome.
>
> --
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.595 / Virus Database: 378 - Release Date: 2004-02-25
>|||I've the SP3 on a Windows 2000 Server + SP4 server.
I've 3 drives, and I've splitted differents database files on these 3 drives
(tempdb on 1; data of my data warehouse on 2; data warehouse log file on 3)
"Allan Mitchell" <allan@.no-spam.sqldts.com> a crit dans le message de
news:OCcnxMCDEHA.580@.TK2MSFTNGP11.phx.gbl...
> Have a look at perfmon
> If you have to do this every day/after using it a while then a resource is
> being depleted seriously and a reboot frees it up again.
> Look at Memory, Processor to start with.
> What version and SP of SQL Server ?
>
> --
> --
> Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP)
> www.SQLDTS.com - The site for all your DTS needs.
> I support PASS - the definitive, global community
> for SQL Server professionals - http://www.sqlpass.org
>
> "Jj" <willgart@._A_hAotmail_A_.com> wrote in message
> news:e6Ke4vrCEHA.2908@.TK2MSFTNGP09.phx.gbl...
> long
> performance.
>
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.595 / Virus Database: 378 - Release Date: 2004-02-25
Showing posts with label reboot. Show all posts
Showing posts with label reboot. Show all posts
Wednesday, March 28, 2012
Monday, February 20, 2012
performance monitor lifetime
from what i've read so far, the performance monitor's data is not
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthurThere's no one setting that control all counters. You have to read the counter definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache counters that now
averages from about 2000 to 3000 last samples (changed in sp3).
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegroups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
--
David Gugick
Imceda Software
www.imceda.com
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthurThere's no one setting that control all counters. You have to read the counter definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache counters that now
averages from about 2000 to 3000 last samples (changed in sp3).
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegroups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
--
David Gugick
Imceda Software
www.imceda.com
performance monitor lifetime
from what i've read so far, the performance monitor's data is not
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthur
There's no one setting that control all counters. You have to read the counter definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache counters that now
averages from about 2000 to 3000 last samples (changed in sp3).
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegr oups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>
|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
David Gugick
Imceda Software
www.imceda.com
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthur
There's no one setting that control all counters. You have to read the counter definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache counters that now
averages from about 2000 to 3000 last samples (changed in sp3).
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegr oups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>
|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
David Gugick
Imceda Software
www.imceda.com
performance monitor lifetime
from what i've read so far, the performance monitor's data is not
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthurThere's no one setting that control all counters. You have to read the count
er definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache c
ounters that now
averages from about 2000 to 3000 last samples (changed in sp3).
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegroups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
David Gugick
Imceda Software
www.imceda.com
instantaneous, rather collected historically since the last reboot. is
this correct? i've also read (though i can't remember where), that
there was some win2k/service pack combination that allowed for
instantaneous monitoring. i've got a feeling the monitors values are
getting skewed by past (good) performance and would like to see
instantaneous values (w/out having to reboot). thanks,
arthurThere's no one setting that control all counters. You have to read the count
er definition about how
it is measured. The ones that you are probably thinking of is Buffer Cache c
ounters that now
averages from about 2000 to 3000 last samples (changed in sp3).
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
http://www.sqlug.se/
"arthur" <alangham@.gmail.com> wrote in message
news:1107372326.151669.237260@.f14g2000cwb.googlegroups.com...
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
>|||arthur wrote:
> from what i've read so far, the performance monitor's data is not
> instantaneous, rather collected historically since the last reboot. is
> this correct? i've also read (though i can't remember where), that
> there was some win2k/service pack combination that allowed for
> instantaneous monitoring. i've got a feeling the monitors values are
> getting skewed by past (good) performance and would like to see
> instantaneous values (w/out having to reboot). thanks,
> arthur
Performance monitor values vary by counter type. You can read about them
in MSDN. If you are dealing with a counter that does not display an
instantaneous value, all you need to do is compare it to the previous
value. That's essentially what sp_monitor does (as does perfmon). That
is, it keeps track of previous values and capture date/time in order to
perform the necessary calculations. Rebooting the server won't help
because once you do, you're back to have to calculate each time.
David Gugick
Imceda Software
www.imceda.com
Subscribe to:
Posts (Atom)