#archlinux32 | Logs for 2020-04-25

Back
[00:01:23] <trotz> 2020/04/25 00:01 CRIT jeti100 Archive Secure HTTP CRITICAL - Invalid HTTP response received from host on port 443: cURL returned 28 - Timeout was reached
[00:11:23] <trotz> 2020/04/25 00:11 OK jeti100 Archive Secure HTTP OK: HTTP/1.1 200 OK - 619 bytes in 0.097 second response time
[00:42:02] -!- deep42thought has quit [Quit: Leaving.]
[02:02:41] <buildmaster> dirty! girls, my database - so dirty :-(
[02:02:42] * buildmaster goes insane.
[02:04:27] <trotz> 2020/04/25 02:03 CRIT buildmaster Git HTTP CRITICAL - Invalid HTTP response received from host on port 80: cURL returned 28 - Timeout was reached
[02:24:28] <trotz> 2020/04/25 02:23 OK buildmaster Git HTTP OK: HTTP/1.1 200 OK - 9490 bytes in 0.415 second response time
[02:59:23] * buildmaster resumes sanity.
[04:52:03] <buildmaster> pentium4/traefik is broken (says nlopc46): https://archlinux32.org
[05:12:36] <trotz> 2020/04/25 05:11 CRIT buildmaster Git Secure HTTP CRITICAL - Invalid HTTP response received from host on port 443: cURL returned 28 - Timeout was reached
[05:32:36] <trotz> 2020/04/25 05:32 OK buildmaster Git Secure HTTP OK: HTTP/1.1 200 OK - 9252 bytes in 0.021 second response time
[05:39:37] <trotz> 2020/04/25 05:38 CRIT buildmaster Git HTTP CRITICAL - Invalid HTTP response received from host on port 80: cURL returned 28 - Timeout was reached
[05:56:27] <buildmaster> i686/traefik is broken (says nlopc46): https://archlinux32.org
[05:59:39] <trotz> 2020/04/25 05:58 OK buildmaster Git HTTP OK: HTTP/1.1 200 OK - 9491 bytes in 0.123 second response time
[06:51:40] <trotz> 2020/04/25 06:51 CRIT buildmaster Buildmaster Secure HTTP CRITICAL - Invalid HTTP response received from host on port 443: cURL returned 28 - Timeout was reached
[07:15:42] <trotz> 2020/04/25 07:15 OK buildmaster Buildmaster Secure HTTP OK: HTTP/2.0 200 - 2903 bytes in 5.219 second response time
[07:42:59] -!- buildmaster has quit [Remote host closed the connection]
[07:43:00] -!- trotz has quit [Read error: Connection reset by peer]
[07:44:27] -!- trotz has joined #archlinux32
[07:44:28] <trotz> 2020/04/24 17:04 CRIT buildmaster OS updates 1 updates, 0 ignored
[07:44:30] -!- buildmaster has joined #archlinux32
[07:44:30] <buildmaster> !rq buildmaster
[07:44:31] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[07:49:28] <trotz> 2020/04/25 07:48 CRIT buildmaster Git HTTP CRITICAL - Invalid HTTP response received from host on port 80: cURL returned 28 - Timeout was reached
[07:49:28] <trotz> 2020/04/25 07:48 CRIT buildmaster Git Secure HTTP CRITICAL - Invalid HTTP response received from host on port 443: cURL returned 28 - Timeout was reached
[07:54:28] <trotz> 2020/04/25 07:54 OK jeti100 Master Mirror HTTP OK: HTTP/1.1 200 OK - 1408 bytes in 0.035 second response time
[07:54:28] <trotz> 2020/04/25 07:54 OK jeti100 Master Mirror Secure HTTP OK: HTTP/1.1 200 OK - 1408 bytes in 0.085 second response time
[08:13:12] -!- bill-auger has quit [Read error: Connection reset by peer]
[08:13:51] -!- bill-auger has joined #archlinux32
[08:18:30] -!- bill-auger has quit [Read error: Connection reset by peer]
[08:19:31] <trotz> 2020/04/25 08:18 OK buildmaster Git HTTP OK: HTTP/1.1 200 OK - 9491 bytes in 0.034 second response time
[08:19:31] <trotz> 2020/04/25 08:18 OK buildmaster Git Secure HTTP OK: HTTP/1.1 200 OK - 9252 bytes in 0.034 second response time
[08:21:18] -!- bill-auger has joined #archlinux32
[08:54:31] <trotz> 2020/04/25 08:53 OK buildmaster OS updates 0 updates, 0 ignored
[09:59:02] -!- DCyrax has quit [Ping timeout: 264 seconds]
[10:15:40] -!- abaumann has joined #archlinux32
[10:15:40] <buildmaster> Hi abaumann!
