Hi Joe, I've just tried the manual clean up using NSVacuum, and that works fine, it did 32 in 30 seconds. It's running at the moment, but has approx 900,000 quantums to clean out. Thanks, John [quoted text, click to view] "Joe Webb" wrote: > Hi John - > > What happens when you disable the scheduled vacuumer and run it > manually? > > I'm not suggesting this is a production-level resolution. But I'd like > to see whether the issue is in the vacuumer determining what to get > rid of, or whether the vacuumer is not running at all. > > HTH... > > -- > Joe Webb > SQL Server MVP > > > ~~~ > Get up to speed quickly with SQLNS > http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 > > I support PASS, the Professional Association for SQL Server. > ( www.sqlpass.org) > > > > > On Thu, 15 Sep 2005 08:48:38 +0100, "J" <J@somewhere.cam> wrote: > > >Hi, > > > >I'm having trouble with the NS<product>Events table not being cleaned down. > >Initially I thought it was because the vacuum event was clashing with the > >backup procedure on the server, so I moved the vacuum event to a time 5 > >hours before hand. Unfortunatley it made no difference, and the number of > >events in the table actually increased! > > > >Has anyone got any ideas as to what maybe happening? > > > >The retention age is set to 1 day, and the event will go off at 9pm for 3 > >hrs every day. > > > >Many Thanks, > >John > >
Hi John - What happens when you disable the scheduled vacuumer and run it manually? I'm not suggesting this is a production-level resolution. But I'd like to see whether the issue is in the vacuumer determining what to get rid of, or whether the vacuumer is not running at all. HTH... -- Joe Webb SQL Server MVP ~~~ Get up to speed quickly with SQLNS http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 I support PASS, the Professional Association for SQL Server. ( www.sqlpass.org) [quoted text, click to view] On Thu, 15 Sep 2005 08:48:38 +0100, "J" <J@somewhere.cam> wrote: >Hi, > >I'm having trouble with the NS<product>Events table not being cleaned down. >Initially I thought it was because the vacuum event was clashing with the >backup procedure on the server, so I moved the vacuum event to a time 5 >hours before hand. Unfortunatley it made no difference, and the number of >events in the table actually increased! > >Has anyone got any ideas as to what maybe happening? > >The retention age is set to 1 day, and the event will go off at 9pm for 3 >hrs every day. > >Many Thanks, >John
Hi Joe, Answers to your questions :- 1. Yup, here's the section from the config file: <Vacuum> <RetentionAge>P1DT00H00M00S</RetentionAge> <VacuumSchedule> <Schedule> <StartTime>21:00:00</StartTime> <Duration>P0DT03H00M00S</Duration> </Schedule> </VacuumSchedule> </Vacuum> 2. Yup :) 3. v2.0 4. I'll have to let you know that one tomorrow... I just checked, and the State returned the time from todays manual run, and the Schedule said: 2, 01/01/1900, 21:00:00, 10800 Thanks,
Hi Joe, No problem, we're in Southampton, England. So we shouldn't need to adjust the time :) I have a thoery as to why it doesn't look like it's cleaning down. It's taking approx 1hr to clean down 200,000 events in our system. I'm wondering if it is working, but we're generating more than it can clean down in the specified time interval. What do you think? Thanks, John [quoted text, click to view] "Joe Webb" wrote: > Hi John - > > Thanks for posting the ADF snippet! > > Unless you live along the 0 degree line that more or less crosses > Greenwich, England, I don't think the vacuumer is running when you > think it should. > > SQLNS keeps time in UTC, formerly known as GMT. To get local time from > UTC time, you need to add or substract a number of hours (depending on > the time zone) to/from the UTC. For example, the Central Standard Time > has a -6 hour offset from UTC. So if you live in say Chicago, you're > actually running the vacuumer at 300pm (2100 - 600 = 1500). > > Could that be your issue? > > BTW - here's a link to a site that talks about UTC and the offsets. > http://www.dxing.com/utcgmt.htm > > HTH... > > -- > Joe Webb > SQL Server MVP > > > ~~~ > Get up to speed quickly with SQLNS > http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 > > I support PASS, the Professional Association for SQL Server. > ( www.sqlpass.org) > On Thu, 15 Sep 2005 06:00:08 -0700, "J" <J@discussions.microsoft.com> > wrote: > > >Hi Joe, > > > >Answers to your questions :- > > > >1. Yup, here's the section from the config file: > > > > <Vacuum> > > <RetentionAge>P1DT00H00M00S</RetentionAge> > > <VacuumSchedule> > > <Schedule> > > <StartTime>21:00:00</StartTime> > > <Duration>P0DT03H00M00S</Duration> > > </Schedule> > > </VacuumSchedule> > > </Vacuum> > > > >2. Yup :) > > > >3. v2.0 > > > >4. I'll have to let you know that one tomorrow... I just checked, and the > >State returned the time from todays manual run, and the Schedule said: > > > >2, 01/01/1900, 21:00:00, 10800 > > > >Thanks, > >John
