As an update, there is another thread in the forums where someone has around 30 agents and some random problems were coming up.
After months of testing, we were finally able to isolate the problem and we thought it was fixed.
Additional logs provided to us using the latest version shows that the problem is still there in some form or another.
The agent uses a multithreaded method and it seems that in some cases, one of the threads can spiral out of control causing the agent to miss sending some data now and then.
What's probably happened from looking at your reports is that the agent missed sending pings because the rest of the data looks consistent while the missing pings have no associated missing communications.
In other words, when those pings were missed, the gaps, the agent was still communicating which tells us it failed to send the pings in those times.
When you see gaps, you can always look at your outages to see if there was an outage at the same time. If so, you can discount the missing ping/heartbeat as it makes sense since the agent was not able to communicate. Otherwise, you can attribute the gaps to the problem we are now working on and will release an update once tested.
The biggest challenge with offering online services is that when folks come across problems, they usually prefer to blame the service and tell others how bad it is.
The fact is, if people took the time to let the service know about the problems, they will always do everything possible to fix any problems to keep hard-earned members.
Thank you for pointing this out.