On 12-08-30 2:34 AM, Mike Hommey wrote:
On Wed, Aug 29, 2012 at 07:03:17PM -0400, Ehsan Akhgari wrote:
I don't believe that the current situation is acceptable, especially
with the recent focus on performance (through the Snappy project),
and I would like to ask people if they have any ideas on what we can
do to fix this. The fix might be turning off some Talos tests if
they're really not useful, asking someone or a group of people to go
over these test results, get better tools with them, etc. But
_something_ needs to happen here.
Ideally, we should make talos regressions visible on tbpl as oranges,
and star them as other oranges.
The problem is that with the current state of signal/noise of talos
results, this would lead to
- a whole lot of oranges
- oranges showing up long after the fact (by the time the email
notification comes, plenty more pushes have happened already)
The latter makes it hard to handle such oranges as other oranges.
The way that we handle similar situations with test failures is to close
inbound, back out enough patches so that things go green again, and then
reopen, and ask the people that had been backed out to retry landing
after testing on try.
This is entirely possible for Talos regressions as well if we want to,
it just takes longer, and will mean more patches will get backed out.
Ehsan
_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform
