Bug 142753

Summary:

inspector-protocol/dom-debugger/node-removed.html crashes on EWS

Product:

WebKit

Reporter:

Alexey Proskuryakov <ap>

Component:

Web Inspector

Assignee:

Nobody <webkit-unassigned>

Status:

RESOLVED INVALID

Severity:

Normal

CC:

andersca, burg, cdumez, graouts, joepeck, jonowells, mattbaker, nvasilyev, timothy, webkit-bug-importer

Priority:

Keywords:

InRadar

Version:

528+ (Nightly build)

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
Full crash log	none

Description Alexey Proskuryakov 2015-03-16 15:43:44 PDT

I noticed that inspector-protocol/dom-debugger/node-removed.html crashes a lot on EWS.

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   com.apple.WebKit              	0x0000000106e496bd IPC::Connection::invalidate() + 33 (Connection.h:201)
1   com.apple.WebKit              	0x0000000106f97e6e WebKit::WebInspector::closeFrontend() + 50 (type_traits:3006)
2   com.apple.WebKit              	0x0000000106f99ccc WebKit::WebInspector::didReceiveMessage(IPC::Connection&, IPC::MessageDecoder&) + 116 (HandleMessage.h:16)
3   com.apple.WebKit              	0x0000000106e84f8d IPC::MessageReceiverMap::dispatchMessage(IPC::Connection&, IPC::MessageDecoder&) + 125 (MessageReceiverMap.cpp:87)
4   com.apple.WebKit              	0x0000000107016110 WebKit::WebProcess::didReceiveMessage(IPC::Connection&, IPC::MessageDecoder&) + 28 (WebProcess.cpp:622)
5   com.apple.WebKit              	0x0000000106e48db4 IPC::Connection::dispatchMessage(std::__1::unique_ptr<IPC::MessageDecoder, std::__1::default_delete<IPC::MessageDecoder> >) + 102 (memory:2665)
6   com.apple.WebKit              	0x0000000106e4b048 IPC::Connection::dispatchOneMessage() + 106 (memory:2684)

Comment 1 Radar WebKit Bug Importer 2015-03-16 15:44:13 PDT

<rdar://problem/20180789>

Comment 2 Alexey Proskuryakov 2015-03-16 15:45:44 PDT

Skipped in r181583.

Comment 3 Alexey Proskuryakov 2015-03-16 15:47:26 PDT

Created attachment 248760 [details]
Full crash log

Invalidating the connection from a nested run loop looks really really bad.

Comment 4 Alexey Proskuryakov 2015-03-16 15:51:38 PDT

Hmm, maybe it's just a bad patch (bug 142752). It didn't immediately look like it was causing this test to fail, however this crash didn't happen before.

Comment 5 Alexey Proskuryakov 2015-03-16 16:49:11 PDT

Chris unskipped this test in <https://trac.webkit.org/r181592>. Closing as invalid, we can always re-open if this happens again.