"View Source" window re-requests page if user has disabled the memory cache (and disk cache is disabled or page says to skip it)
Categories
(Core :: Networking: Cache, defect, P3)
Tracking
()
People
(Reporter: flammable, Unassigned)
References
Details
(Whiteboard: plan in comment 34 [necko-backlog])
Attachments
(2 files)
Comment 1•19 years ago
|
||
Reporter | ||
Updated•19 years ago
|
Reporter | ||
Comment 3•19 years ago
|
||
Comment 6•19 years ago
|
||
Reporter | ||
Updated•19 years ago
|
Reporter | ||
Comment 7•19 years ago
|
||
Comment 8•19 years ago
|
||
Comment 9•19 years ago
|
||
Comment 10•18 years ago
|
||
Comment 11•18 years ago
|
||
Comment 12•18 years ago
|
||
Comment 13•18 years ago
|
||
Comment 14•18 years ago
|
||
Comment 15•17 years ago
|
||
Comment 16•17 years ago
|
||
Comment 17•17 years ago
|
||
Comment 18•17 years ago
|
||
Comment 19•17 years ago
|
||
Comment 20•17 years ago
|
||
Assignee | ||
Updated•16 years ago
|
Comment 26•16 years ago
|
||
Comment 27•15 years ago
|
||
Comment 28•15 years ago
|
||
Comment 29•15 years ago
|
||
Comment 30•15 years ago
|
||
Comment 32•15 years ago
|
||
Comment 33•15 years ago
|
||
Comment 34•15 years ago
|
||
Comment 35•15 years ago
|
||
Comment 36•15 years ago
|
||
Comment 37•15 years ago
|
||
Comment 38•15 years ago
|
||
Comment 39•15 years ago
|
||
Comment 40•15 years ago
|
||
Comment 41•15 years ago
|
||
Comment 42•15 years ago
|
||
Comment 43•15 years ago
|
||
Updated•15 years ago
|
Comment 44•15 years ago
|
||
Comment 45•15 years ago
|
||
Comment 46•15 years ago
|
||
Comment 47•15 years ago
|
||
Comment 48•15 years ago
|
||
Updated•15 years ago
|
Comment 49•15 years ago
|
||
Comment 50•15 years ago
|
||
Comment 55•11 years ago
|
||
Comment 56•10 years ago
|
||
Comment 57•10 years ago
|
||
Comment 58•10 years ago
|
||
Comment 59•10 years ago
|
||
Comment 60•10 years ago
|
||
Comment 61•10 years ago
|
||
Comment 62•10 years ago
|
||
Comment 63•10 years ago
|
||
Comment 64•10 years ago
|
||
Comment 65•10 years ago
|
||
Comment 66•10 years ago
|
||
Comment 67•10 years ago
|
||
Comment 68•9 years ago
|
||
Comment 69•9 years ago
|
||
Updated•9 years ago
|
Comment 70•7 years ago
|
||
Comment 71•7 years ago
|
||
Updated•7 years ago
|
Comment 72•6 years ago
|
||
Comment 73•4 years ago
|
||
Happened to me just now in Firefox 83.0 (64-bit) on Windows 10.
Comment 74•4 years ago
|
||
This is still occurring in Firefox 85.0 (64-bit) on Windows 10.
Seemingly with random pages, trying to view the source is instead showing the source of the login page for my app, as though the cookie isn't transmitted with the source request. However, after reading the comments here, I checked the Disk Cache and found it was maxed out with only 8k of free space left.
After manually deleting the Disk Cache files (about:cache), the page source is again showing correctly.
Comment 75•4 years ago
|
||
Can the title be updated to indicate that cookies are not sent on the re-request? Severity P3 does not seem to reflect the fact that you can't view the source of pages that require you to be logged in.
Comment 76•4 years ago
|
||
(In reply to mfishe from comment #75)
Can the title be updated to indicate that cookies are not sent on the re-request?
That's a symptom, not a cause, and should not be added to the title. OTOH, the title is already incorrect, so, I'm not sure it would matter. People have been tolerating this bug for 16 years now – so, in other words, there's zero incentive to fix it and, consequently, it won't be fixed. Just keepin' it real...
Anyway, the problem is simply that once the cache memory is maxed out, the older entries are not deleted to make space for new entries. Clearing the cache manually works 100% of the time for this issue and you can then immediately view the source of the page(s) that was/were problematic. For a dev machine, though – one where you might need to be able to see page source regularly – it is a hassle to have to keep clearing the cache, so, I disabled caching entirely. And now every page's source is showing correctly without fail. Here's the configs I used:
browser.cache.disk.capacity = 0
browser.cache.disk.enable = false
browser.cache.memory.capacity = 0
browser.cache.memory.enable = false
Sidenote: Amazon Prime now has a hard time starting up movies and it usually takes a couple refreshes before a movie will stream, although it does stream without glitching once it's started. Pandora will not work in this state at all. And FF seems to be hogging a lot of extra resources and requires more frequent restarts to keep it snappy.
Comment 77•3 years ago
|
||
Firefox 91.0.2, 16 years later still shows login screen source. Common guys...
Comment 78•3 years ago
|
||
FWIW I'm not comfortable with the distinction many seem willing to draw between "web devs" and "everyone else" here. It implies that the remedy being sought is something that only should matter to a web dev, which I'm afraid is bunk. There are actual real-world examples above from people who are not developers but just have their reasons to need to read the static source of the page they've been served.
