Skip to main content

The No-BS Guide to Logging - Part 2

A vendor-neutral checklist to help you get your log strategy straight
Sven Dummer


Start with The No-BS Guide to Logging - Part 1

Coming off of the last post outlining the necessity for log management, the process of choosing logging software can seem daunting. The following are major elements of a good log strategy and can also serve as checklist items when you shop for a log management solution:

Collect, Aggregate, Retain

It's crucial to think about your data retention needs and the costs associated with storing them. How long do you need to keep the logs? Do you need them just for troubleshooting, or also for business intelligence type of analysis? Are there regulatory or audit requirements that require you to keep the logs for a certain period of time?

Your daily log volume might already be large, but keep in mind that it doesn't take much to multiply the volume temporarily. For example, a component failure and the resulting log messages in a complex system could easily quadruple the amount of log messages. An external event could have the same effect: if you run an online store, Black Friday might balloon your sales as well as your log volumes. If your log aggregation doesn't scale, you could lose your main troubleshooting foundation when you need it most.

Handle Log Diversity

Log files come in a variety of formats, some following standards and conventions, others completely custom. Your log solution should be able to parse and present the data in a comprehensive form in near real-time, and it should allow to define custom parsing rules. A desirable feature is the ability to add metadata.

Reveal What Matters

Just having a search tool is not enough. To make sense of your log data and the correlation between different data points, you need real-time indexing and parsing, grouping, along with powerful analytics, customizable dashboards, and data visualization. Your log analytics solution should provide a treasure map to the contents of your logs, not just a metal detector that you must use to scan indiscriminately.

Detect Anomalies

Given the volume and complexity of log data, you can't rely on searching for problems. Things you never anticipated happening are typically the type of problems that hurt the most. A good log analytics solution should be able to learn what is “normal” in your log data, and automatically identify and highlight any deviations from norms.

Make Your Own Apps Log

If you write your own code, your log management solution must be able to parse and analyze it. Consider using a well-established data format like JSON (our recommendation) or XML. Whatever you choose, make sure it's plain text format (not binary), that it is human-readable, and easy to parse. Your log solution should be able to easily receive the logs from your application and allow you to set up custom parsing rules if needed.

Be Alert(ed)

Just like every good monitoring application, every good log management solution should allow to send you and your teams alerts based on defined events, like error messages. It should be possible to send these alerts through common third party collaboration tools.

Don't Break the Bank

Cloud technologies made running distributed systems and elastic compute farms affordable for SMBs. The bill for the troubleshooting tools should be affordable, too. There are fully cloud-based SaaS solutions out there, as well as on-premise products and hybrids, which typically come at higher costs (including those for hardware and datacenter footprint).

Key criteria to decide if SaaS or on-premise solutions are right for you are the sensitivity and volume of your data. Security or privacy concerns or regulatory requirements may keep you from transferring data across public networks. Similarly, the sheer data volume could make this impossible or too expensive.

Sven Dummer is Senior Director of Product Marketing at Loggly.

Hot Topics

The Latest

64% of enterprise networking teams use internally developed software or scripts for network automation, but 61% of those teams spend six or more hours per week debugging and maintaining them, according to From Scripts to Platforms: Why Homegrown Tools Dominate Network Automation and How Vendors Can Help, my latest EMA report ...

Cloud computing has transformed how we build and scale software, but it has also quietly introduced one of the most persistent challenges in modern IT: cost visibility and control ... So why, after more than a decade of cloud adoption, are cloud costs still spiraling out of control? The answer lies not in tooling but in culture ...

CEOs are committed to advancing AI solutions across their organization even as they face challenges from accelerating technology adoption, according to the IBM CEO Study. The survey revealed that executive respondents expect the growth rate of AI investments to more than double in the next two years, and 61% confirm they are actively adopting AI agents today and preparing to implement them at scale ...

