mozilla :: #firefox

13 Sep 2017
00:28solenodicwhat's with all these new cross-origin policy errors in nightly
00:29solenodicis there some about:config option to loosen the restrictions to pre-firefox 57
00:30solenodicwhat's with all these new cross-origin policy errors in nightly
00:30solenodicis there some about:config option to loosen the restrictions to pre-firefox 57
00:31solenodicspecifically all the "unsafe eval" errors
00:31solenodicinline scripts blocked by default now or something?
00:32Caspy7solenodic: you may also try #security
00:36solenodichmm nevermind
00:36solenodicit has to be an addon
00:37solenodicactually hmm
00:39solenodicI'll collect a list of sites that it happens on without any addons enabled
01:04grahamperrin-MWhat might be the best workaround to Firefox representing stale content after restoring a session? Should I experiment with a change of `browser.cache.disk.enable` to `false`?
01:20grahamperrin-MMost noticeable with GitHub issues. I post a comment, quit Firefox, start Firefox, the GitHub Page is as if the comment was not posted. A terribly frequent problem.
01:24katniplike it's not saved eh
01:59firebotJust appeared in Planet Mozilla - :
01:59firebot Mozilla Open Policy & Advocacy Blog: Announcing the 2017 Ford-Mozilla Open Web Fellows!
05:41adrian_1908 <- Does &quot;(obsolete)&quot; refer to the setting (3) itself or?
05:44adrian_1908Found another article, looks like the setting itself (value 3) is no longer used.
06:16grahamperrin-Mkatnip: the opposite, in a way.
06:18grahamperrin-MFirefox restores the page as if my comment to GitHub was not saved (to GitHub) when truly, the comment was saved. In addition it&#39;s commonplace to find the text of the comment in the comment field, ready to submit.
06:19grahamperrin-MIn these cases I rely on my own memory (not the memory of Firefox): I&#39;m almost certain that what&#39;s on screen is false so I (Control-A), Control-X) cut what&#39;s in the comment field then (Control-F5) reload the page to reveal the truth.
06:20grahamperrin-MAFAIR if I don&#39;t cut the comment from the comment field then a simple reload might present both my posted comment, plus duplicate text in the comment field, ready to (re-)post.
06:28Corkgrahamperrin-M: might be that github have scripts that tries to stor old comment data and it conflicting with the built in solution in firefox
06:28Corkgithub uses a lot of javascript inline modifications for the comment system
06:29grahamperrin-MCork: thanks. Where, on disk, does the Firefox solution store its data? In the (normal) disk cache?
06:33grahamperrin-MNow might be a good time to mention that I never have a sessionstore.js in my profile directory whilst using Firefox. For example:
06:33* grahamperrin-M sent a long message: grahamperrin-M_2017-09-13_06:33:18.txt <>
06:42* grahamperrin-M finds <>
06:43grahamperrin-MFirefox, the Browser that has your Back[up] | Il y a du th renvers au bord de la table there are any new information about to how can be customize the scrollbars on Nightly (size, colour)?
06:44grahamperrin-M > snapshot of the state of the browser whenever anything happens, whether the user browses, fills a form, scrolls, or an application sets a Session Cookie, Session Storage, etc. (this is actually capped to one save every 15 seconds, to avoid overloading the computer).
06:45grahamperrin-M> Whenever a user interaction or a script requires it, Firefox writes the contents of Session Restore to a file called sessionstore-backups/recovery.js.
06:47grahamperrin-MOK, that essential detail seems to be missing from e.g.
06:52grahamperrin-M may be similarly misleading. If the 2014 article by yoric is accurate, then I&#39;ll update the Mozilla wiki.
07:12grahamperrin-MFor a long time, my `browser.sessionstore.interval` has been cautiously set to `60000` (sixty seconds, four times as long as the default). That has not worked around the issue
07:15grahamperrin-M so, Cork, I might raise something in Bugzilla@Mozilla. Maybe there should be enhancements to session storage to work around issues with sites such as GitHub.
