I want to store the frontend logs into promethus, an example of frontend log metric looks like this:
{
name: 'fcp',
Labels:
- pageName
- traceID
- userAgent
}
{
name: 'component_render_times',
Labels:
- componentID
- pageName
- traceID
- userAgent
}
the traceID will be different for every page visit, it is unchanged if user stay at the same page. My question is: will this cause cardinality explosion? if so, is there an alternative? I want to analysis the time series data for every traceID, like is the performance get worse if the page openned for days? whats the total components it rendered for that page? if some data point's performance metric very bad, I need the traceid to find more related metrics to find out whats wrong with it