Image
IBM

 

A major architectural shift is underway across enterprise networks, according to a new global study from Cisco. As AI assistants, agents, and data-driven workloads reshape how work gets done, they're creating faster, more dynamic, more latency-sensitive, and more complex network traffic. Combined with the ubiquity of connected devices, 24/7 uptime demands, and intensifying security threats, these shifts are driving infrastructure to adapt and evolve ...

Image
Cisco

The development of banking apps was supposed to provide users with convenience, control and piece of mind. However, for thousands of Halifax customers recently, a major mobile outage caused the exact opposite, leaving customers unable to check balances, or pay bills, sparking widespread frustration. This wasn't an isolated incident ... So why are these failures still happening? ...

Cyber threats are growing more sophisticated every day, and at their forefront are zero-day vulnerabilities. These elusive security gaps are exploited before a fix becomes available, making them among the most dangerous threats in today's digital landscape ... This guide will explore what these vulnerabilities are, how they work, why they pose such a significant threat, and how modern organizations can stay protected ...

The prevention of data center outages continues to be a strategic priority for data center owners and operators. Infrastructure equipment has improved, but the complexity of modern architectures and evolving external threats presents new risks that operators must actively manage, according to the Data Center Outage Analysis 2025 from Uptime Institute ...

As observability engineers, we navigate a sea of telemetry daily. We instrument our applications, configure collectors, and build dashboards, all in pursuit of understanding our complex distributed systems. Yet, amidst this flood of data, a critical question often remains unspoken, or at best, answered by gut feeling: "Is our telemetry actually good?" ... We're inviting you to participate in shaping a foundational element for better observability: the Instrumentation Score ...

We're inching ever closer toward a long-held goal: technology infrastructure that is so automated that it can protect itself. But as IT leaders aggressively employ automation across our enterprises, we need to continuously reassess what AI is ready to manage autonomously and what can not yet be trusted to algorithms ...

Much like a traditional factory turns raw materials into finished products, the AI factory turns vast datasets into actionable business outcomes through advanced models, inferences, and automation. From the earliest data inputs to the final token output, this process must be reliable, repeatable, and scalable. That requires industrializing the way AI is developed, deployed, and managed ...

The No-BS Guide to Logging - Part 2

A vendor-neutral checklist to help you get your log strategy straight
Sven Dummer


Start with The No-BS Guide to Logging - Part 1

Coming off of the last post outlining the necessity for log management, the process of choosing logging software can seem daunting. The following are major elements of a good log strategy and can also serve as checklist items when you shop for a log management solution:

Collect, Aggregate, Retain

It's crucial to think about your data retention needs and the costs associated with storing them. How long do you need to keep the logs? Do you need them just for troubleshooting, or also for business intelligence type of analysis? Are there regulatory or audit requirements that require you to keep the logs for a certain period of time?

Your daily log volume might already be large, but keep in mind that it doesn't take much to multiply the volume temporarily. For example, a component failure and the resulting log messages in a complex system could easily quadruple the amount of log messages. An external event could have the same effect: if you run an online store, Black Friday might balloon your sales as well as your log volumes. If your log aggregation doesn't scale, you could lose your main troubleshooting foundation when you need it most.

Handle Log Diversity

Log files come in a variety of formats, some following standards and conventions, others completely custom. Your log solution should be able to parse and present the data in a comprehensive form in near real-time, and it should allow to define custom parsing rules. A desirable feature is the ability to add metadata.

Reveal What Matters

Just having a search tool is not enough. To make sense of your log data and the correlation between different data points, you need real-time indexing and parsing, grouping, along with powerful analytics, customizable dashboards, and data visualization. Your log analytics solution should provide a treasure map to the contents of your logs, not just a metal detector that you must use to scan indiscriminately.

Detect Anomalies

Given the volume and complexity of log data, you can't rely on searching for problems. Things you never anticipated happening are typically the type of problems that hurt the most. A good log analytics solution should be able to learn what is “normal” in your log data, and automatically identify and highlight any deviations from norms.

