I’m sometimes seeing mismatched timestamps in influx when using telegraf to poll port channel interfaces on two separate network switches. When I look at table view in Grafana I see that port one of the channel group answers at a certain time and a few seconds later the second port will also fill in it’s data. Most times the two ports answer the query at the same time. But the every once in a while time differences are throwing off the graphs.
Shown directly below is a table view of the port-channel separated by host names. The next picture is that of merged table and the sums are getting thrown off somehow. It’s returns a number much larger that the true sum of the parts. The merged table is the data is what I want my graph to show but it’s getting skewed by those random time differences. Is there a way to rectify this issue or ignore those large numbers?
