Support and Maintenance en New York
How we work in New York
“It loads fine for us” does not close a report
It is the most repeated answer and it is usually true: from here it really is fine. It is also useless, because whoever complained was not here. Before dismissing anything we look for the record from that window and that origin; if there is no record, the problem is that we were not measuring.
Automated checks from several points around the world
We schedule requests from different locations so that a degradation shows up without waiting for someone to report it. With its limit stated up front: an automated check uses a clean machine and a good line, so it warns early but does not replace people.
The alert fires on the high percentile, not on the average
An average absorbs one in twenty people having a terrible time, and that person exists. We set the alerts on the high percentiles, which is where the problem nobody reports lives, because they leave before writing. The average is for the report, not for monitoring.
A release can add a round trip without changing anything visible
A new tag, a redirect rule or an extra call to an outside service do not show on screen and do show at a distance. After every change we compare against the previous version the number of requests, the domains contacted and the hops: that is where these additions surface.
Answering your questions
What people ask us about Support and Maintenance in New York.
If your question is not here, tell us about the project and we will answer with real context.
Ask us directlyReal work
Projects we have built
Full service · All the detail
Support and Maintenance: stack, process, use cases and FAQs