07:16grahamperrin-MOTOH I see the issue so frequently, I&#39;m surprised that there&#39;s no mention elsewhere. Maybe it&#39;s an issue with Session Manager extension.
07:16Corkgrahamperrin-M: the best suggestion i can give is to put the profile in a ram drive
07:16Corkit is how i solved it when i had those problems
07:16grahamperrin-MAh, did you, too, have problems with/affecting GitHub?
07:17Corkno not github, but the session problem
07:17Corkit was a long time ago now, on ssd since many years
07:17grahamperrin-MCan you recall which domains were affected in your case?
07:18Corkuh, seams i read too much into browser.sessionstore.interval >_>
07:18Corkfor me i had hangs or disk io problems cause of lots of tabs
07:18Corknot sure what else you meant by that line
07:19Cork*could have meant
07:21grahamperrin-MCurrently I have 706 tabs across thirteen windows, but only 28 loaded.
07:23grahamperrin-MSo, yeah, I occasionally wondered whether Firefox and/or Session Manager occasionally had difficulty writing to required files.
07:24grahamperrin-MCERTAINLY Firefox is sometimes slow to quit, a remarkable delay between disappearance of the GUI and ending of the firefox process.
07:25grahamperrin-MA long shot, atime. line 23
07:33grahamperrin-M output from htop not long before a quit of Firefox with ~28 of 700+ tabs loaded. Quit completed in less than thirty seconds, which is relatively good compared to some quits.
07:35* grahamperrin-M uses Session Manager, after launching Firefox without automatically restoring the session, to load 704 tabs across thirteen windows
07:40grahamperrin-M with disk cache and multiprocess disabled, with ninety-two extensions enabled (one hundred, according to `about:healthreport`). All thirteen windows rendering after ~155 seconds, which is good, given the set of extensions.
08:48Jan\omg! 100 addons ?
08:55DaggerI have 128 installed in one instance of Firefox... only 83 enabled though
08:55Daggerit&#39;s not particularly crazy. modern Firefox has at least that many broken features, deliberately or otherwise
09:32Jan\you must be excited about the 57 release then
09:36Dagger57 has problems that have nothing to do with number of extensions :/
13:15firebotJust appeared in Planet Mozilla - :
13:15firebot The Mozilla Blog: Mozilla Announces 15 New Fellows for Science, Advocacy, and Media
13:32grahamperrin-M[Jan]( I&#39;m this excited:
13:32* grahamperrin-M uploaded an image: 2017-09-13 14:06:33.png (311KB) <>
13:33grahamperrin-MDagger: test, please ignore
13:34Kylek14ignoring :P
13:34grahamperrin-M[Jan]( test, please ignore. Finished testing.
14:10firebotJust appeared in Planet Mozilla - :
14:10firebot Mozilla Open Innovation Team: Open Source Needs Students To Thrive
14:17DuClare> &quot;how can I pay my student loans with open source&quot;
14:17DuClareImportant question, unassessed
14:18MardegDuClare: did you mean &quot;unaddressed&quot;?
14:19DuClareSorry, I&#39;m drunk from not having drunk enough coffee
14:23Kylek14Drinking tea here. However, &quot;You can help pay your loans in experience!&quot;
14:24DuClareWhat does that mean?
14:29Kylek14it&#39;s a play on the &quot;it&#39;s an unpaid internship, but you&#39;ll get paid in experience!&quot;
14:30DuClareYeah I thought it was something like that. It&#39;s terrible :(
14:32Kylek14never had an internship when I was in college, it was either get an internship or not pay for school the next semester :D
14:33Mardegthe outternship is the way of the future
14:44Kylek14never heard that term before
14:45DuClareI feel a little left out when everyone&#39;s so concerned about students
14:45DuClareGSoC etc. are not accessible for non-students
14:46DuClareCompanies look for students to intern
14:46DuClare.. then they look to employ senior software developers
14:47DuClareWhat about all the people who aren&#39;t students and who aren&#39;t senior developers? :(
14:51Kylek14yeah, not gonna lie, the work search sucks right now with that. Need to have 10 years experience upon leaving school, but nobody will hire/intern you without experience.
