Hey folks! I ran into the following issues when creating a log monitoring dashboard:
Currently, Grafana’s Loki is used to fetch the change logs, and each log is sent out by a certain Tacacs server. One of the items in a log record’s label, namely taIP, represents the source IP of the log (corresponding to a certain server). I need to determine the abnormal situations of each server based on the statistical data of the source server IP of the logs. (If a server hasn’t sent out any logs for more than a week, it will be regarded as being down and requires troubleshooting.)
The current mode of the Grafana panel is that it filters and displays the data it has collected. That means when it hasn’t received any log data from a certain server, the panel simply won’t display anything related to that server.
There are a total of 10 servers, and they will send out different amounts of log data which are then received and managed by Loki. I want to use Grafana to show the number of logs sent by each of these 10 servers within a specific time period. Here’s the problem: During this period, it’s possible that some servers didn’t send any logs, and thus there’s no such data on Loki. At this time, the Grafana dashboard will only display the existing logs according to the data from the data source. But what I want is that the number of logs sent by all 10 servers should be shown on the dashboard, even if it’s zero. So, how can I achieve this?
I guess it might be necessary to first determine the style of the panel to be displayed and then fill in the data by querying.
Any help would be greatly appreciated! Thanks in advance!