Server-side Monitoring Tips to Ensure Optimal Web Performance
An overloaded Web server? Your users might get a "503 Service~ Unavailable" error
A slow database connection? Your users might get a "502 Bad Gateway " message
A saturated application server? Your users~ might get a "500 Internal~ Server Error"
Don't wait to realize that you don't have enough infrastructure capacity until is too late and Web transactions and requests cannot be completed.~~ Ensuring high-performing Websites and optimal users' experience starts with a healthy, reliable and top-performing infrastructure.
The following checklist will walk you through some key areas to consider to help you build comprehensive server-side monitoring plans.
1. Assess what elements to monitor. Your plans should be comprehensive and include ALL elements responsible to present Web content to your users, including:
- Physical datacenter systems: Oversee DNS, mail servers, Web servers, application servers, databases and any other server part of your Web content delivery chain. Besides monitoring critical server metrics such as CPU, Disk, Memory, Process, Services or Network Utilization, you should oversee connectivity and response time. For example, continuously test and validate that your DNS look-ups are working and your DNS server is resolving domain names correctly. Similarly monitor performance of Mailboxes, Hub Transport, Client Access, Unified Messaging and Edge roles in your Exchange deployment. Track complete mail server round trip time and oversee also database operations by caller (item, inventory, category, etc).
- Cloud instances and Virtualization resources : Monitor VMware vSphere hosts and virtual machines (VMs) performance to quickly identify ESX/ESXi Servers running short of resources, as well as performance for EC2 & RDS instances.
2. Configure thresholds and alerts. Define SLA agreement and take the time to identify who should be alerted and notification mechanism (SMS, email, push notifications, Twitter, etc...) , and the type of violation (warning or critical) that will trigger an alert. For example, some organizations like to get early-sign notifications (e.g. Disk space - 80% full), while others like to get alerted as it gets closer to a catastrophic event (Disk space - 98% full). In addition, consider who is responsible for each element of your infrastructure to determine notification policies (e.g. DBA for a database problem, Exchange Administrator for an email problem, etc) and make sure that alerts are routed to the right teams.
3. Automatically share key reports. Define custom views and reports (high-level dashboards views as well as very granular details) so you can easily troubleshoot the performance across all monitored areas. Automatically send out reports ( CSV, PDF, Excel or Email formats) to key stakeholders on a weekly, bi-weekly or monthly basis, and keep everybody informed and aligned.
4. Create a baseline and compare performance evolution over time. That way you can easily identify trends and troubleshoot deviations (e.g. over-strained servers) and take corrective actions much early (capacity planning, different resource allocation, etc)
5. Don't forget to build a comprehensive Web performance monitoring plan that includes server-side monitoring, app code monitoring and Web transaction monitoring to detect ANY type of problem that could hinder online user experiences.
When your underlying app infrastructure sneezes, your online users get a cold, so take action now. Sign-up for a free trial of Site24x7 Monitoring today!