14:51Kylek14on a side note. I hate Flash with a passion. Causing all sorts of headaches at work.
16:10The0x539Is there any good way to cut down Nightly battery usage on Win10?
16:13Kylek14What do you have it doing during the night?
16:14The0x539haha :|
16:15Kylek14no seriously, what exactly is running during the evening that&#39;s killing the battery?
16:16Kylek14The obvious answer would be just to turn it off but I&#39;m assuming you want to leave it on for a reason. So probably killing bluetooth, disabling wifi and turning the screen brightness all the way off would help significantly.
16:17Kylek14also look at what you have running in the background like Steam or other programs that like to run in the system tray.
16:18The0x539I said Nightly battery usage
16:18The0x539as in power usage due to running &quot;Firefox Nightly&quot;
16:18Kylek14crap. my bad
16:20Kylek14Can&#39;t help ya there. I&#39;ve been out of the loop with just regular builds for the last year or so.
16:21Kylek14so nightly are a whole nother level
16:29Caspy7The0x539: one factor that comes to mind is the extra telemetry in Nightly. Would at least be interesting to contrast to 57 Beta. Though perhaps more significant is the fact that its your web browser based on what it does and how much you use it compared to the other apps in that list and it&#39;s obviously going to use more energy. The fairer comparison would be to other versions of Firefox and...
16:29Caspy7...other web browsers. Is it worst than earlier versions of Firefox or Chrome.
16:29Caspy7*worse than
16:30The0x539I probably *should* do a comparison but even without one I&#39;m under the impression that it&#39;s using a sizable amount
16:30The0x539almost certainly a fair chunk more than Edge does
16:30The0x539if only I were willing to use edge
16:32TimvdeFirefox definitely uses more battery than Edge
16:32TimvdeBut what you get in return is of course a lot more ;)
16:33Caspy7Timvde: not that much
16:33Kylek14well that&#39;s interesting
16:34TimvdeCaspy7: Afaik, Edge has improved even more since then, while Firefox hasn&#39;t
16:34TimvdeChrome also claims to have surpassed us
16:35Caspy7Timvde: the prevailing idea that Firefox was so much worse than edge was perpetuated by MS with a lab test that they concocted and performed one one type of computer. But their own usage numbers contradict them
16:36Caspy7Timvde: cool, did they provide some numbers. Preferably usage comparisons.
16:36TimvdeI know, but the numbers are old
16:36TimvdeCaspy7: You have definitely seen the comparison, it&#39;s also not new
16:37Caspy7Timvde: I&#39;m talking about your Chrome claim
16:37TimvdeCaspy7: Chrome did implement background tab throttling in the meantime
16:38The0x539should probably bring up that I have yet to switch off irccloud :|
16:38Caspy7Timvde: I&#39;m not saying Firefox can&#39;t get better or Chrome hasn&#39;t, but people keep treating it as foregone conclusion and it&#39;s often based on marketing
16:39Kylek14Marketing + PR always dominates, regardless of reputation.
16:39TimvdeCaspy7: Seems like I was wrong, they only compared against themselves:
16:40Caspy7and you seemed so confident
16:41TimvdeCaspy7: I was probably confused with Microsoft&#39;s update to their original test:
16:42TimvdeAgain, vimeo though.
16:43Caspy7again, rigged marketing hubbub
16:43Caspy7their original numbers didn&#39;t seem to reflect reality and nothing changed, so
16:46Caspy7MS has a long and glorious history of creating rigged tests and funding rigged research. So that&#39;s really not helping.
