THIS BUG WAS CREATED IN RESPONSE TO https://bugzilla.mozilla.org/show_bug.cgi?id=339445#c25 Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b99) Gecko/20090605 Firefox/3.5b99 Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b99) Gecko/20090605 Firefox/3.5b99 Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b99) Gecko/20090605 Firefox/3.5b99 The test case attached to bug 339445 no longer appears to work in the above builds. I've included the test case in the URL field of this bug. STR: 1. Navigate to the test case 2. Trigger a crash using the crashme! extension 3. Restart Firefox 4. Restore the session RESULT: Step 1: 1st field = null, 2nd field = 123 Step 4: 1st field = null, 2nd field = 123 EXPECTED: Step 1: 1st field = null, 2nd field = 123 Step 4: 1st field = 123, 2nd field = 123 I'm going to work on a regression range for this.

Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b1) Gecko/20081007 Firefox/3.1b1 I've checked this in the above build (3.1b1 release). As per bug 339445, the test case should work. It does not however. First field is always null, second field is always 123. As far as I can tell, this is not a regression. It just never worked.

Please make sure to set the pref browser.sessionstore.privacy_level to 0 when testing, otherwise this is not supposed to work per design.

Ok...thanks. I've now narrowed it down to between Firefox 3.1b3 and Firefox 3.5b4

Ok...I've got us down to a 24-hour window: Firefox 3.5b4 20090414 - pass Firefox 3.5b4 20090415 - fail

That's a mighty long list of check-ins. My guess would be on bug 486654...

Bug appears to still be extant in trunk