777849 - Decrease the timeout value for chromehang telemetry

Status: NEW

(Core :: General, defect)

Description

[(no longer active)]

I don't think it makes too much sense for us to ignore tons of jank which can happen for less than 5 seconds.  I propose that we lower this value to 500 milliseconds, and perhaps split the incoming data into buckets (500ms-1s, 1s-2s, etc.) on the server side.

Then maybe we can talk about the collected data in the snappy meetings and get some people to fix some of them.  ;-)

What do you think Taras?

Comment 1

[(dormant account)]

I think we should wait until our existing chromehang data makes sense. Vladan only got ~40 chromehangs out of last month of nightly reporting.

Comment 2

[(no longer active)]

(In reply to comment #1)
> I think we should wait until our existing chromehang data makes sense. Vladan
> only got ~40 chromehangs out of last month of nightly reporting.

Yep, he mentioned that to me.  Still I don't see why we should not try to work on this in parallel.  Vladan told me that the patch will be fairly simple.

Comment 3

[(dormant account)]

(In reply to Ehsan Akhgari [:ehsan] from comment #2)
> (In reply to comment #1)
> > I think we should wait until our existing chromehang data makes sense. Vladan
> > only got ~40 chromehangs out of last month of nightly reporting.
> 
> Yep, he mentioned that to me.  Still I don't see why we should not try to
> work on this in parallel.  Vladan told me that the patch will be fairly
> simple.

Because I'm worried that the problem might be in the chromehang-reporting mechanism and this would distract us from fixing that.

Comment 4

[(no longer active)]

(In reply to Taras Glek (:taras) from comment #3)
> (In reply to Ehsan Akhgari [:ehsan] from comment #2)
> > (In reply to comment #1)
> > > I think we should wait until our existing chromehang data makes sense. Vladan
> > > only got ~40 chromehangs out of last month of nightly reporting.
> > 
> > Yep, he mentioned that to me.  Still I don't see why we should not try to
> > work on this in parallel.  Vladan told me that the patch will be fairly
> > simple.
> 
> Because I'm worried that the problem might be in the chromehang-reporting
> mechanism and this would distract us from fixing that.

I think these are two orthogonal issues, and I don't see any reason why fixing one of them should block the other.

Comment 5

[Vladan Djeric (:vladan)]

We can do these things simultaneously