Jump to content

johk

Member
 View Profile  See their activity

  • Posts

    49

  • Joined

  • Last visited

  • Days Won

    2

  • Donations

    0.00 USD 

  • Country

    Spain

 Content Type 

Everything posted by johk

  1. johk
    I'm not going to get in to details, but I had to switch to the old resolver to test a couple of things and I also tested yt-dlp. And surprise (what else would I be here for?), the curl_cffi component of yt-dlp that was theoretically responsible for the DNS extensions request problem, look like doesn't require the DNS extensions now :-? I just wanted to post it. I don't care about the reasons (I'm not going to dig on it). Just mention it.
  2. johk
    I hope that is a fix for the regular and constant CPU spikes happening after visit sites with loads of JavaScript (example comes to mind: Twitter). After leave the site, is a constant rate of CPU spikes wherever you browse then. For now I leave the update for tomorrow or later, but I really hope it fixes it. With the summer heat, having spikes constantly rising the CPU temperature is insane, even when you are browsing plain text sites. OFF-topic, isn't the Google chatty a bit off-topic here? I don't use Google since about 10 years, preferring, by far, Bing/Yahoo (same engine), and since Bing hardcoded redirects as links to sites, duckduckgo now. Google is not a trustable search engine since long ago. I have a personal example that... I prefer not to share... but if you find my content, won't be because Google, so make a favour to yourself and use any other search engine that actually index Internet.
  3. johk
    From the engine (omni.ja>goanna.js): As always, that can only be understood by developers.
  4. johk
    I'm using the win64 version of NewMoon 28, so, I guess that won't help me. Whatever, I'm just reporting what I see, simply that I takes a lot of CPU and hangs/unresponsive when before didn't 🤷 As a shame, more than a complain. I don't know how Mozilla did with 115ESR but I really wish it is achieved the same smoothness (without hangs/locks/unresponsiveness) with Goanna/UXP or which ever is the root cause. Thanks for your work, yours, Moonchild's and whoever tries to make this alternate engine/browser alive. I could use alternatives, or buy a more powerful second hand computer, but sadly, is not possible for everyone.
  5. johk
    After these days using the latest build of NewMoon 28, I would say, that, despite I don't have BSODs, it hangs quite a lot, and by hang I mean that the window becomes unresponsive a lot. If the page has lots of javascript, prepare yourself to have an unresponsive window. Before, it just, well, got the CPU high and no more, but now... oh my... high CPU usage and totally unresponsive. In my case, might be the patch applied to win32k.sys, but one way or the other, it is crazy how often it happens compared to previous versions. As I said previously, this didn't happen on 20260411 and earlier. I don't know if I would agree what modnar said above for NewMoon 28 too, but whatever CPU threading that the engine has now is a killer at whatever level.
  6. johk
    Answered through PM to avoid off-topic.
  7. johk
    So, in the end, I was right with my early findings about the CPU usage even Moonchild had to confirm I just really hope that those thread parsing corrections help to make NewMoon/Goanna engine smoother, on CPU I mean. The actual smooth would be accomplished on the javascript engine (as Firefox 115), someday... ------- FOLLOWS, AFTER ALL OFF-TOPIC ----- Sorry if I haven't replied these days, I've been testing RAM (that you know it takes loooooooot of time) with thunderstorms in the middle that didn't allow me to even plug the PC and now I'm fixing problems on a family member PC. And the RAM and processor are fine. In the meantime I rechecked all my previous debuglogs (made with debug wizard help) and, even the "probably caused by: win32k.sys" is very... (as any other with microsoft) imprecise, and I never considered it to be the real problem, I searched and applied the latest(?) win32k.sys update named KB3095649 which, in theory, might help, but as told, I haven't had time to test yet, because I was able to reproduce intentionally the BSOD, so I guess I could test. Answering to other mentions, the disk is not the problem as, long ago, decided to run the profile on a ramdisk (made with ImDisk) thanks to the decision long ago form Mozilla, and so MoonChild, and so Gecko and Goanna engine, to use an sqlite database for the history. You can't believe the performance hit that is managing those databases on disk, really. It is a continous read/write. And Firefox, using localStorage on sqlite databases for browser cache... you couldn't believe the performance difference of running it on a ramdisk. Anyway, I'm digressing right? I will test if NewMoon now crashes the system (not because of NewMoon but because I found a good testing vector), but that is something on my side. EDIT 2026-05-01T03:30+2: for what I tested before go to sleep, I couldn't reproduce the BSOD, so, fingers crossed, let's hope I don't have new BSODs with NewMoon as vector for win32k.sys BSOD, nor any other program. Or maybe today wasn't the same environment and didn't crash. The upcoming days will tell. And I hope the newer NewMoon builds, without the Goanna engine CPU hog, would be helpful as well.
