Lag time introduced when switching servers. This is a really "interesting" issue

We recently switched data base server that was being monitored by Grafana and telegraf on a windows 2016 box. After the switch the four Grafana boards we use to monitor the server, three of them dropped an hour behind, and then will lag up to 20 minutes before refreshing the boards. But will keep up with the data for about 40 minutes before the lag sets in.
No changes to Grafana or InfluxDB configuatins.

  • Restarted Grafana, influxdb and telegraf service numerous times.
    -Setting UTC on Grafana does not change anything on the time zone.
  • Changed telegraf interval = “15s” from 10s.
    -All out of Ideas as to where the issue might be.