John - Okay, so the schedule is the problem. Forgive the simple questions, but sometimes it's best to start there. You are specifying 900pm in UTC time right, not local times? And you're running NSControl update after modifying the ADF? What version are you running? v2.0 or 2005? After the vacuumer was supposed to have run, what results do you get from the following T-SQL statements? SELECT * FROM dbo.NSVacuumSchedule SELECT * FROM dbo.NSVacuumState -- Joe Webb SQL Server MVP ~~~ Get up to speed quickly with SQLNS http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 I support PASS, the Professional Association for SQL Server. ( www.sqlpass.org) On Thu, 15 Sep 2005 04:39:03 -0700, "J" <J@discussions.microsoft.com> [quoted text, click to view] wrote: >Hi Joe, > >I've just tried the manual clean up using NSVacuum, and that works fine, it >did 32 in 30 seconds. It's running at the moment, but has approx 900,000 >quantums to clean out. > >Thanks, >John > >"Joe Webb" wrote: > >> Hi John - >> >> What happens when you disable the scheduled vacuumer and run it >> manually? >> >> I'm not suggesting this is a production-level resolution. But I'd like >> to see whether the issue is in the vacuumer determining what to get >> rid of, or whether the vacuumer is not running at all. >> >> HTH... >> >> -- >> Joe Webb >> SQL Server MVP >> >> >> ~~~ >> Get up to speed quickly with SQLNS >> http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 >> >> I support PASS, the Professional Association for SQL Server. >> ( www.sqlpass.org) >> >> >> >> >> On Thu, 15 Sep 2005 08:48:38 +0100, "J" <J@somewhere.cam> wrote: >> >> >Hi, >> > >> >I'm having trouble with the NS<product>Events table not being cleaned down. >> >Initially I thought it was because the vacuum event was clashing with the >> >backup procedure on the server, so I moved the vacuum event to a time 5 >> >hours before hand. Unfortunatley it made no difference, and the number of >> >events in the table actually increased! >> > >> >Has anyone got any ideas as to what maybe happening? >> > >> >The retention age is set to 1 day, and the event will go off at 9pm for 3 >> >hrs every day. >> > >> >Many Thanks, >> >John >> >
Hi John - Thanks for posting the ADF snippet! Unless you live along the 0 degree line that more or less crosses Greenwich, England, I don't think the vacuumer is running when you think it should. SQLNS keeps time in UTC, formerly known as GMT. To get local time from UTC time, you need to add or substract a number of hours (depending on the time zone) to/from the UTC. For example, the Central Standard Time has a -6 hour offset from UTC. So if you live in say Chicago, you're actually running the vacuumer at 300pm (2100 - 600 = 1500). Could that be your issue? BTW - here's a link to a site that talks about UTC and the offsets. http://www.dxing.com/utcgmt.htm HTH... -- Joe Webb SQL Server MVP ~~~ Get up to speed quickly with SQLNS http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 I support PASS, the Professional Association for SQL Server. ( www.sqlpass.org) On Thu, 15 Sep 2005 06:00:08 -0700, "J" <J@discussions.microsoft.com> [quoted text, click to view] wrote: >Hi Joe, > >Answers to your questions :- > >1. Yup, here's the section from the config file: > > <Vacuum> > <RetentionAge>P1DT00H00M00S</RetentionAge> > <VacuumSchedule> > <Schedule> > <StartTime>21:00:00</StartTime> > <Duration>P0DT03H00M00S</Duration> > </Schedule> > </VacuumSchedule> > </Vacuum> > >2. Yup :) > >3. v2.0 > >4. I'll have to let you know that one tomorrow... I just checked, and the >State returned the time from todays manual run, and the Schedule said: > >2, 01/01/1900, 21:00:00, 10800 > >Thanks,
Hi, I'm having trouble with the NS<product>Events table not being cleaned down. Initially I thought it was because the vacuum event was clashing with the backup procedure on the server, so I moved the vacuum event to a time 5 hours before hand. Unfortunatley it made no difference, and the number of events in the table actually increased! Has anyone got any ideas as to what maybe happening? The retention age is set to 1 day, and the event will go off at 9pm for 3 hrs every day. Many Thanks, John
Well, your timezone certainly makes like a bit more simple from this point of view. :) Anyway, your theory may very well be correct. Tomorrow morning, check the result of this T-SQL statement to see how many events and notifications were removed during vacuuming. SELECT * FROM dbo.NSVacuumState BTW - You can also run the NSSnapshotApplications stored procedure from the instance database to essentially get the same info. If you not cleaning as much as you'd like, you can specify a couple of vacuuming periods - one before the backups and another after the backups. But, I wouldn't really expect the backups to interfere with vacuuming if they are running simultaneously - other than further taxing the system resources and making things run a bit slower. HTH... -- Joe Webb SQL Server MVP ~~~ Get up to speed quickly with SQLNS http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 I support