  8. johk
    Ok, you are the programmer :) But it was very coincidental that at the same time NewMoon freeze, I went to try to close it and BSOD :S Anyway, I won't use the 20260425 build as, for whatever reason, gives me problems.
  9. johk
    And you know what they say, that if you play with fire your end burning yourself. Well I got a BSOD (I didn't have one since a couple of years) when testing the 20260425 build in safe mode. I send the dmp minidump privately to you @roytam1 as I think it might have some personal information. You don't allow PMs, OK, if you want the dmp tell me and I'll send the link though K-meleon forum PMs. I have to go to sleep now. See you.
  10. johk
    And it works on the last week build 20260418 too @VistaLover Well, for the sake of testing while I have it on a ramdisk, I didn't notice that palefill stopped working, so I did the fix mentioned, changing strict compatibility to false on install.rdf, and while now can be installed and, I guess, works, on the latest build, 20260425 it still hangs/freezes, so the palefill availability is not the culprit of the hangs/freezes. Why doesn't happen to you @roytam1? No clue. Is just visit github.com/notifications and after load, freeze (unresponsive window to any interaction, except close). I don't have the means to debug it.
  11. johk
    I'm not under XP, I use NewMoon x64 under Windows 7. One way, or the other, good or bad job (as I'm not blaming roytam at all, but reporting), the thing is that now NewMoon freezes more, after the 20260411 build. Even if 20260418 doesn't freeze that much, it just did on this page. On 20260411 and earlier, didn't. It is true that GitHub is very intensive, but until 20260425, and particularly about February or March builds, GitHub rendering on NewMoon has enhanced a lot, as there were some glitches with some site functions that drive you to empty pages, as for example mark threads as read/done, and now (until 20260418 because 20260425 just freezes) works better. There shouldn't be a difference if it runs on XP or Win7. In my opinion.
  12. johk
    Updated from 20260411 to 202650425 of NewMoon 28 64 bits resulted in high and long CPU usage and now it hangs on loading github notifications. Back to 20260411 and all is fine. The latest build is, by far, slower on my Core2Duo CPU. If I have to bet, might be the following update: "- Issue #3045 - backport of bug 1350760 - atomization fast-path improvements (b44216a883)" When I read atomization.... I thought... this is going to give problems... I haven't tested that intermediate build, as it was introduced on 20260418, because I can't download it at this rate (barely 10KiB/s) This is the reason I don't like to update, if I don't need to :/ (un)expected consequences. If it works, don't do anything. EDIT 20260425T20:50+2: Damn it, after I wrote the slow download it downloaded a minute later. With 20260418 doesn't crash on github.com/notifications and it is smoother. Changed back to 20260425 and it hangs and I have to force close
  13. johk
    Don't worry, as I knew this was going to happen soon or later, I had to "resign" and I spent time on testing another version and alternate software. But thanks to give this extra time to read, test, etc. So far so good, tested this new yt-dlp build, and works. But it is a shame because, as every software nowadays (despite the version I'm using now is still old, but isn't less true that the newer, in regard to the OPT DNS extension, hasn't evolved since) it lacks features the original software had and quite a few more issues. Newer doesn't mean better and almost all at MSFN, or these threads, more or less, agreed with this statement, and that is the sad story :/ EDIT: still, I don't find the need for this tool to require it. It doesn't add anything special, less as the behavior described earlier. Nonsense.
  14. johk
    @Mark-XP yes,that is the message; that page is actually for developers, that is the reason you can't see anywhere. It was created by Cloudflare after the block to Palemoon and any based browser and was published on Palemoon forums so people could test. I use myself to test if my browsers work. Maintain a list of sites using Cloudflare's trunstile would be endless, just only in my country. The only one able to rise this to CloudFlare is roytam, registering as a developer and signing some undisclosing contract. I don't remember the exact procedure told a year ago in the Palemoon forums.
  15. johk
    Yes, that is almost my thought too, but maybe not the computer, but the JavaScript engine on NewMoon. Unless I have some option missing to enhance its performance. Because I'm on a Core 2 Duo (we would agree it is slow and old), but with Firefox 115, that runs JavaScript waaaaayyyy smoother, as the tests are completed in a split second, not as NewMoon, that struggles, instantly shows the test is OK at https://browser-compat.turnstile.workers.dev/
  16. johk
    Just suffered the Cloudflare issue. I didn't have problems along the week til now, using the 20th of March version of NewMoon 28. But if you have problems too, is not the version. "All fine, but we block you because... we can."
  17. johk
    yes, clear as water, sorry. As I go the default way, I was really missing something
