How to use Fastly + Logentries for insight into log data
Content Marketing Manager, Fastly
Can you give a brief overview of your background and role at Logentries?
I’m the Director of Product Marketing at Logentries, the leading log management and analytics company with over 40,000 users worldwide.
What makes Logentries unique/different as a log provider?
First and foremost, we’ve built Logentries to be elastic — i.e. scalable to ingest and interpret any volume of log data, in any format, from any environment. Logentries is also a real-time technology, preprocessing log data to provide features like “Live Tail,” which enables users to watch their aggregated log events stream as events occur. Finally, Logentries’ query language (LEQL) is easy to learn but powerful enough to analyze large datasets.
You announced the Fastly Community Pack on Logentries last year — why did you build it, and how does it work?
Logentries Community Packs are sets of pre-built queries, tags, alerts and dashboards configured for a specific environment. Built by both Logentries engineers and members of the Logentries community, these packs make it easy for any user to get immediate insights from their log data with zero setup. Logentries’ Fastly Community Pack offers out-of-the-box log analytics to Fastly customers who want to better understand what’s happening within their CDN.
Is there a real-world (or generalized) example you can give of how the Fastly Community Pack works?
Simply download the free Fastly Community Pack and upload it to your Logentries account. Your account will immediately be populated with:
Saved queries that reveal stats like page hits over time and requests per point of presence.
Real-time alerts on events like large file downloads, inactivity, and 400 errors. Alerts can be triggered off specified patterns in your logs, inactivity across your environment or specific changes in system or user behavior. Alerts can be sent to email, popular tools like Slack, and HipChat, or to a webhook.
Top URLs by response time, geography and more.
Custom dashboards visualizing information like data center distribution and requests by country.
Would different verticals/industries use Logentries differently?
Absolutely. Ecommerce customers, for example, rely heavily on real-time analysis, so they’ll spend most of their time using features like Live Tail to watch their logs stream and catch errors in real time, and Inactivity Alerts to be immediately notified of the absence of certain events. These real-time features are crucial as even seconds of downtime can often result in immediate and significant revenue loss. Other companies use Logentries to maintain PCI compliance, relying on anomaly detection, log aggregation, and data retention. Customers who produce mobile applications depend on our application performance libraries to monitor their apps and understand end-user behavior from our client libraries. Finally, we’ve been seeing more SaaS companies using Logentries for technical support teams who want to live-troubleshoot customer issues and product teams using logs to better understand how customers are using their product.
You recently announced SORT to the Logentries Query Language (LEQL) — can you describe what this is? Why has it been so compelling to your customers, and what are some of the use cases you’re seeing?
Yes — SORT enables customers to specify the order in which they view their query results. Just as you would use SORT in a SQL query, LEQL enables users to specify whether they’d like to review query results in ascending or descending order. When using the SORT function for logging from a CDN like Fastly, users can calculate value sets such as the top 10 URLs being requested, the most infrequently used features, or the most popular Fastly points of presence (POPs). This data can be grouped by country, device type, status code, IP address, and region.
What are the business benefits of using Logentries and Fastly?
Monitoring and analyzing your Fastly logs with Logentries can reveal insight into what content is most valued by site visitors, where your customers are located, which assets take the longest for customers to access and more. These insights can help you dramatically improve your customers’ user experience. Fastly logs are particularly great, given the amount of data you can log and how deep you can go with pulling insights out of your data.
Any best practices for using Logentries and Fastly together?
I would say the first thing you should do is set up logging from your Fastly account to start streaming logs. Fastly doesn’t store logs, so you’ll want to create a free Logentries account and follow our docs for sending logs to Logentries for search, analysis, and storage. Next, download and install the Logentries Fastly Community Pack to instantly populate your account with useful analytics you can use out-of-the-box. Finally, I would recommend configuring your alerts to be sent to the appropriate team members via email, webhook, or chat tool like Slack or HipChat.