PASS, the Professional Association for SQL Server. ( www.sqlpass.org) On Thu, 15 Sep 2005 07:07:07 -0700, "J" <J@discussions.microsoft.com> [quoted text, click to view] wrote: >Hi Joe, > >No problem, we're in Southampton, England. So we shouldn't need to adjust >the time :) > >I have a thoery as to why it doesn't look like it's cleaning down. It's >taking approx 1hr to clean down 200,000 events in our system. I'm wondering >if it is working, but we're generating more than it can clean down in the >specified time interval. > >What do you think? > >Thanks, >John > > >"Joe Webb" wrote: > >> Hi John - >> >> Thanks for posting the ADF snippet! >> >> Unless you live along the 0 degree line that more or less crosses >> Greenwich, England, I don't think the vacuumer is running when you >> think it should. >> >> SQLNS keeps time in UTC, formerly known as GMT. To get local time from >> UTC time, you need to add or substract a number of hours (depending on >> the time zone) to/from the UTC. For example, the Central Standard Time >> has a -6 hour offset from UTC. So if you live in say Chicago, you're >> actually running the vacuumer at 300pm (2100 - 600 = 1500). >> >> Could that be your issue? >> >> BTW - here's a link to a site that talks about UTC and the offsets. >> http://www.dxing.com/utcgmt.htm >> >> HTH... >> >> -- >> Joe Webb >> SQL Server MVP >> >> >> ~~~ >> Get up to speed quickly with SQLNS >> http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 >> >> I support PASS, the Professional Association for SQL Server. >> ( www.sqlpass.org) >> On Thu, 15 Sep 2005 06:00:08 -0700, "J" <J@discussions.microsoft.com> >> wrote: >> >> >Hi Joe, >> > >> >Answers to your questions :- >> > >> >1. Yup, here's the section from the config file: >> > >> > <Vacuum> >> > <RetentionAge>P1DT00H00M00S</RetentionAge> >> > <VacuumSchedule> >> > <Schedule> >> > <StartTime>21:00:00</StartTime> >> > <Duration>P0DT03H00M00S</Duration> >> > </Schedule> >> > </VacuumSchedule> >> > </Vacuum> >> > >> >2. Yup :) >> > >> >3. v2.0 >> > >> >4. I'll have to let you know that one tomorrow... I just checked, and the >> >State returned the time from todays manual run, and the Schedule said: >> > >> >2, 01/01/1900, 21:00:00, 10800 >> > >> >Thanks, >> >John
hi, Noticed you are in Southampton. We are just up the road from you in Portsmouth, so if you want to compare notes on NS applications then let me know. It is always useful to know more about what kind of apps are being created. Kate MBCS Alert Technologies http://www.alert-technologies.co.uk Get your notification services implementation going in minutes not weeks ......
Hi Joe, Sorry for the delay, have been off sick. I couldn't find the stored procedure you mentioned on our installation, so just ran ther select queries instead... Anyways, they did go off as normal and did remove quite a few recordas which is good. I just don't think they're removing enough so we'll scheduling a second clean up after the backup as you suggested. :) I found out why the backup was upsetting the clean up as well, it's turns out they stop all services fro running when it is performed! So that would of upset it quite nicely. Thanks for the help,
Hi Kate, I may well take you up on the offer sometime :) Thanks, John
John - [quoted text, click to view] >I couldn't find the stored procedure you mentioned on our installation, so >just ran ther select queries instead...
No problem. Both report basically the same information. BTW - The NSSnapshotApplications is in the Instance datatabase, not the Application database. [quoted text, click to view] >Anyways, they did go off as normal and did remove quite a few recordas which >is good. I just don't think they're removing enough so we'll scheduling a >second clean up after the backup as you suggested. :)
Ah...good! Thanks for posting the follow up! [quoted text, click to view] >I found out why the backup was upsetting the clean up as well, it's turns >out they stop all services fro running when it is performed! So that would >of upset it quite nicely.
That would definitely upset the apple cart. :) Cheers! -- Joe Webb SQL Server MVP ~~~ Get up to speed quickly with SQLNS http://www.amazon.com/exec/obidos/tg/detail/-/0972688811 I support PASS, the Professional Association for SQL Server. ( www.sqlpass.org) On Mon, 19 Sep 2005 02:23:02 -0700, "J" <J@discussions.microsoft.com> [quoted text, click to view] wrote: >Hi Joe, > >Sorry for the delay, have been off sick. > >I couldn't find the stored procedure you mentioned on our installation, so >just ran ther select queries instead... > >Anyways, they did go off as normal and did remove quite a few recordas which >is good. I just don't think they're removing enough so we'll scheduling a >second clean up after the backup as you suggested. :) > >I found out why the backup was upsetting the clean up as well, it's turns >out they stop all services fro running when it is performed! So that would >of upset it quite nicely. > >Thanks for the help,
Don't see what you're looking for? Try a search.
|