16:46jeroen7sgoogle: owns youtube, performs tests with vimeo, posts video of tests with vimeo to youtube
16:48Caspy7well that&#39;s not too surprising
16:49Caspy7for those who haven&#39;t seen yet, there is apparently a new and great horror for the computer world
16:49Caspy7basically every BlueTooth enabled device is vulnerable to infection
16:52jeroen7sseen it alreaddy
16:55firebotJust appeared in Planet Mozilla - :
16:55firebot Air Mozilla: Weekly SUMO Community Meeting September 13, 2017
17:00decebalusyeah..another big panic
17:01The0x539conspiracy by the headphone jack removers :o
17:01Caspy7decebalus: nothing to worry about, you think
17:02decebalusnot at all
17:04Caspy7good thing you&#39;re here to keep us grounded
17:05Caspy7decebalus: oh, forgot to ask, what are you basing that on?
17:14decebalusCaspy7: I don&#39;t use smartphone/bluetooth/wireless connexions..;)
17:15Caspy7decebalus: so, your non-use of bluetooth makes this described exploit &quot;another big panic&quot; and nothing to worry about for others?
17:20decebalusCaspy7: man became too dependent of this devices
17:21alex_mayorgaCaspy7: Just stay away from the delivery guy, per the video =)
17:21dark-knighthello my friends, i have a question. i am on 56.0b11 (64-Bit) and the site i use is HTTPS (lets encrypt) but firefox still shows a formular warning that the form data is not secured
17:21dark-knightany ideas why?
17:21Caspy7alex_mayorga: I think you&#39;re on to something
17:22Mossopdark-knight: Maybe the form will submit to an insecure page?
17:22Caspy7dark-knight: try clicking on the (i) icon in the URL bar. See if it says something more
17:23Caspy7Mossop: I think in mainly (only?) happens when the https security is actually broken
17:23alex_mayorgadark-knight: is the login form a frame, perhaps?
17:23dark-knightthe cert is valid and everytrhing is fine (greein)
17:23dark-knightthere are no frames on that page
17:23Caspy7can you link the page?
17:24dark-knightah, stupid me ... the site is TLS with strict transport security, but the form still points to a http resource ...
17:24dark-knightsorry for asking :(
17:25Caspy7Mossop: you were right
17:32decebalusCaspy7: I think this guys are much happier than the majority of the rest of the world :D
17:33Caspy7decebalus: your attempt at redirection is stellar
17:34decebalusyou think
17:55firebotJust appeared in Planet Mozilla - :
17:55firebot Air Mozilla: The Joy of Coding - Episode 112
18:25firebotJust appeared in Planet Mozilla - :
18:25firebot Dave Townsend: How do you become a Firefox peer? The answer may surprise you!
19:30Jan\Dagger: problems like what?
19:52Caspy7The0x539: note bug 1397092
19:52firebot ASSIGNED, High Cpu/GPU usage due to &quot;Ping Pong&quot; loading indicators in Firefox 57 affecting browser performance
19:53The0x539Caspy7: relevant to Win10?
19:54Caspy7The0x539: you were complaining of battery usage, yes? Higher CPU may contribute to that
19:55The0x539Seems linux-specific to me?
19:57Caspy7The0x539: why do you say that? The reporter explicitly specified windows 10
19:58The0x539Whoops, wrong bug tracker
19:59Caspy7I assumed you&#39;d click the link that firebot provided
19:59The0x539I have firebot ignored and used a search engine
19:59The0x539First result was for redhat...
20:01Caspy7funny, I find firebot to be very helpful (such as for bug links of course)
20:02SeburoI do not ignore Firebot, mainly because I am scared to.
20:03Caspy7back a few years ago it would frequently post newly filed bug in the channel. I can understand wanting to ignore it and limit the noise
20:11DuClareCaspy7: Does it post them on another channel now?