It's no good to say, here in this secluded spot, that users need to purge their caches beforehand if they definitely don't want this to happen to them. Even if by chance they eventually find this information squirrelled away here, by then it is usually too late. It's a situation that shouldn't arise in the first place.
I'd go as far as to say that a source-view that requires the page to be re-requested is not what anyone wants, because the content may be different for any number of reasons - and even worse, that variability is something the viewer may not even be conscious of, so as such what's presented is misleading to the uninitiated. Absent a full fix for this issue, I'd actually advocate for the re-request not to be automatic, but rather to present an advisory with a confirmation dialog if the user still wishes to proceed.
Comment 79•3 years ago
|
||
This bug is almost old enough to enlist in the military. Then it will be old enough to buy booze. Then it will be old enough to collect social security.
Long story short: it's not getting fixed.
Comment 80•3 years ago
|
||
I just observed this behavior in Chrome today, so that's disappointing.
Comment 81•3 years ago
|
||
This seems as if it was fixed in Firefox version 92.0, and continues to be fixed today (95.0.2). Whether it was accidentally or on purpose, who knows. The latter would obviously be preferred.
Comment 82•3 years ago
|
||
(In reply to Jan Kyu Peblik from comment #81)
This seems as if it was fixed in Firefox version 92.0, and continues to be fixed today (95.0.2). Whether it was accidentally or on purpose, who knows. The latter would obviously be preferred.
I cannot confirm this.
When following my reproduction steps in Comment 55, when trying to load the Main Thread > bitbucket.org > account/settings > (index)
debugger source code, it shows "Error: Incorrect contents fetched, please reload." in the debugger source code window.
Using: Firefox 95.0.1 on Ubuntu 20.04
Comment 83•3 years ago
|
||
(In reply to Daniel Roesler from comment #82)
Seems to work as desired for me, as just tested in Firefox version 95.0.2 on Windows 10. The Debugger tab shows the expected source and not the login page source. The 'profile page' for step 2 from comment 55 I have used is 'https://id.atlassian.com/manage-profile/'. Please advise if another address should be tested against. (I can also test on Ubuntu if we continue to have differing experiences. It would surprise me if the OS-specifics of the build would be at issue, but I suppose such things have happened before.)
Comment 84•3 years ago
|
||
(In reply to Jan Kyu Peblik from comment #83)
Please advise if another address should be tested against.
Attached is a screen capture video (Error_Incorrect_contents_fetched_please_reload_2022-01-03_22.30.30.mp4
) showing the error.
Comment 85•3 years ago
|
||
(In reply to Daniel Roesler from comment #84)
I do see that same behavior over here. My observations are:
To be clear, this is to do with the Debugger part of the Web Developer Tools pane, and not the ordinary View Source system as per this bug's title. The ordinary View Source system does not appear to exhibit this undesired behavior at this time. It's possible you will have to file another bug for this (if this current bug is ever looked at by another developer, anyway).
This page seems to be heavy JavaScript nonsense, and roughly coded at that. I'm afraid I have limited understanding and interest, but offhand it looks like the JavaScript on which the page rendering depends attempts to act upon, for example, events that would only take place during page load, and which would therefore not necessarily be available just by opening the Web Developers Tools pane. It seems possible to me that this JS could by "design" not be compatible with this feature of the Debugger system that works with so many other more ordinary pages. It could be a fundamental, feasibly unavoidable side effect of using so much client-side code.
Anyway, I'll leave further comment on your particular Debugger issue to others in future. Again, the View Source system appears apparently resolved at this time, and that was my own interest. Best of luck.
Comment 86•3 years ago
|
||
(In reply to Jan Kyu Peblik from comment #85)
To be clear, this is to do with the Debugger part of the Web Developer Tools pane, and not the ordinary View Source system as per this bug's title. The ordinary View Source system does not appear to exhibit this undesired behavior at this time. It's possible you will have to file another bug for this (if this current bug is ever looked at by another developer, anyway).
Just to clarify that statement, as it is not immediately clear.
-
This bug (bug 307089) is to do with the fact that using the 'view source' feature will cause the page to be reloaded, rather than showing the source that was fetched in the original page request. It does not cover the source tab in the debugger at all and it does not cover error messages or other incorrect renderings of the source. Comment 81 indicates that this issue may well be resolved now, and this bug is awaiting further confirmation of that.
-
The issue reported by Daniel Roesler is related to the the debugger source tab and therefore should be logged as a separate bug (though, based on comment 85, it may be a site-specific issue in which case it may not be resolvable).
Updated•2 years ago
|
Comment 87•2 years ago
|
||
The severity field for this bug is relatively low, S4. However, the bug has 11 duplicates, 32 votes and 55 CCs.
:kershaw, could you consider increasing the bug severity?
For more information, please visit auto_nag documentation.
Comment 88•2 years ago
|
||
The last needinfo from me was triggered in error by recent activity on the bug. I'm clearing the needinfo since this is a very old bug and I don't know if it's still relevant.
Comment 89•1 year ago
|
||
This bug is still relevant. Today, I wanted to view source on a private page on my site but instead got the login screen.
Description
•