m60freeman · Senior Database Administrator · ✭✭✭✭✭

I could see this working much like the Registered Servers | Local Server Groups feature of SSMS (add and remove a group and add and remove a server from a group features). The Group, Server, and ServerGroup tables to implement this could be stored in the Repository. Being able to have global and per-user Groups would be…

My highest priorities out of this would be: a) Allow enough space to at least hold a full stored procedure name, which is 128 characters in SQL Server. A full database.schema.name would be 3 times that long (plus the periods), but 128 would be a great improvement. b) Allow special characters so I can use a name like…

Is this essentially a duplicate of

30 characters isn't even enough for us to fit most full stored procedure names, much less anything meaningful to tell us which statement in a stored proc the hash represents. I'd like to see at least 128 (SYSNAME length), but preferably enough for DatabaseName.SchemaName.ProcName plus a little extra room for a line number…

This has been implemented in 10.2 and this suggestion should therefore have its status changed.

I think that another way to approach this would be to lock in the custom range but be able to click left and right arrows to shift it by exactly one day in either direction. This would make it easy to compare the selected period with the same period in the previous days (one at a time). jadams@shopathome.com‌ would need to…

Please see related feature requests:

If it would be easier, even displaying the historical data as a line (like the baseline display for Resources) would be helpful.

Being able to find the statement holding a lock that is part of an idle blocker by clicking on the words "idle blocker" (which would need to be a link to show the previous SQL statements for the outermost current transaction for that session) would be a big help. In no case have I ever been able to hunt through the UI in…

See related suggestion:

Any news on the Columbus meeting? I haven't heard a word since I sent my RSVP. If I show up, will anyone from Solarwinds be there?

it is well known that the optimizer works in mysterious ways. The decisions about how the engine manages the procedure cache seems equally mysterious. 

>Do you think people would get the line scale and how it may relates to the bars? >If it's not obvious or intuitive, how could we make it so? On the one hand, Total Waits will always be the larger scale, so that helps. On the other hand, hover text (for each Y axis scale) that mentions whether that scale is for Total Waits…

siggy's line for Total Wait is brilliant! I like his option #1 as well, but that might be complex to implement. I'd prioritize #2 over #1 but would like them both.

I've been running with optimize for ad hoc since 3/3 and have had an hourly job to execute DBCC FREESYSTEMCACHE('SQL Plans') running since 3/5. Neither of these resulted in a change to the pattern of the Procedure Cache Hit Ratio chart. I can live with 75%, but I'm still curious as to the reason for the ratio dropping when…

Thanks, the links are working properly now. I wish they didn't keep breaking...

Do you mean UPDATE ignite.CONST_ID SET ST = NULL WHERE TRUNCATED = 'Y' AND PNAME IS NULL or DELETE ignite.CONST_ID WHERE TRUNCATED = 'Y' AND PNAME IS NULL?

Further investigation showed that the Grails application is making a SOAP call to a third-party in the middle of a transaction (which could take a few seconds to a minute or two to return), so you were right on the money. We're looking into how to change the application code to avoid that.

>Drill into tighter and tighter time frames to see top 15 specific for that timeframe The issue is that without the "other" showing up, I don't know where I need to drill into. >If we were to show the top 15 + "other" in many systems with something like an ad hoc workload, you'd lose the relative impact of the top 15…

I think the issue may be that the Drive Path needs to be set to the designation that starts with "naa", but I'm not sure. If I click on "Show Files", I see the files that I expect. However, I'd like the drive letter to be what is displayed on the Disk Devices tab within the Resources tab.

What product were you using?

When the Procedure Cache Hit Ratio is low, so is the SQL Compilation/sec. Therefore, I don't think that there is a situation where too many compilations are driving down the ratio.

Done: https://thwack.solarwinds.com/ideas/4864

mandevil, Exactly right. But we have lots of old code that uses web services and other external calls, possibly inside transactions, and rather than creating a project to find and fix all of them, it would be helpful to give the developers the information they need to find the one that is causing the specific problem I'm…

Interesting and possibly useful. I may play around with that. But it doesn't address the order in which the colors are used in the chart. My particular issue is that a number of adjacent column segments will be minor shade variations of the same color, making it difficult to visually match the column segment with the…

Done: I would encourage people to up-vote this if it is relevant to their interests.

No use of nolock or non-default transaction isolation levels. Mark Freeman Database Administrator | Rogue Fitness 1080 Steelwood Road, Columbus, OH 43212 MFreeman@RogueFitness.com On Fri, Jun 20, 2014 at 5:18 PM, mandevil <

Is this fixed only for new SQL not in the repository prior to upgrading or should it be working for all? I'm not seeing new results for older statements.

Excellent! I'll test it out.

I upvoted the feature request. That workaround might be barely acceptable if it actually worked, but there is too much likelihood that the query won't be listed there. In my case, I did find it, but when I stick a 0x on the front and plug it in to a query and run it on the correct instance, I still get nothing: SELECT…