[10:15:40] <buildmaster> !rq abaumann
[10:15:40] <phrik> buildmaster: <abaumann> I'm pretty convinced upstream doesn't favour the use of telnet in productive installations. :-)
[10:16:16] -!- deep42thought has joined #archlinux32
[10:16:17] <buildmaster> Hi deep42thought!
[10:16:17] <buildmaster> !rq deep42thought
[10:16:17] <phrik> buildmaster: <deep42thought> rcf you just detuned and are now receiving a different program - please execute /join #debian
[10:16:33] <abaumann> hi deep42thought
[10:16:37] <deep42thought> Hi abaumann!
[10:16:41] * abaumann plays with his OLPC XO-1, finally
[10:16:47] <deep42thought> :-)
[10:17:16] <abaumann> the thing is frustrating. I either get 'Illegal Opcode' on the screen when booting or just blank nothing
[10:17:39] <abaumann> I might have to connect this serial console, otherwise I cannot see a damn thing..
[10:17:41] <deep42thought> what opcodes does it support?
[10:17:51] <abaumann> that's exactly the tricky question.
[10:18:04] <abaumann> It claims MMX, but some adressing mode are defenitely not supported.
[10:18:06] * deep42thought always asks the tricky questions :-D
[10:18:21] <abaumann> First I thought i686 should work (by the specs), but obviously it doesn't
[10:19:29] <deep42thought> then you should reall connect the serial console - but do not make the mistake which I did: cross TXd and RXd - I fried my banana pi's serial console (and possibly my usb-serial converter's one), now I'm *really* flying blind with the banana pi :-)
[10:20:01] <abaumann> huh. good advice. :-)
[10:20:14] <abaumann> I'm missing a voltage amplifier for the serial anyway.. :-(
[10:20:47] <deep42thought> MAX232
[10:20:49] <abaumann> ah: "General Protection Exception"
[10:20:50] <abaumann> that's new
[10:22:31] <abaumann> I'm not even sure modern kernels work at all on this kind of devices.
[10:22:53] <deep42thought> what is "modern" about the kernel that should prevent it from working?
[10:23:09] <abaumann> KASLR, spectre mitigation, switches off branch prediciton.
[10:23:20] <abaumann> this here sounds like a memory protection scheme going wrong..
[10:24:03] <abaumann> with all the modern cpu bugs it would not be unreasonable to assume, that you have to switch off quite some "fixes" so that you can boot an old CPU.
[10:24:09] <abaumann> OTOH, the ALIX boots.
[10:24:17] <abaumann> and that one is quite similar to an XO-1
[11:09:37] <buildmaster> girls, my database is dirty again ...
[11:09:37] * buildmaster goes insane.
[11:11:34] <deep42thought> these were always transient network errors ...
[11:13:02] * buildmaster resumes sanity.
[12:42:55] <abaumann> hah!
[12:43:15] <abaumann> use the original config-x01 from kernel 4.6 for the newest kernel, reconfig and voila! :-)
[12:43:47] <deep42thought> summa sumarum: we should adopt our i486 kernel config?
[12:44:05] <abaumann> aeh.
[12:44:07] <abaumann> not really.
[12:44:23] <abaumann> I tried that. the problem is, it always resulted in all kind of weird kernel errors
[12:44:37] <abaumann> I feel confortable with a linux-olpc, actually, for now
[12:45:02] <deep42thought> but it only differs in the config, right?
[12:45:05] <abaumann> there is special libertas firmware and some other special stuff which makes it necessary to have specialized packages anyway.
[12:45:08] <abaumann> yep.
[12:45:18] <deep42thought> ok, so linux-olpc it is :-)
[12:45:29] <abaumann> we could make a huge diff and see what causes the thing to not work.
[12:45:38] <deep42thought> is it worth it?
[12:46:50] <abaumann> maybe not.
[12:46:55] <deep42thought> hmm, current blacklistling logic does not permit do build a package *only* for i486 :-/
[12:47:08] <deep42thought> and I'm not really inclined to change that ...
[12:47:09] <deep42thought> too
[12:47:10] <deep42thought> much
[12:47:12] <deep42thought> effort :-D
[12:47:17] <abaumann> I wouldn't change that too..
[12:47:29] <abaumann> ..there is nothing wrong with some self-compiled packages for that purpose
[12:48:10] <deep42thought> I'm not against having them as official archlinux32 packages
[12:48:24] <deep42thought> just put them in extra/linux-olpc and you're done
[12:48:46] <abaumann> ok.
