SQL Monitor currently allows only one time setting for the maintenance window that can be applied to the chosen days of the week.
I would like to set day by day i.e. Monday from 11pm to 4am etc.

When I access the user interface (hosted on IIS) it is quite slow to use.
It frequently takes several seconds between clicking on something and things happening.

It does not appear to be a resource problem, as both the IIS/Monitor server and the database server has plenty of resources to spare.

Create an alert to monitor instance global configuration settings.

i.e. the below retrieves the global configuration settings.

EXEC sp_configure 'show advanced option', '1'
RECONFIGURE
EXEC sp_configure

A scheduled check against default results could discover any unauthorized changes such as 'allow updates', 'awe enabled', 'blocked process threshold', 'c2 audit mode'. etc

Create an alert to monitor server SERVERPROPERTY results.
i.e. select SERVERPROPERTY('Edition') retrieves instance edition.

A scheduled check against default results could discover any unauthorized changes to the server instance such as IsClustered, IsFullTextInstalled, IsIntegratedSecurityOnly, IsSingleUser, IsSyncWithBackupetc, etc.

Create an alert to monitor Database properties.
i.e. select * from sys.databases retrieves the settings for all the databases.

A scheduled check against default results could discover any unauthorized changes to the databases such as compatibility_level, UserAccess, Collation, IsTornPageDetectionEnabled, SQLSortOrder, etc.

It would be very helpful if I could save 'Analysis' graph after I select multiple counters. So the next time I could just click the Name and bring back the multi-counter analysis. In fact redgate can have some pre-canned favorites for Memory/IO counters etc. and others can contribute to different combinations of counters.

It seems that this is not possible with the custom metric solution.

To get a instant view of the unread alerts and their context/importance it might be a good idea to create a tickerbox at the leftmenu of the global overview page.

Currently to know what kind of alert has occured, I have to zoom in/navigate everytime to the alert page.

Layout example: Grouped by server, level, category, datetime

As addition it might be an option to create a tickerbox hide option .

UI analysis hangs/slow(10 seconds) when metricselection is made. Version 3.2, 3.3, browser IE 8, builtin webbrowser.
The slow responsetime makes the use of this interface less attractive.

I would like to suggest "Data Usage" metrics as part of the Analysis section.

Currently, we do have "Data Size" even "Log space usage" but not "Data Usage". Ability to see how much of the allocated space for a database (mdf file) is being used or how is growing over the time , allo a DBA to take corrective actions in terms of storage and SAN planning.

I am currently taking this metric on my own, using Reporting Services and do not understand why SQL Monitor can replicate same functionality.

To determine performance problem b it is needed to monitor several counter values at once. For example when counter x and counter y and counter z are all below 10 then you have a memory problem. I would be nice to contruct that kind of logic to get true performance alerts.

Have the ability to add more columns. specifically path, this way on servers with multiple instances we can see which sqlservr process/instance is using the CPU/Memory quickly.

I would like to see which alerts are firing more frequent over time.
It would be nice to be able to see alert statistics over time, tabluar or graphical.

We would like to collect the "dbcc showcontig" table scan density percentage so that we can alert on levels below a certain percentage.

We would like the ability to quickly turn off and on all monitoring and alerting globally. This would be usefull when we have major outage that we become aware of and start to work on, however alerts get generated very quickly and can overwhelm the email system or support phone. It would be great if we could continue to collect performance metrics but stop all alerts until the problem is resolved.

We would the ability to change/modify the alert description information so that emails sent to a phone can be more descriptiive to our error