1615290 - Validate the 'async init' improvement once it hits Fenix

Status: RESOLVED FIXED

(Data Platform and Tools :: Glean: SDK, task, P1)

Description

[Alessio Placitelli [:Dexter]]

This is to verify that bug 1610574 did the right thing.

Comment 1

[Alessio Placitelli [:Dexter]]

I put together this query for checking the volume and the number of clients from the past 3 weeks up until today.

The Fenix Nightly that includes the async init is the one from the 19th of February 2020. The preliminary data we have shows that volumes are in the expected range and client counts are nominal.

The ingestion console shows nothing out of the ordinary with respect to missing fields.

Comment 2

[Alessio Placitelli [:Dexter]]

I executed the query from comment 1 again and the volumes seem to be in line with our expectations. Chris, could you give this a sanity check (you volunteered!) so that I can close this?

Comment 3

[Chris H-C :chutten]

The queries look good and your conclusion seems sound.

The ping volumes (specifically the dip on the 21) seem interesting, though. Do we have an explanation for those?

Comment 4

[Alessio Placitelli [:Dexter]]

(In reply to Chris H-C :chutten from comment #3)

The ping volumes (specifically the dip on the 21) seem interesting, though. Do we have an explanation for those?

Not really. The new build with the async changes was released on the 19th (see this chat), so the dip on the 21st can be related but... doesn't really match up. I would also expect that change to be something that would trigger a much bigger data loss or ping corruption. We saw neither :(

Does the above make sense?

Comment 5

[Chris H-C :chutten]

Oh yes, sorry I wasn't clear: I don't know that it would have anything at all to do with the changes (and even if it did, it seems to be working out quite well for us). I was just noticing that it's a peculiarity and wondering if we already knew what caused it.

Comment 6

[Alessio Placitelli [:Dexter]]

(In reply to Chris H-C :chutten from comment #5)

Oh yes, sorry I wasn't clear: I don't know that it would have anything at all to do with the changes (and even if it did, it seems to be working out quite well for us). I was just noticing that it's a peculiarity and wondering if we already knew what caused it.

Good call! Let me flag Frank in case he knows :)

Comment 7

[Frank Bertsch [:frank]]

Nope, I'm not familiar with any issues relating to that dip. It's small enough that I'd imagine it's not a big deal.

Comment 8

[Alessio Placitelli [:Dexter]]

All right, closing this then! Thanks Frank!