Whenever I load bug 59108, it will "double load" - before the page finishes loading, it will stop and start over. This used to happen to me in Netscape 4, and it might even be the same problem that caused the double loading before. This happens on other pages, but seems to be 100% reproducible with that URL. Build: 2001091608 W98
This is a dup... pretty sure it's bug 88701, but check the screenshot there to make sure it's what you're seeing. *** This bug has been marked as a duplicate of 88701 ***
Status: NEW → RESOLVED
Last Resolved: 17 years ago
Resolution: --- → DUPLICATE
No, it's not doubling the form, it's loading the page twice.
Status: RESOLVED → REOPENED
Resolution: DUPLICATE → ---
I see what you mean (0.9.4/Linux); sorry for the hasty dup. I'm not sure this is a network bug, though -- the whole page loading apparatus looks like it's starting over. I wonder if there's anything weird in the source of that page.
OS: Windows 98 → All
I'm seeing a much more serious problem. When I edit the bug that comes before bug 59108 on the list, the returned page stops, then I get a message asking me if I want to resend the POSTDATA. If things like this happen on credit card transaction pages or other such important things, we could have VERY serious problems. Try it -- get a bug list and edit the bug that falls before bug 59108!
Severity: normal → critical
Status: REOPENED → NEW
Keywords: mozilla0.9.6, nsbeta1, nsCatFood, nsdogfood
Summary: Bug page loads twice → Some pages load twice
Assignee: neeti → darin
skewer: what version of mozilla were you using when you first/last noticed this problem?
Darin: You can see from my comments, this was reported with 2001091608 W98. This still happens in today's build. 100% reproducible after shift+reload.
this is key: "100% reproducible after shift+reload" skewer: can you reproduce it without shift+reload?
No, because normal reload does not download the page; it only checks for updates and displays the cached page, which would not exhibit this bug (leading to suspicion that this is a networking problem). This also happens the first time you load the page since it has expired from the cache.
yeah.. i'm seeing this too.. starting with an empty disk cache shows the problem every time. -> 0.9.6
Severity: critical → major
Status: NEW → ASSIGNED
Priority: -- → P2
Target Milestone: --- → mozilla0.9.6
just tried this with the 2001103113 linux build, and i can't repro this anymore. marking WORKSFORME.
Status: ASSIGNED → RESOLVED
Last Resolved: 17 years ago → 17 years ago
Resolution: --- → WORKSFORME
skewer: if this doesn't happen to you anymore, can you mark as VERIFIED?
It appears that bug 59108 is still loading twice, but only the scrollbar in the listbox is being drawn (instead of the whole page) in the first load. Since this *virtually* solves the problem, I'll verify. There may still be a performance issue but when AMD releases their ClawHammer CPU's it won't matter anyway.
Status: RESOLVED → VERIFIED
On second thought, the problem may not be visible anymore, but I can still see Mozilla clearly loading the page twice in my proxy log. This bug may be less visible but it still happens. Reopening.
Status: VERIFIED → REOPENED
Resolution: WORKSFORME → ---
Created attachment 99058 [details] http log demonstrating the double load yup, i'm able to repro this too (linux trunk build 2002091108). from the log it appears that someone above necko is canceling the first load... not sure why, but i'm looking to see.
Try deleting this line form your prefs.js file: user_pref("intl.charset.detector", "zh_parallel_state_machine");
another to solve this bug is by selecting the menu: View->Character Coding->Auto-Detect->(Off)
-verifyme & mozilla1.2 Is this a config problem or a bug in Autodetect? Should this be moved to HTTP or some other component? Do we need to get this assigned->fixed for the Mozilla 1.3a?
Keywords: mozilla1.2, verifyme
cc'ing nhotta and shanjian... looks like a problem related to charset auto-detection.
bugzilla does not have charset meta tag, so when autodetector in on, new charset will be detected if the page contains some special character. The character in the link can be found in comment #68. So far, we do network reload if doc can be found in cache for charset reload. It might also be possible the because charset reload is issued before the first reload is ready, so there is no cache, and network reload happens. If the concern is about how charset reload should happen, we have other bugs and this behavior won't be changed soon. For Post data concern, that has been addressed. Autodetection will be disabled for post page.
via mail: the previous comment referenced http://bugzilla.mozilla.org/show_bug.cgi?id=59108
change component to i18n. resolve it as wontfix.
Status: REOPENED → RESOLVED
Last Resolved: 17 years ago → 15 years ago
Component: Networking → Internationalization
Resolution: --- → WONTFIX
Mark as verified as won't fix per comment #20.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.