I think the first instance of this crash was http://build.webkit.org/results/SnowLeopard%20Intel%20Release%20(WebKit2%20Tests)/r89604%20(12857)/results.html, but I am not absolutely sure. If that was the first instance, then it was probably one of the following revisions that started it: http://trac.webkit.org/changeset/89599 http://trac.webkit.org/changeset/89604 Since then, I have seen it twice: http://build.webkit.org/results/SnowLeopard%20Intel%20Release%20(WebKit2%20Tests)/r89609%20(12858)/results.html http://build.webkit.org/results/SnowLeopard%20Intel%20Release%20(WebKit2%20Tests)/r89623%20(12863)/results.html I am going to add this test to the mac-wk2 skipped list in order to get the bots green.
I'm OK with rolling out both patches to see if the crashing stops. abarth?
Note that my patch only affects MHTML related code. Since none of these tests is related to MHTML (and since MHTML is not enabled by default) I don't see how it could have broken things. But I have been wrong in the past... so feel free to roll out.
Of those two patches, http://trac.webkit.org/changeset/89604 seems more likely given that it has the word "inspector" in the diff and the other patch isn't compiled into that port. However, it does seems like a reasonably safe patch. :) I'd be inclined to just roll out 89604 to see if that stops the crashes.
(In reply to comment #3) > Of those two patches, http://trac.webkit.org/changeset/89604 seems more likely given that it has the word "inspector" in the diff and the other patch isn't compiled into that port. However, it does seems like a reasonably safe patch. :) > > I'd be inclined to just roll out 89604 to see if that stops the crashes. Sounds good to me. I'll roll it out and re-enable the tests.
It looks like the test hasn't been disabled yet (at least, I can't seem to find network-cachedresources-with-same-urls.html in any of the Skipped files). This test also hasn't crashed since build 12863 yesterday. I'm not sure if I should roll out or not. It's too bad we don't have more historical data about this test.
Maybe we should gather more data? Your patch looked pretty darn safe.
(In reply to comment #6) > Maybe we should gather more data? Your patch looked pretty darn safe. I agree that we shouldn't necessarily roll out the patch. I think a better approach would be to have somebody run the test a bunch of times, trying to get it to crash, and see if they can debug from there. If we do see the crash again on the bots, I think it would be best to disable it or skip it.
Closing as invalid, as this bug pertains to the old inspector UI and/or its tests. Please file a new bug (https://www.webkit.org/new-inspector-bug) if the bug/feature/issue is still relevant to WebKit trunk.