Make Your Own Apps Log

If you write your own code, your log management solution must be able to parse and analyze it. Consider using a well-established data format like JSON (our recommendation) or XML. Whatever you choose, make sure it's plain text format (not binary), that it is human-readable, and easy to parse. Your log solution should be able to easily receive the logs from your application and allow you to set up custom parsing rules if needed.

Be Alert(ed)

Just like every good monitoring application, every good log management solution should allow to send you and your teams alerts based on defined events, like error messages. It should be possible to send these alerts through common third party collaboration tools.

Don't Break the Bank

Cloud technologies made running distributed systems and elastic compute farms affordable for SMBs. The bill for the troubleshooting tools should be affordable, too. There are fully cloud-based SaaS solutions out there, as well as on-premise products and hybrids, which typically come at higher costs (including those for hardware and datacenter footprint).

Key criteria to decide if SaaS or on-premise solutions are right for you are the sensitivity and volume of your data. Security or privacy concerns or regulatory requirements may keep you from transferring data across public networks. Similarly, the sheer data volume could make this impossible or too expensive.

Sven Dummer is Senior Director of Product Marketing at Loggly.

Hot Topics

The Latest

64% of enterprise networking teams use internally developed software or scripts for network automation, but 61% of those teams spend six or more hours per week debugging and maintaining them, according to From Scripts to Platforms: Why Homegrown Tools Dominate Network Automation and How Vendors Can Help, my latest EMA report ...

Cloud computing has transformed how we build and scale software, but it has also quietly introduced one of the most persistent challenges in modern IT: cost visibility and control ... So why, after more than a decade of cloud adoption, are cloud costs still spiraling out of control? The answer lies not in tooling but in culture ...

CEOs are committed to advancing AI solutions across their organization even as they face challenges from accelerating technology adoption, according to the IBM CEO Study. The survey revealed that executive respondents expect the growth rate of AI investments to more than double in the next two years, and 61% confirm they are actively adopting AI agents today and preparing to implement them at scale ...

Image
IBM

 

A major architectural shift is underway across enterprise networks, according to a new global study from Cisco. As AI assistants, agents, and data-driven workloads reshape how work gets done, they're creating faster, more dynamic, more latency-sensitive, and more complex network traffic. Combined with the ubiquity of connected devices, 24/7 uptime demands, and intensifying security threats, these shifts are driving infrastructure to adapt and evolve ...

Image
Cisco

The development of banking apps was supposed to provide users with convenience, control and piece of mind. However, for thousands of Halifax customers recently, a major mobile outage caused the exact opposite, leaving customers unable to check balances, or pay bills, sparking widespread frustration. This wasn't an isolated incident ... So why are these failures still happening? ...

Cyber threats are growing more sophisticated every day, and at their forefront are zero-day vulnerabilities. These elusive security gaps are exploited before a fix becomes available, making them among the most dangerous threats in today's digital landscape ... This guide will explore what these vulnerabilities are, how they work, why they pose such a significant threat, and how modern organizations can stay protected ...

The prevention of data center outages continues to be a strategic priority for data center owners and operators. Infrastructure equipment has improved, but the complexity of modern architectures and evolving external threats presents new risks that operators must actively manage, according to the Data Center Outage Analysis 2025 from Uptime Institute ...

As observability engineers, we navigate a sea of telemetry daily. We instrument our applications, configure collectors, and build dashboards, all in pursuit of understanding our complex distributed systems. Yet, amidst this flood of data, a critical question often remains unspoken, or at best, answered by gut feeling: "Is our telemetry actually good?" ... We're inviting you to participate in shaping a foundational element for better observability: the Instrumentation Score ...

We're inching ever closer toward a long-held goal: technology infrastructure that is so automated that it can protect itself. But as IT leaders aggressively employ automation across our enterprises, we need to continuously reassess what AI is ready to manage autonomously and what can not yet be trusted to algorithms ...

Much like a traditional factory turns raw materials into finished products, the AI factory turns vast datasets into actionable business outcomes through advanced models, inferences, and automation. From the earliest data inputs to the final token output, this process must be reliable, repeatable, and scalable. That requires industrializing the way AI is developed, deployed, and managed ...