Closed Bug 934378 Opened 11 years ago Closed 11 years ago

Intermittent test_bug339494.xhtml | application terminated with exit code 1536 (from "Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967")

Categories

(Core :: Security: PSM, defect, P1)

defect

Tracking

()

RESOLVED FIXED
mozilla28
Tracking Status
firefox26 --- unaffected
firefox27 --- unaffected
firefox28 --- fixed
firefox-esr24 --- unaffected

People

(Reporter: cbook, Assigned: briansmith)

References

()

Details

(Keywords: assertion, intermittent-failure)

Attachments

(1 file)

      No description provided.
Ubuntu VM 12.04 mozilla-inbound debug test mochitest-1 on 2013-11-04 00:05:18 PST for push b4143e04bea1

slave: tst-linux32-ec2-001

https://tbpl.mozilla.org/php/getParsedLog.php?id=30065588&tree=Mozilla-Inbound

-1355257024[88e21e8]: Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967
Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967
TEST-UNEXPECTED-FAIL | /tests/browser/base/content/test/general/test_bug364677.html | application terminated with exit code 1536

00:11:41     INFO -  -1355257024[88e21e8]: Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967
00:11:41     INFO -  Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967
00:11:42  WARNING -  TEST-UNEXPECTED-FAIL | /tests/browser/base/content/test/general/test_bug364677.html | application terminated with exit code 1536
00:11:42     INFO -  INFO | runtests.py | Application ran for: 0:00:48.466696
00:11:42     INFO -  INFO | zombiecheck | Reading PID log: /tmp/tmpW4A01tpidlog
00:11:54  WARNING -  PROCESS-CRASH | /tests/browser/base/content/test/general/test_bug364677.html | application crashed [@ linux-gate.so + 0x424]
00:11:54     INFO -  Crash dump filename: /tmp/tmpG6Rtbi/minidumps/2880c2da-8d6c-c4f6-7ee0d3b6-0be876e1.dmp
00:11:54     INFO -  Operating system: Linux
00:11:54     INFO -                    0.0.0 Linux 3.2.0-23-generic-pae #36-Ubuntu SMP Tue Apr 10 22:19:09 UTC 2012 i686
00:11:54     INFO -  CPU: x86
00:11:54     INFO -       GenuineIntel family 6 model 26 stepping 5
00:11:54     INFO -       1 CPU
00:11:54     INFO -  Crash reason:  SIGABRT
00:11:54     INFO -  Crash address: 0x8ec
00:11:54     INFO -  Thread 4 (crashed)
00:11:54     INFO -   0  linux-gate.so + 0x424
00:11:54     INFO -      eip = 0xb7792424   esp = 0xaf385ec4   ebp = 0xaf386048   ebx = 0x000008ec
00:11:54     INFO -      esi = 0xb761fd9c   edi = 0xb761eff4   eax = 0x00000000   ecx = 0x000008f7
00:11:54     INFO -      edx = 0x00000006   efl = 0x00200282
00:11:54     INFO -      Found by: given as instruction pointer in context
00:11:54     INFO -   1  libssl3.so!ssl3_AuthCertificateComplete [ssl3con.c:b4143e04bea1 : 9965 + 0x3b]
00:11:54     INFO -      eip = 0xb723177a   esp = 0xaf386050   ebp = 0xaf386078
00:11:54     INFO -      Found by: previous frame's frame pointer
00:11:54     INFO -   2  libssl3.so!SSL_AuthCertificateComplete [sslsecur.c:b4143e04bea1 : 1566 + 0x11]
00:11:54     INFO -      eip = 0xb7249163   esp = 0xaf386080   ebp = 0xaf3860a8   ebx = 0xb726b2e4
00:11:54     INFO -      esi = 0xb09b6ba8
00:11:54     INFO -      Found by: call frame info
00:11:54     INFO -   3  libxul.so!nsNSSSocketInfo::SetCertVerificationResult(int, mozilla::psm::SSLErrorMessageType) [nsNSSIOLayer.cpp:b4143e04bea1 : 519 + 0x8]
00:11:54     INFO -      eip = 0xb3dc4cff   esp = 0xaf3860b0   ebp = 0xaf3860d8   ebx = 0xb6bdbde8
00:11:54     INFO -      esi = 0xb09b6ba8
00:11:54     INFO -      Found by: call frame info
00:11:54     INFO -   4  libxul.so!mozilla::psm::::SSLServerCertVerificationResult::Run [SSLServerCertVerification.cpp:b4143e04bea1 : 1368 + 0xe]
00:11:54     INFO -      eip = 0xb3dd79d5   esp = 0xaf3860e0   ebp = 0xaf3860f8   ebx = 0xb6bdbde8
00:11:54     INFO -      esi = 0x0af4b0d8   edi = 0xaf386128
00:11:54     INFO -      Found by: call frame info
00:11:54     INFO -   5  libxul.so!nsThread::ProcessNextEvent(bool, bool*) [nsThread.cpp:b4143e04bea1 : 622 + 0x13]
00:11:54     INFO -      eip = 0xb4479b2a   esp = 0xaf386100   ebp = 0xaf386158   ebx = 0xb6bdbde8
00:11:54     INFO -      esi = 0x088e1ee0   edi = 0xaf386128
00:11:54     INFO -      Found by: call frame info
00:11:54     INFO -   6  libxul.so!NS_ProcessNextEvent(nsIThread*, bool) [nsThreadUtils.cpp:b4143e04bea1 : 251 + 0xf]
00:11:54     INFO -      eip = 0xb4433708   esp = 0xaf386160   ebp = 0xaf386198   ebx = 0xb6bdbde8
00:11:54     INFO -      esi = 0x088e14c0   edi = 0x088e1ee0
00:11:54     INFO -      Found by: call frame info
00:11:54     INFO -   7  libxul.so!nsSocketTransportService::Run() [nsSocketTransportService2.cpp:b4143e04bea1 : 688 + 0x9]
00:11:54     INFO -      eip = 0xb30e0103   esp = 0xaf3861a0   ebp = 0xaf3861f8   ebx = 0xb6bdbde8
Component: Security → Security: PSM
Product: Firefox → Core
This was probably caused by the landing of NSS 3.15.3 BETA 2 in bug 898431.
Blocks: 898431
Assignee: nobody → brian
Blocks: 713933
(Tweaking summary to try and avoid the html escaping issue)
Summary: Intermittent TEST-UNEXPECTED-FAIL | /tests/browser/base/content/test/general/test_bug364677.html | apAssertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967 → Intermittent test_bug339494.xhtml | application terminated with exit code 1536 (from "Assertion failure: ss->ssl3.hs.ws == wait_new_session_ticket || ss->ssl3.hs.ws == wait_change_cipher || ss->ssl3.hs.ws == wait_finished, at ssl3con.c:9967")
FYI I hit this running a local debug Win32 build going to http://talky.io, every time (ok, 3 for 3).