  18. johk
    This won't make you happy, but in my case is true that if I don't use quickjs, which I usually don't, the m3u8 formats are not listed, what I don't care much anyway, but that is what happens to me. There is a difference in the video information retrieval, when not using the JS runtime, it downloads only the adroind JSON, but when using the JS runtime downloads the safari player JSON, the player Javascript and then downloads the m3u8. But I don't know why this happens to me and not to you. [youtube] kS-9ISzMhBM: Downloading android sdkless player API JSON [youtube] kS-9ISzMhBM: Downloading web safari player API JSON [youtube] kS-9ISzMhBM: Downloading player c1c87fb0-main [youtube] [jsc:quickjs] Solving JS challenges using quickjs [youtube] kS-9ISzMhBM: Downloading m3u8 information
  19. johk
    http://kmeleonbrowser.org/forum/read.php?19,154431,158596,page=35#msg-158595 Just in case you need something to add.
  20. johk
    It works now. But let me say one thing. While it is legit to complain as I did, and also helps to know what is the issue and see if I can fix, or can be fixed, if it is necessary to use the latest curl_cffi, just do it. Don't downgrade because a person complains. If I can use the x86, awesome, I'll use. In cases as with this tool, how many times are we dealing with over 2³² of data/memory needs to use the x64. It is more a matter of slightly better performance. off-topic from here Now, about the "modern DNS extensions". Actually they aren't modern... I mean, the original proposed standard RFC is from 1999, and then ratified as internet standard on 2013. It is fair to use it, but, why now? If I could I would update the hardware, OS, software, but... I don't live in a world of magic.
  21. johk
    That's it!!! Oh lord , what a waste of day doing tests back and forth, searching for alternatives, finding BIND9 stopped support for Windows on 9.17.14.... Anything, but not with the x86 build... I'll end with a correction to myself. I should have said that EDNS is not for security only. I always mix it with DNSSEC (which actually is a feature of the upstream curl utility), despite this requires EDNS, too.But, anyway, there is no need for such implementation, IMHO.
  22. johk
    It is actually the curl stage, as it is what it fails when can't do an OPT DNS query type 41 hex (65 is the decimal conversion)). No, my DNS server/resolver software doesn't support it. I should have told, but I got tired of test, renames, edits... oh dear... I know the software is old, but trusted, but, also, there shouldn't be need to hardcode EDNS queries on curl. Too paranoid, in my opinion. I tested with the "--force-ipv4" and it is the same. If the curl update is really necessary... I'll have to change the software, but it is really dumb reason :/ It works if I redirect queries to an external resolver (example: 1.1.1.1)... but defeats the purpose of using a local one in first place Do you know what is really curious? That I almost always perform formats queries first, and it does this without a problem (querying correctly an A record for www.youtube.com), but when queries it again (or direct download, without query formats first) for download... In other words, it happens at download time In the version of curl I have I don't have such an EDNS option to test.
  23. johk
    As I don't think that it is fair to report this to yt-dlp, as I'm using your builds under Windows 7 (and the officials aren't valid for this), and so, just in case, the fault is in this build, I want to tell that your latest, this is 2026.01.19.143412, but not the previous, this is 2026.01.18.152406, there is a problem with DNS resolving. Instead request the OS to resolve an A DNS record (this is the IPv4 IP) for the host of the video www.youtube.com (after re-review is not the host of the video, it doesn't even reach that step, is www.youtube.com), requests a ?65? DNS record (question mark, 6, 5, question mark), that is absolutely invalid (there is no such record in the standard and so replies the Google DNS server). I use my own DNS resolver (local resolver, I mean, not external) and so I can inspect what is going on with DNS requests and so I'm seen it and reporting it. Where is the fault coming? I don't have the knowledge to tell. Just that in the 18th of January, it works OK. Regards. EDIT: just to add up, I double checked the hash and file download isn't corrupted. EDIT2: now that I think about it... 65 matches to the decimal value of the character A (capital A)... Might be some sort of UTF or ASCII thing? DNS is pure ASCII (or maybe ANSI), so no clue what has been introduced this issue. LAST EDIT: ok I wasted the time digging it, using wireshark in loopback mode, and 65 means code 41, which means EDNS. Why on earth yt-dlp is now requesting an EDNS record for www.youtube.com :-?
  24. johk
    From my experience, Japanese content is highly suitable to be geo-fenced. They protect a lot of things from overseas access.
  25. johk
    As of... about a month, or two, can't tell exact, NewMoon 28 is unable to login to AOL nor Yahoo (same parent company, same back/front systems). After input the password, nothing but a spinning circle on the submit button. Latest version/build, the same issue persists.
×
×
  • Create New...