Why Switch to SQL Sentry?
Mark Lessard | TrustRadius Reviewer
Updated April 30, 2020

Why Switch to SQL Sentry?

Score 10 out of 10
Vetted Review
Verified User
Review Source

Overall Satisfaction with SQL Sentry

SQL Sentry is being used to monitor 260 SQL instances across or enterprise. We are in the process of integrating the alerting capabilities with SCOM via SNMP. Along with alerting SS greatly increase our efficiency in resolving ongoing and past performance issues. SS also provides a central repository for reporting at the enterprise and client level. We previously used a competitors tool for 7 years but moved to SQL Sentry due to the products inability to function in our environment and more so a lack of response to the issues we were experiencing by the vendor. Our experience with SQL Sentry support has been excellent to date.
  • Collecting and rolling up SQL statements with the associated plan. The ability to see aggregate and individual execution of SQL statements along with the associated plan is key in rapid diagnosis of bad plans. Bad plans are the most common performance issue we see in the big data environment.
  • The ability to drill into disk IO. Being able to easily see IO per a file can help identify issues with slow / stressed storage and or poorly distributed tables.
  • Easily focus on time-frame. Being able to click and drag to select a time frame to analyze from a prior time frame keeps maintains the flow of troubleshooting.
  • Alerting is inherited and highly configurable.
  • The collection can be distributed over multiple servers allowing scaling out horizontally and fault tolerance.
  • AD roles can be used in configuring access to the subsets of the servers being monitored. Access can be configured such that adding a user to an Active Directory role will provide access to just the servers then need to see. This works very well for our client specific operations.
  • The presentation of SQL server waits needs work. A large value on a throw away wait like cxpacket will change the scale such that an important wait like sos_schedular _yield is not readable. I have been told this is being addressed in the next release.
  • Installation of the client requires sysadmin access on the repository server for the initial install. Once the install is complete sysadmin can be removed.
  • There is a web interface but this does not have much of the functionality yet.
  • I have not found a way to view page splits over time.
  • We are more efficient in diagnosing performance issues. Time spent on this type or diagnosis has decreased by about 75%. The tool makes it much easier to discriminate between issues due to SQL and environment issues versus solution architecture and data distribution issues.
A common experience is the SQL optimizer choosing to use a loop join on a large join. We see this even after stats update with full-scan on the associated tables. We have tried the optimizer trace flag.

Find the SQL statement in Top SQL
Look at the historical run times/plans grid. Notice run time and color of dot. High run-time and different color, you have a bad plan.
Open the plan to review operators
Add hash join query option or join hints. I know this is not best practices according to MS, but the trade-offs in our OLAP environment make it the best choice.

  • Diagnostic Manager
Diagnostic Manager has a better interface to view waits. This is an issue for me. SQL Sentry does a better job with showing SQL statments, rolling up the metrics, providing the code and is able to show the SQL plans. The alerting in SQL Sentry is easier to use. We are having issues with integrating SQL Sentry with SCOM via SNMP, SQL Sentry is working on this.

The big difference in these products is support. Both of these products have bugs and in some cases stop functioning. You can rollback to prior versions but all of this is time spent on the tool as opposed to doing the job. The support provided by SQL Sentry is magnitudes better than the support we got from Diagnostic Manager.
I only see one scenario for this type of tool. I want all tools in this category to collect OS and SQL performance metrics over a time frame and present and or alert on them in a way that allows me to diagnose availability and performance issues on the servers. I need this information in real time and historically. I need to be able to report on this information at the micro and macro level. I need a tool that just works and a company that stands behind there tool. SQL Sentry meets these needs.

The only scenario I see were SQL Sentry would not work is were purchase cost of the product is weighted very heavily as opposed to all other factors.

SQL Sentry Feature Ratings

Performance dashboard
Intelligent alerting
Historical trend data
Virtualization support

Using SQL Sentry

20 - The tool is used directly to analyze performance and alert on on issues by the database administrators. The alerts generated by the tool are used by support personel to route issues appropriately. There is a small number for developers and operations personal that use the tool to evaluate issues with campaign runs.
4 - Our SQL Sentry installation is realitively complex. We monitor over 250 instances using multiple monitors hosted on multiple hosts providing access into various firewalled networks. In general SQL Sentry requires little support on a day to day bases other than backups of the repositroy. Upgrades are a different story and require a person with database administration skills, basic networking skills and the abiltiy to work with services on a windows operating system. We spend about 8 manhours on an upgrade. That said Sentry One's latest release has an automated upgrade feature, I hope it works as designed.

If you want to make use of the custom alerting capabilities then be prepared for a steep learning curve.
  • Performance tuning
  • Alerting for performance and availability
  • Responding to runtime issues either real time of forensicly
  • Direct integration with our help desk application
  • Identification of connections to a specified server to assist in firewall configuration
  • Identification of unused logins
  • Make better use of alerting and metrics to identify performance issues.
  • Greater use by operations teams to provide visibilty into production run status
  • Far less time is spent on resolving performance issues and the data to perform forensic analysis.
  • the ability to address SLA violations with data.
The primary reason I rate SQL Sentry at a ten is there support. I have worked with support from Idera, IBM, AWS , Google and Microsoft. SQL Sentry support is by far the best in all regards. This includes response times, timely resoution of issues, follow up on going isssues and willingness to escalate an issue as needed.

SQL Sentry Support

They usually respond within a few hours and will pull in needed resources without the need for me to escalate the issue. They take the initiative to follow up. They are open to suggestions for future enhancements and actually implement them. They give the impression they want to resolve your issues.
Quick Resolution
Good followup
Knowledgeable team
Problems get solved
Kept well informed
No escalation required
Immediate help available
Support understands my problem
Support cares about my success
Quick Initial Response
It is the overall support I am impressed with. I expect the vendor to work with us and keep us informed until the issue is resolved. I want a sense that they are committed to our success. Out of the vendors we use IBM, Microsoft (Premier support), AWS, Hortonworks and SQL Sentry, SQL Sentry is the only one that has met my expectations. Based on my experience with the other vendors the majority of my support experiences have been exceptional.