my client has an AWS VPC with several Windows EC2 app servers that talk to each other and one outside server. We are seeing issues with what appear to be DNS resolution errors timing out the communications periodically. The issue has been happening for several months but seems to be getting worse. This is a set of applications that my company has taken over from a former programmer who put together this architecture. An expert we hired identified spikes in 4xx errrors at specific times. From a few to server thousand for a VERY short period. He is a Linux admin though and we are windows servers. We are looking for someone who can look in the Windows logs and see if there is a correlation that we can find. We are not experts at AWS, we get around but that's about it. We need someone to review the site with a particular focus on what is the root cause of these errors and how to fix them. Here are some notes from the primary programmer on this project:
1. have these failures just started happening or have they been on going and not recognized?
answer: I've seen some of them for several months now. It does appear to be getting worse, however.
2. you think they might be related to timeouts in the route53? Is that right?
answer: Yes, it's like the dns resolution wigs out every once in a while for several servers (client, script). Things just time out, and then a few minutes later, everythings back to normal
(since this was written it seems less likely this is route53)
3. am I understanding correctly that the evidence is timeout in finding one of the app sites? Is it always the same site?
answer: Yes, but it seems to be 2, 'client' and 'script' (which are the most used servers) that have the issue the most (there are 6 other servers). There error as seen by the user is not available. 505 time out
About the recuiterMember since Jul 18, 2017 Marko N.
from Utah, United States