Closed
Bug 77514
Opened 23 years ago
Closed 23 years ago
History data lost with Mozilla open but idle when crash from other causes occurred
Categories
(Core Graveyard :: History: Global, defect)
Tracking
(Not tracked)
VERIFIED
INVALID
People
(Reporter: xanthian, Assigned: xanthian)
References
Details
(Keywords: dataloss)
Josh Soref requested that I file here this as a separate bug (originally filed as a "me too" to bug 74215, since it seems like a related problem to the loss of bookmarks because of a crash), so here goes, cut and pasted from the entry there: Under Windows 98b, similarly, my history.dat file of perhaps 1.3Mbyte was replaced by a ~365 byte default file after a crash of my entire system with Mozilla launched but idle. The crash wasn't due to Mozilla, but Mozilla should not be leaving files for long periods in unstable states, the crash was probably several hours after the last interaction with Mozilla. This way lies madness. Recommend status for this bug be raised to critical. [Never mind, I started this one out that way.] Losing a history.dat file is, for accessing something like the mailgate.org Usenet interface, like having a lobotomy, all clues differentiating accessed from unaccessed news articles are gone, which can reflect 100's of hours of "work". [End of cut and paste of original.] To answer further an objection from Josh that history.dat is a "perk" and merely "nice to have", making its loss by implication "no big deal", I think this bug is "critical" to product success of Mozilla because losing end user data is pretty much guaranteed to lose that end user and everyone that end user can find to listen, potentially causing the Mozilla product to fail to thrive in the field. End users have a _lot_ higher regard for their time and effort accumulating any flavor of "data" than apparently developers do. A little cut and paste from email sent to Josh: "History.dat" is the browser version of the "ratchet" so ably described by Robert Persig as the main topic of his _Lila_, successor to his _Zen and the Art of Motorcycle Maintenance_; including the "ratchet is the only way progress can exist. [end cut and paste of email sent to Josh] For the end user, the history mechanism is the only way to avoid revisiting all of the pages from potentially, at last report, some 6 billion web pages (only about one per hominid, we are nowhere close to begun) without depending on human memory to identify stuff in a search engine output already seen from stuff worth a look, an absurd alternative in practice. To me, this makes history.dat a critical usability issue for Mozilla, not a "perk". End of sermon, for now.
Comment 1•23 years ago
|
||
this is inane. file a new bug with the cause of the problem. I don't need a soliloquy on the importance of global history.
Status: UNCONFIRMED → RESOLVED
Closed: 23 years ago
Resolution: --- → INVALID
Assignee | ||
Comment 3•23 years ago
|
||
Reopened, since cold crash of a Windows 98B system again caused all history data to be lost. Since alexf asked for a debugging guess rather than a report: Probable cause, Mozilla keeping history file open and writing to it continuously, or just holding it open in general, rather than intermediately doing an open, write, close to limit the dataloss vulnerability window. See the discussion in bug 63292. Bug 63292 may have the same cause but a different outcome, because there Mozilla crashes and the OS closes the open file, apparently without the session updates being committed. Here the OS crashes and, probably, the open history.dat file is left open, and therefore (probably) zero length and replaced or rewritten from the origin with next use of Mozilla, or (possibly) removed by the OS on reboot disk scan cleanup from cold crash, and treated by next launch of Mozilla as never having been initialized. The above is guesswork, and the guesses are not internally consistent, but it was requested guesswork. [The OS crash is partially due to Mozilla, but other browsers interacting with the same disk thrashing background process suffer the same kind of crashes, though the same background process run while thrashing violently without a forground browser in use does not crash even when run for hours. It isn't worth reporting as a fixable bug, it takes hours and lots of good luck to get the particular background process into thrashing mode, but once it is there, browser use makes a crash inevitable.] Anyone who would rather criticize the report than fix the bug is requested to age gracefully but quickly.
Comment 4•23 years ago
|
||
I asked for a NEW bug report, not more rambling in this bug. Please. I beg of you.. use the bugzilla helper: http://www.mozilla.org/quality/help/bugzilla-helper.html
Status: UNCONFIRMED → RESOLVED
Closed: 23 years ago → 23 years ago
Resolution: --- → INVALID
Assignee | ||
Comment 5•23 years ago
|
||
Amazingly, raised by an English teacher, I parsed what you wrote. You didn't parse what I wrote, either time. You have already had the response to your initial unprofessional behavior. Learn to live with it, it will help you recognize humility when you see it in others. Perhaps advice from some outside commentators would help: "IMHO Mr. Dolan should be given an award. He does not suffer fools gladly, and has the nerve to take actions. Although his reaction in this case was a bit over the top, his presence on the net is an asset." -- Bernd Felsche. "Kent is a little touchy sometimes. However, he's usually right." -- Dan Zerkle, in response. Terry_Palfrey@mindlink.bc.ca (Terry Palfrey): >> I stated that you might learn something by stitching the >> whole thread together again and re-reading it. No one >> said you were wrong, just using the wrong perspective, >> you argued with Kent like he was your peer or something, >> he's not. karlt96@nox.nyx.net (Karl Thomas): > Let's try this again. You seem to have the same problem > that he had. Care to come up with something more technical > than he was able to muster? Or do you just idolize Kent > so much if he said the world was flat you would believe him? Uh, let me put this to you gently, if the world was really flat he would be one of the people who knew. Do you have a problem being instructed by someone much smarter and a hell of a lot more experienced than you? -- Terry_Palfrey@mindlink.bc.ca (Terry Palfrey)
Status: RESOLVED → UNCONFIRMED
Resolution: INVALID → ---
Assignee | ||
Comment 6•23 years ago
|
||
Oh, and continuing to mark as "verified invalid" a bug for which you are simultaneously checking in fixes doesn't do much for your image as a rational human being. If you bother to read a few of my several other bug reports and then go back and read this one with your eyes open, you will notice that the entirely different style here was in direct response to Mozilla maintainer requests and inputs, not something I did for the sheer joy of tormenting you, though that latter certainly was an added unexpected benefit you contributed by your reaction. Xanthian. -- Kent, the man from xanth. | Can we hurry this up? | Atheist raising a future Kent Paul Dolan. | I have places to go, | LDS seer and revelator <xanthian@well.com> | and people to insult. | for fun and prophet.
Comment 7•23 years ago
|
||
and THIS is why your bugs are not CONFIRMED. I'm tired of dealing with this bug. I have better things to do with my time than read obtuse quotes about why reading obtuse comments is so important. You seem to be missing my point: I asked for a NEW bug, no matter WHAT you put in this bug, because this bug is already too difficult to understand in under 10 seconds. You can keep adding junk to this bug, but I will NEVER mark it fixed.
Assignee: alecf → xanthian
Comment 8•23 years ago
|
||
I don't know what you're talking about, I don't think many others do either. How about just doing what he asked and filing a newer, clearer bug?
Status: UNCONFIRMED → RESOLVED
Closed: 23 years ago → 23 years ago
Resolution: --- → INVALID
Comment 9•23 years ago
|
||
Did Kent Paul Dolan just quote someone else quoting Kent Paul Dolan? That's classic! this bug report has sunken below the threshold of useful and must go away now. If anyone has a reproducible testcase and cares to elucidate it in a concise manner feel free to do so and cite the bug number here. VERIFIED Invalid
Status: RESOLVED → VERIFIED
Updated•6 years ago
|
Product: Core → Core Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•