This is a problem in that talky.io is one of the primary webrtc sites we're working with and use for dogfooding.

Can we either solve this ASAP or backout until it's dealt with?
Severity: normal → critical
Flags: needinfo?(brian)
Also hit it on Bugzilla.  Makes browsing really problematic.  And what happens on a non-debug build without the assertion?
I hit this assertion failure about 70% of the time when I start a debug build on page [1] and then click a bookmark (on my bookmark toolbar) for page [2]. I can't reproduce any crashes or hangs when using the same STR with a release build.

This assertion is not Linux-specific because I'm running OS X 10.9.

[1] http://www.washingtonpost.com/blogs/the-switch/wp/2013/11/04/how-we-know-the-nsa-had-access-to-internal-google-and-yahoo-cloud-data/

[2] https://panopticlick.eff.org/index.php?action=log&js=yes
OS: Linux → All
Hardware: x86 → All
(In reply to Chris Peterson (:cpeterson) from comment #31)
> I hit this assertion failure about 70% of the time when I start a debug
> build on page [1] and then click a bookmark (on my bookmark toolbar) for
> page [2]. I can't reproduce any crashes or hangs when using the same STR
> with a release build.
> 
> This assertion is not Linux-specific because I'm running OS X 10.9.
> 
> [1]
> http://www.washingtonpost.com/blogs/the-switch/wp/2013/11/04/how-we-know-the-
> nsa-had-access-to-internal-google-and-yahoo-cloud-data/
> 
> [2] https://panopticlick.eff.org/index.php?action=log&js=yes

(In reply to Randell Jesup [:jesup] from comment #28)
> FYI I hit this running a local debug Win32 build going to http://talky.io,
> every time (ok, 3 for 3).
> 
> This is a problem in that talky.io is one of the primary webrtc sites we're
> working with and use for dogfooding.
> 
> Can we either solve this ASAP or backout until it's dealt with?

Thanks. This is awesome info. I will debug this in the morning. Either I'll back out the patch or I'll check in an immediate fix.
Flags: needinfo?(brian)
If you guys hit this assertion again, can you break into the debugger and find out what the value of ss->ssl3.hs.ws is? Either the symbolic name or the numeric value. Thanks.
Flags: needinfo?(rjesup)
Flags: needinfo?(cpeterson)
Flags: needinfo?(brian)
ss->ssl3.hs.ws is wait_server_key

I'll try to leave it sitting in the debugger
Flags: needinfo?(rjesup)
Attached file minidump of crash on talky.io —
This should help
I am working on the patch in NSS bug 935847. Please continue to star *this* bug in TBPL so the NSS bug doesn't get the TBPL Robot noise. I will mark this bug fixed after we update NSS to contain the fix.

Special thanks to :jesup for giving me the information I needed to diagnose the problem.
Flags: needinfo?(cpeterson)
Flags: needinfo?(brian)
Whiteboard: [testing patch]
Target Milestone: --- → mozilla28
Whiteboard: [testing patch]
This was reproducible in bughunter which can used to verify this is fixed after it lands.
https://hg.mozilla.org/mozilla-central/rev/61fb80e560de
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: