Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #14 by [EMAIL PROTECTED]:
crash fix committed in r4624
unit test committed in r4714
verified success on the subsequent chromebot runs: {4629, 4641, 4657, 4698}
Moving t
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #13 by [EMAIL PROTECTED]:
Ok cool, chromium buildbot 4522 has some useful stacks which hold the
answer:
---
Problem sequence
---
1. Ma
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #12 by [EMAIL PROTECTED]:
Yep, we are in fact hitting case (1).
And your suspicions of socket reuse are spot on.
The new data points are:
- establishing_tunnel_ is never
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #11 by [EMAIL PROTECTED]:
For case (1), it's also possible that reused_socket_ is true, so
we skip connect and proceed to write the request headers. But I
can't explain
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #10 by [EMAIL PROTECTED]:
Silly me, swapped case (1) and case (2) in the comment.
--
You received this message because you are listed in the owner
or CC fields of thi
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #9 by [EMAIL PROTECTED]:
wtc points out that establishing_tunnel_ is set to false inside of the 200
case, so we still need to
explain the mismatch of establishing_tunne
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #8 by [EMAIL PROTECTED]:
Instrumented code results are in, and we now know:
(1) |this| is a valid initialized instance of (HttpNetworkTransaction)
(5) this->url.SchemeIs
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #6 by [EMAIL PROTECTED]:
The results from instrumentation confirm the virtual dispatch on bad object
theory.
The magic value put on the stack is 0xDEADBEFC, which mean
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #5 by [EMAIL PROTECTED]:
> I think the easiest thing to do now is do some runs of chrome-bot to
collect full-
memory dumps.
Alas, Huan says we can't do full memory dum
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #4 by [EMAIL PROTECTED]:
[update with some findings and speculation; nothing conclusive so don't
obligated to read]:
The dumps I looked at were fairly consistent, bein
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #3 by [EMAIL PROTECTED]:
The value of STATE_READ_HEADERS_COMPLETE is 15. I wonder if there's
something special about the value (15 = 0xF). It's strange that we
always c
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #2 by [EMAIL PROTECTED]:
I don't have a good idea on what might have caused this crash.
In HttpCache::Transaction::BeginNetworkRequest() we have just
created a transactio
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
Comment #1 by [EMAIL PROTECTED]:
From report id 63fd8e4f802a3ff:
Thread 4 *CRASHED* (EXCEPTION_ACCESS_VIOLATION @0x625c3a43)
0x02e18909 [chrome.dll - http_network_tra
Issue 3772: Distributed reliability test failures with NewHTTP
http://code.google.com/p/chromium/issues/detail?id=3772
New issue report by [EMAIL PROTECTED]:
crash: http://www.metacafe.com/
crash: http://www.orkut.com/
crash: http://www.passport.net/
crash: http://www.blogger.com/
Issue attri
14 matches
Mail list logo
