Closed
Bug 127546
Opened 22 years ago
Closed 14 years ago
crash report should include alltime's high for mtbf
Categories
(mozilla.org :: Talkback Server & Webtool, task)
Tracking
(Not tracked)
RESOLVED
INVALID
People
(Reporter: bernd_mozilla, Assigned: jay)
References
()
Details
as pointed out in the mainfesto a criteria for 1.0 would be # upward MTBF trend, all-time maximum it would be good to have a graph but at least in the report the alltime maximum should be shown now.
Comment 1•20 years ago
|
||
changed URL from "ftp://ftp.mozilla.org/pub/data/crash-data/seamonkey-crash-analysis.txt" to "ftp://ftp.mozilla.org/pub/mozilla.org/data/crash-data/seamonkey-crash-analysis.txt"
Comment 2•18 years ago
|
||
Reporter, is this bug still valid? The linked-to crash report is 3 years old.
Comment 3•18 years ago
|
||
I'm updating the Product and Component, since the crash analysis is not part of the Talkback client itself. Also changing the URL, since the crash analysis is now part of Talkback Public's Reports section. Just select the desired Product and then click on either the Simple Reports or Smart Analysis links on the left. Note that the Talkback client will likely be replaced by Airbag in the new future.
Severity: normal → enhancement
Component: Talkback Client → Talkback Server & Webtool
OS: Windows 98 → All
Product: Core → mozilla.org
Version: Trunk → other
Comment 4•18 years ago
|
||
resetting to default owner... I'm pretty sure it's not Shiva these days. :)
Comment 6•18 years ago
|
||
(In reply to comment #3) > I'm updating the Product and Component, since the crash analysis is not part of > the Talkback client itself. I think you're assuming every single crash will be reported to the site. This is rarely the case, so I believe it will be useful for the talkback client to keep mtbf/crash statistics internally.
Comment 7•18 years ago
|
||
(In reply to comment #6) > I think you're assuming every single crash will be reported to the site. This > is rarely the case, so I believe it will be useful for the talkback client to > keep mtbf/crash statistics internally. > But I believe the *mean* time between failures (MTBF) is only meaningful if you average over a large number of users, platforms, and days. Otherwise it will jump all over the place depending on how you use the browser.
Comment 8•18 years ago
|
||
(In reply to comment #7) Well there are different kinds of means. You may be interested in the mean time between failures on a specific user's system or on the systems of a small number of users. Plus, what you said does not contradict my argument, as even if you want to take the mean over many users, you still need to count the crashes that happen but aren't reported.
Assignee | ||
Comment 9•18 years ago
|
||
MTBF numbers have been inaccurate for a while (not sure exactly why), but most likely due to A LOT more users submitting bad data than we used get in Talkback reports (back at Netscape, it was mostly nightly users and internal engineers/qa...in a more controlled environment). I don't think I'm going to have time to work on this bug in the near future, but if I will try to recruit volunteers to help once I get a few "Talkback Reports" projects organized for the community.
Comment 10•14 years ago
|
||
Talkback isn't used anymore: R.Invalid now.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → INVALID
You need to log in
before you can comment on or make changes to this bug.
Description
•