So i am unsure how to troubleshoot this issue. or even what the application connections are.
thanks in advance for your help.
What's between the two machines? Is it just a simple layer 2 connection or does traffic cross falf the internet to get between machine A and B?
Can you monitor the application on the other side of the connection? That could help narrow things down.
Without more information, it's hard to say for certain. I have seen similar behavior in the past occur as a result of antivirus software on the machine. Specifically, a packet filter driver that was installed as part of an endpoint protection software, of which antivirus was only one component. I'm not saying that's what's going on here, but it's worth taking a look. If you find something that has an IDS/IPS or firewall running on either side, temporarily uninstall it and see if the condition improves. If it does not, then don't forget to reinstall it back.
So each is a virtual machine on a different box, there are 10Gig links between each box and connected to a Nexus switch. do you think it will be the physical connections. I thought that maybe there is a metric that is dropping traffic if it goes over a counter?
Ill have look at the interfaces and see if there is large amounts incoming traffic that is getting dropped.
thanks
I would suggest looking at each of the interfaces in between the two servers, looking for interface errors and discards. Are these two machines on the same subnet, or does their traffic pass through a router to get to each other?
Another possibility for what might cause this is resource contention. Given their 10Gb interfaces, it's probably not network congestion. Since this appears to be happening high up the OSI model, this same behavior can also be the result of high CPU or memory pressure on the machine. In extreme conditions, It could even be caused by bottlenecked disk I/O. Is the machine performing well aside from what ADM reports? Is either machine laggy when you RDP into them and launch applications?
The device has some of the logical processors of the CPU running at 100. could this cause latency and packet loss?
Could someone tell me specifically what this is measuring? Is it just TCP traffic being sent from the node to the application?
This article refers to an 80% packet loss, but may still have some helpful suggestions.https://support.solarwinds.com/SuccessCenter/s/article/Incorrect-80-packet-loss-statistics-returned-by-Connection-Quality-polling-in-SAM?language=en_US
