Database observability: faster, more secure DevOps

June 9, 2023
Database observability: faster, more secure DevOps

The case for database observability is simple. You can’t fix what you can’t see.

For your org’s databases, that means you can’t improve reliability, security, efficiency, or productivity. You can’t rapidly find and chase down bugs or easily and quickly run root cause analyses. You’re left playing catch-up. 

You’re working with a database ecosystem that’s slowing down the development and innovation your company hinges on. And you’re worried about internal and external risks – the ones you know of and even more concerning, the ones you don’t. 

Database observability installs a window into your database operations so you can see clearly. This visibility and clarity become a superpower in the face of security risks. Even more, it becomes a central diagnostic tool that unearths actionable insights and informs intelligent decisions for your database DevOps

Bringing Database Observability to DevSecOps webinar

Read on for an overview of database observability and its game-changing impact on DevSecOps. For an in-depth expert explanation – including DB observability demos using dashboards on Elastic, Datadog, AWS Cloudwatch, and Splunk – head to the webinar replay

Observable = actionable

Database observability is the ability to measure a system’s current state based on the data it generates, such as logs, metrics, and traces.

Think of it like the warning lights and information center on your car’s dashboard. The speedometer and odometer get most of the attention as they measure your velocity and progress – analogous to your data and development itself. But the car’s systems – engine, fuel, brakes, and more – are also monitored to make sure they adequately support that velocity and progress. 

Database observability is like a vehicle dashboard

Database observability can function like a Check Engine Light, warning you of issues that may need maintenance, repair, or investigation. Or to take the analogy even further, observability is like your car alarm, alerting you of suspicious activity. All in all, these monitors give you visibility into the health and reliability of your vehicle. 

Database observability monitors and understands what’s happening in your database ecosystem. Plugged into your DevOps dashboard and reporting platforms, it allows you to pop the hood on your database change management and functionality to improve, maintain, diagnose, and supercharge with speed, agility, and confidence. No matter how far you’ve progressed in your database automation initiatives, observability can drive faster, more impactful strategies. 

Structured logging is the cornerstone of database observability

Structured logging is crucial to database observability

Logs are everything

If you’re not collecting database change logs or those logs aren’t accurate, you can’t have observability. 

To leverage log files for communicating information into observability tools and dashboards, they needed to be reorganized. Before updating, logs are difficult to parse.

structured logs - comparison

Liquibase’s approach, dubbed structured logging, groups together relevant pieces of information in JSON blocks. These can easily be sent to dashboards and subsequently parsed into actionable insights. 

After adopting structured logging, these JSON blocks include: 

  • Timestamps
  • Log levels
  • Classes
  • Count of executed threads
  • Command outcome messages
  • Target
  • Catalog name
  • Schema name
  • …and more

Across more than 50 supported database types, this structured logging provides data that can be sliced and diced for operational insights such as:

  • Deploy Frequency
  • Success/Failure
  • Lead Time
  • Long Running Jobs
  • Deployment Size
  • Drift Detection
  • Change Author
  • Root Cause Analysis

See the database DevOps observability dashboard in action (live demo) in the on-demand webinar.

New levels of security thanks to database observability

Security investigations are easier with database observability

If you can easily see what’s happening with your databases, you can monitor for outlier events and activity. Thanks to structured logging feeding observability data into your dashboard, you can detect things like:

  • Odd Hours
  • Malicious Drift
  • Unauthorized Users
  • Permission Changes
  • Post Mortem

A database security analyst might otherwise spend hours scouring logs for clues and trying to understand what happened, what’s at risk, and who’s to blame. But with database observability fueled by Liquibase, that analyst can waste no time searching – they’ll easily recognize the concerning event from the dashboard – and devote time to fixing and preventing security issues, instead. 

Observability allows database teams to streamline their reactive approach to security breaches and reallocate that time saved to optimizing a better, more secure database and DevOps process. 

Database observability is a catalyst for better investigations and higher quality

Capturing observability data about your databases gives you a powerful performance and security dashboard – but the specific use cases are possibly even more compelling. 

With Liquibase’s distinct approach to structured logging, you can introduce third-party data to mix-and-match information that answers tricky questions. Taking the example used in the webinar, query time, you can ask, “why has query time gone through the roof?” Perhaps website users are experiencing frustrating lags or slow-loads that disrupt their experiences. 

Reviewing Liquibase data, you can identify database changes that may have contributed to the query slowdown. Thanks to database observability, you can quickly find the problem and strategize a resolution – such as, in some cases, a Targeted Rollback to the pre-issue state. 

You can then deploy Quality Checks to make sure the improvements you’ve identified are applied to every subsequent database change. And just like that, observability elevates not only your ability to hunt down issues but your standards of quality regarding all future database changes. 

Feature sneak peek: custom fields and enriched log data

Custom fields and enriched log data for even more database observability

Adding data to your logs enriches the data with additional context that can give you even more granular and specific layers of observability. For instance, including “line of business” or “app name” allows you to zero in on teams, departments, or programs to make more valuable recommendations. 

Catch the webinar replay for an exclusive look at this upcoming feature, plus live demos of observability insights using popular dashboards.