Realizing a potential disaster for their online business, the IT department sought a third-party external monitoring service provider to monitor their e-commerce site 24/7 and identify the issues and dependencies that were causing website outages. After one-minute monitoring each day for 7 days, the following issues were identified:

1. Slow Website Response

Dotcom-Monitor found that, although the e commerce site was hosted on a dedicated shared server, during certain times in a day, other applications on the server consumed most of its resources. This led to slow site response. This happened only during certain hours, as the conflicting application was run only a few times a day.

2. Slow Web Page Response

department also discovered that several pages within a transaction, such as shipping history and invoice printout were extremely slow. They took a long time to load, causing shoppers to abandon the site and instead call customer service. Dues to this discrepancy within the network, Dotcom-Monitor was able to identify the specific online pages that were slow and traced the problem to the database table that those pages were accessing.  

3. Problem in Accessing the Website

During certain intervals in a day, the retailer’s site became inaccessible altogether. Dotcom-Monitor tested the retailer’s DNS servers by querying specified servers performing Every minute, website monitoring showed that one DNS server was occasionally unable to resolve the address. This implied that the DNS server supporting the organization was not working consistently. The retailer upgraded the DNS server and the problem was resolved.

4. Varying Website Response from Different Geographical Locations

Persistent monitoring showed that this retailer’s shopping cart website responded at varying speeds from different locations. Some customers complained that the site took too long to open, while others had no issues. After analyzing the data, it was found that the ISP hosting the site used several providers that did not peer properly with certain networks. This network peering problem was identified and brought to the ISP. They took the necessary actions to resolve the problem by altering their BGP configuration.

5. Website Downtime required frequent monitoring

The retailer’s website occasionally suffered downtime, which led to customer dissatisfaction. The outages were mostly related to software glitches that the customer was unable to fix in a short period of time. These glitches usually required manual intervention. Enter Dotcom-Monitor. The goal to reduce downtime was to alert key personnel of any issue immediately. Once website monitoring detected any problems with the web server, immediate alert by phone and SMS would be sent to on-call personnel. This decreased downtime periods from an average of 25 minutes to less than 5 minutes.

The author has worked on traffic signals and landscape Auckland and provides SEO services for more than 5 years.

Tags: , , , , ,