20:12Caspy7I don&#39;t remember
20:26firebotJust appeared in Planet Mozilla - :
20:26firebot This Week In Rust: This Week in Rust 199
20:56firebotJust appeared in Planet Mozilla - :
20:56firebot Will Kahn-Greene: Socorro and Firefox 57
21:11firebotJust appeared in Planet Mozilla - :
21:11firebot Mozilla Security Blog: Verified cryptography for Firefox 57
22:25ciaranI&#39;m having some trouble with WebSockets I&#39;m using a default test cert on my server which it&#39;s having trouble verifying and so it rejects the connection, I want to disable this verification
22:26MossopIf you visit the server in a tab it will let you set a security exception for the cert
22:29ciaranMossop: awesome thanks
22:37nemodonHow do I build a code coverage build without crashing on asserts? I tried xpcom_debug_break env variable as warn, but that doesn&#39;t seem to work.
22:47Caspy7nemodon: try #introduction
22:50nemodonOk. Thanks.
22:54zokumI have a recurring problem with all recent versions of firefox. after a while they stop rendering windows properly, either they start repeating the content or add large areas of black or white where the rendered content should be. I used to have to kill the browser to get it to work properly. These days I only kill the rendering thread, and I can fairly quickly get back to work. The problem
22:54zokumseems to happen when the thread is using somewhere bove 2.5 gig memory, it was 2.6 when i just killed it
22:57zokumthe problem persisted after a complete reinstall of os on another harddrive, windows 7
22:57Caspy7zokum: what you&#39;ve described is usually (well, pretty well always) to do with graphics drivers. You can prove it by disabling hardware acceleration and restarting, though this is a pretty long step so best to start with checking that your graphics drivers are completely up to date according to the GPU manufacturers website
22:57Caspy7Microsoft/windows update cannot be trusted for this
22:58Caspy7once I had a laptop and found that the drivers on it were years old. I only had it for 6 months. I came with old drivers and Windows didn&#39;t find updates
22:59zokumdrivers should be fairly up to date
22:59zokumthis has happened for years, among numeros driver revisions
23:01zokumthe gfx card is ~5 years old, and drivers are from 2017, by now the basic bugs should have been ironed out
23:02Caspy7zokum: please go to about:support in your URL bar and copy the Graphics section and paste to and share the link here
23:03kbrosnan2.5gb is around the 32 bit process limit. x64 firefox may be helpful. though up to date gfx drivers would be good too
23:03zokumok, it lists errors in the log, so hopefully, we can get this bug fixed
23:06zokumeven if a 64bit version would &quot;fix&quot; this, it shouldn&#39;t corrupt the output all over the place and go on rendering garbage
23:06Caspy7zokum: what OS? 32 or 64 bit?
23:06zokum64bit, windows 7 enterprise
23:07zokum8 gig ram, 1 gig on gfx if i&#39;m not mistaken
23:09Caspy7zokum: as I said, that garbage rendering is a description of graphics issues. Likely a leak leading to crashing. And on 32 bit, as was said, around 2.5-2.6GB is where Firefox starts having issues and crashing on windows
23:09Caspy7zokum: according to AMD&#39;s site, there is a newer version of drivers - from this month actually. I would start there
23:10zokumwell, i have had this problem for ~3 years, i doubt newer drivers will help, but i can give it a go
23:15zokumis there a location where I can get decent 64bit builds?
23:15Caspy7whatever the outcome, what you&#39;ve said is a spot on description of a graphics leak leading to corruption and then a crash
23:17zokumoh, i didn&#39;t know you had 64bit binaries distributed officially, when&#39;d that start?
23:19Caspy7um, I forget, more than a year ago I know.
23:19Caspy7new installs are now being given 64 bit for qualified systems
23:19Caspy7and I think it&#39;s the next version that&#39;s supposed to upgrade current 32 bit to 64 bit for qualified systems
23:19zokumodd how I ended up with a 32bit version then on a 64 bit os when i reinstalled like half a year ago
23:20kbrosnanthe stub installer has only perfered x64 build for a month or two
23:20Caspy7zokum: if you just got the default installer it would have offered 32 bit. It&#39;s only recently that it preferred 64
23:20zokumok, that&#39;s it then
14 Sep 2017
No messages
Last message: 8 days and 5 hours ago