We can try to shrink pings if they exceed 1MB. Among other things we could drop whole sections, e.g.: sql, chromehangs, bhr.
If we are really worried about session chain consistency, we could even consider dropping histograms and other non-core data.
If we do this, it would be nice to have visibility into what was dropped - we could add a field listing what was removed and the size of the chunk removed.
sql/chromehangs/bhr are all telemetry-only, right? I think we should probably be moving all of those into separate pings anyway.
(In reply to Benjamin Smedberg [:bsmedberg] (away until 27-July) from comment #3) > sql/chromehangs/bhr are all telemetry-only, right? Yes. > I think we should probably be moving all of those into separate pings anyway. Wouldn't we want them correlated with specific sessions or subsessions? We could still link them via (sub)session ids, but then we'd end up stitching together server-side what could have been submitted together in the first place.