996811 - Certain inbound-bisections end prematurely

Status: RESOLVED DUPLICATE of bug 1084962

(Testing :: mozregression, defect)

Description

[William Lachance (:wlach)]

Reported from hurda on github:

--

I got a nightly-bisection down to two dates. (Logs are shortened to the necessary bits, bisection done with pull-request #81 applied)
```
C:\mozreg>mozregression --good 2013-11-06 --bad 2013-11-07 --bits=32
Got as far as we can go bisecting nightlies...
Last good revision: 9ba3faa35c96 (2013-11-06)
First bad revision: 70de5e24d79b (2013-11-07)
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=9ba3faa35c96&tochan
ge=70de5e24d79b
```
Then mozreg downloads Nightly 2013-11-05 to catch previous inbound-patches which haven't been landed in Nightly 2013-11-06:
```
... attempting to bisect inbound builds (starting from previous day, to make sur
e no inbound revision is missed)
Downloading build from: http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2
013/11/2013-11-05-03-02-06-mozilla-central/firefox-28.0a1.en-US.win32.zip
===== Downloaded 100% =====
Installing nightly
Getting inbound builds between 770de5942471 and 70de5e24d79b
```
And after a few steps I get the URL for the regression range...
```
Testing inbound build with timestamp 1383745416, revision a775f0d923a62777fc4fcf
02828641afa88c884c
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): good

Testing inbound build with timestamp 1383763046, revision 2b6e3ca41348953bc5915b
419c1c8030c1248cc3
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

Testing inbound build with timestamp 1383755487, revision dc6beafe17c9372a7293a7
6ea6be7f4da6bde8a4
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

Testing inbound build with timestamp 1383751498, revision b6b291d68a7f851241b4d8
28bf96c06f9a676f7b
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

Testing inbound build with timestamp 1383747058, revision 61d3b1f4614c033c336687
db86ef062f76ad6e0a
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

Testing inbound build with timestamp 1383746458, revision e495c0d3c240e9a9dfaf96
b6928e30071800679d
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

Testing inbound build with timestamp 1383746218, revision ffe4494f1333179bcabb84
77e756216bcc6306e7
Was this inbound build good, bad, or broken? (type 'good', 'bad', 'skip', 'retry
', or 'exit' and press Enter): bad

No more inbounds to bisect
Last good revision: 770de5942471
First bad revision: ffe4494f1333
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=770de59
42471&tochange=ffe4494f1333
```
... but as you can see, the "last good revision" is not the revision of the last one with "good", but just the revision of Nightly 2013-11-05.
Also, opening the m-i pushlog-url shows that there are still plenty of revisions to check.

"Restarting" the bisection with
`mozregression --inbound --good-rev=770de5942471 --bad-rev=ffe4494f1333 --bits=32`
eventually allowed me to find the proper m-i regression range.

Comment 1

[Logan Rosen [:Logan]]

Duping this, as it seems to refer to the same bug as the one Bug 1084962 describes. Revert if I'm wrong.