Re: Question about 'by Unpin'
Hi Bret, Actually, this run of 14 minute sp_sysmon included a few other indexes before the example one (some of them were APL). The example one actually completed within this window. The Housekeeper...
View ArticleRe: Question about 'by Unpin'
Hi Anil, I don't really have an opinion on the housekeeper settings, but then I don't do that much tuning work. Having said that, I realized that I do have another observation on the sysmon output,...
View ArticleRe: Question about 'by Unpin'
Hi Bret, Great! One more thing to try after the current test that is running as I write this. I believe I will need to change the following parameters to enable use of worker processes: max parallel...
View ArticleRe: Question about 'by Unpin'
Hi Bret, I am done with the test. A result file is attached. It consists of: 1. sp_sysmon for the duration of the index creation2. Output of the SQL for the difference between before and after image of...
View ArticleRe: Question about 'by Unpin'
Hi Bret, I just ran the test again with the Housekeeper turned off. There is no difference in the 'by Unpin' data point. The log file for this run is attached. Any further ideas? Thanks,Anil
View ArticleRe: Question about 'by Unpin'
Hi Anil, I think my next step would be to break the activity down into component parts. Currently you rely on rebuilding the clustered index to rebuild the nonclustered indexes. I would break that...
View ArticleRe: Question about 'by Unpin'
Do you have Async Log Service enabled for your database by chance? post output of : sp_helpdb and sp_configure
View ArticleRe: Question about 'by Unpin'
Kevin, Are you referring to the 'allow sql server async i/o' parameter? If so, this is turned on by default. Also, that is a server level parameter. I am not sure I understand what you mean when you...
View ArticleRe: Question about 'by Unpin'
Bret/Kevin, I have some good news! One thing about our tempdb environment... it is always configured with mixed data and log. For this test host, I split it into 2 different devices; one device for log...
View ArticleRe: Question about 'by Unpin'
Hi Anil, The Asynch Log Service feature Kevin is asking about, your sp_helpdb doesn't show it turned on so we don't need to worry about it (though you might want to give it a try in the future)....
View ArticleRe: Question about 'by Unpin'
Bret/Kevin, I will look into ALS at a later time. I am repeating the test once again. I will keep you posted about the results. Meanwhile, I would appreciate your insight about my question in my...
View ArticleRe: Question about 'by Unpin'
The main issue is that you will hit 1105 errors more often/sooner with separate data and log segments than with combined. With them separate you hit 1105 as soon as segment fills, with them combined...
View ArticleClik here to view.
Re: Question about 'by Unpin'
Bret/Kevin, My apologies about the false alarm. The 'by Upin' behaviour still persists. So I am back to square oneBret,You wanted some details about the table. Here is the DDL for it.IF...
View ArticleRe: Question about 'by Unpin'
Bret, As per your suggestion, I ran the test with dropping both the indexes and then creating the clustered index first and then the non-clustered index. I captured separate sp_sysmon outputs for both...
View ArticleRe: Question about 'by Unpin'
Bret, Any updates or thoughts as to the direction to take from here on? Thanks,Anil
View ArticleRe: Question about 'by Unpin'
Hi Anil, I now believe the "by Unpin" flushes are due to interaction with the checkpoint process. The checkpoint process runs regularly and flushes all dirty buffers in cache to disk. Some of those...
View ArticleRe: Question about 'by Unpin'
Hi Bret, Thanks for the explanation. Our weekend maintenance process sets the 'trunc log on chkpt' for the database before it starts and turns it off after it is done. The database is then dumped later...
View ArticleRe: Question about 'by Unpin'
Hi Anil, When the "truncate log on checkpoint" option is on for a database, the checkpoint process ignores the "recovery interval in minutes" setting. Instead the checkpoint process will check once a...
View ArticleRe: 2k Unit - How to calculate
To avoid confusion it's easy to calculate size in (2K Pages) by multiplying the value in MB by 512Ex. 400 MB * 512 = 204800 (2K Pages)And vise versa: Size in 2K Pages / 512 = Size in MB
View ArticleRe: NEED to configure SAP Sybase Ase replication server
Hi Ibrahim, Regarding your second question about space increase per week, it's a good idea to keep records about space used weekly in a table and then, query this table to get the info. that you want....
View Article