Tackling support issues is a lot like a doctor treating a patient – a list of symptoms is all that the doctor has to go on. Using these, a diagnosis is made to identify the cause(s) and then a remedy is suggested. This is an iterative process till the affliction (issue) is cured(solved). In most cases, good doctors can identify the cause in one or two iterations, However, sometimes we encounter some unique cases where a doctor/support executive is forced to treat the symptom rather than the cause since there is insufficient information to clearly identify the cause. Vishnu, from the ResellerClub’s able support team was faced with one such instance. One of our resellers, Peter Van, was unable to access any of the websites hosted on his VPS due to an issue with the Apache server. In his characteristic diligent style, Vishnu devised a solution that addressed the issue immediately and set up safeguards that could help counter a recurrence in the future.
Our Support Executive:
Vishnu has been with ResellerClub for about 2 years and has established himself as our go-to guy for out of the box solutions and impromptu scripting to resolve complex technical issues! No sooner is he given an issue to deal with than he is ready with a solution!
All websites hosted on the reseller’s VPS package were down due to an issue on the Apache web server.
Due to the nature of the VPS set up, logging had been disabled and hence all efforts to identify the cause of the issue were in vain.
Since none of the websites on the server was accessible, Vishnu checked the logs to find any information which could help him pin the cause for the downtime. However, the absence of logged data thwarted this approach.
Since the Apache service was not active, Vishnu restarted the service in an attempt to see if the websites were down purely due to this. It appeared that this was indeed the case and the websites were now accessible!
Beyond just resolving the issue at hand, Vishnu made sure that diagnostics and logs would be available, should a similar instance occur in the future. He installed systat- a monitoring tool, on the server to check on the resource usage history and also wrote a bash script to notify the reseller via email in the event of Apache being down.
The script would also give the reseller the time stamp of the event and would restart the service to prevent a long downtime. Once the service is up and running, the reseller would be notified to contact the support team who can check the logs and resource usage history to identify the cause and address the issue.
Peter was very happy with the solution and lauded Vishnu’s ingenious streak that would help save the day for him should his issue resurface.