[12:49:04] <abaumann> and: there will not be an ISO or USB stick or so.. too much hazzle
[12:49:20] <abaumann> everybody should get confortable with the 'architecture' pacstrap trick
[12:49:33] <abaumann> and just write it to a bootable SD card
[12:49:42] <deep42thought> what boot medium do you install from?
[12:49:48] <deep42thought> or can you take out the hdd?
[12:50:12] <abaumann> aeh: it only has internal NAND storage
[12:50:26] <abaumann> and that one I would like to keep original, I.E. Fedora OS whatever
[12:50:37] <abaumann> but it has an SD card slot :-)
[12:51:55] <abaumann> the sad story is: now that booting it is so easy, I have to do something about graphical X output..
[12:52:05] <abaumann> ..and we know what that means :-)
[12:52:19] <deep42thought> rustrustrust?
[12:52:20] <abaumann> pacstrap and netctl doesn't install dialog?
[12:52:33] <abaumann> yep and some nasty bootstrapping
[12:52:47] <deep42thought> what's dialog needed for?
[12:52:53] <abaumann> for wifi-menu
[12:53:01] <abaumann> ok, I recon, nobody uses that anymore :-)
[12:53:10] <deep42thought> I don't use wifi (much)
[12:53:16] <deep42thought> but I use netctl alot
[12:53:21] <abaumann> the OLPC XO-1 has _only_ wifi
[12:53:29] <abaumann> so..
[12:53:31] <deep42thought> "A tool to display dialog boxes from shell scripts"
[12:53:36] <deep42thought> ???
[12:53:53] <abaumann> a turbo vision like, ncurses-based menu system, yes
[12:54:00] <abaumann> drawing text-dialogs and stuff.
[12:54:02] <abaumann> menues.
[12:54:12] <abaumann> was also used for Linux installers (still in text mode)
[12:54:25] <deep42thought> debian comes to my mind
[12:54:50] <abaumann> genfstab adds a /dev/sda3 as swap
[12:55:02] <abaumann> there might be some bugs also there.. :-)
[12:55:15] <deep42thought> is that mounted as swap currently=
[12:55:16] <deep42thought> ?
[12:55:36] <abaumann> On an eeepc, please :-)
[12:55:40] <abaumann> ah.
[12:55:42] <abaumann> you could be right.
[12:55:57] <abaumann> I'm on a 'modern' X61S laptop writing to the sdcard and doing genfstab
[12:56:20] <abaumann> yeah. ok fair enough. it was designed for installation environments and not for chrooting
[12:56:52] <deep42thought> I think, the intended way is to issue `genfstab /mnt > /mnt/etc/fstab` from the outside
[12:57:11] <abaumann> the openfirmware with Forth of that thing is really cool
[12:57:22] <abaumann> sad EFI won over BIOS, both a nightmare
[12:58:23] <abaumann> "Open Firmware "freezes" the screen (by using the DCON chip to hold the last graphical image) just before starting the OS. This hides Linux's scrolling text messages to make the boot sequence prettier."
[12:58:28] <abaumann> darn it!
[12:58:31] * buildmaster goes insane.
[12:58:37] <abaumann> no need for that ;-)
[12:59:19] <abaumann> prettier... schmittier.. I want to see error.. dammit.
[13:01:20] <deep42thought> buildmaster: what's up?
[13:01:21] <buildmaster> up? I'm up for 5 hours, 17 minutes, load average: 0.63, 1.10, 1.33 ... and I'm insane :-D
[13:02:09] <abaumann> maybe the new chromium was too big to swallow? ;-)
[13:02:10] <deep42thought> btw: the new vm on archlinux32.org seems to have improved the situation: the site loads faster (for me) and I see less (no?) timeouts anymore
[13:02:21] <deep42thought> probably a network error again
[13:02:32] <deep42thought> then the buildmaster doesn't see the packages on the master mirror and halts
[13:02:49] <abaumann> the new archlinux32.org is definitely much better, I think so too,.
[13:03:09] <abaumann> -rw-r--r-- 1 http http 58525159 Apr 25 12:46 pool/chromium-81.0.4044.122-1.2-pentium4.pkg.tar.zst
[13:03:13] <abaumann> uff, I can test..
[13:03:33] <abaumann> https://bbs.archlinux32.org
[13:03:57] <abaumann> big shoutout for schnitzeltony providing a seccomp-sandbox patch for chroium
[13:04:01] <abaumann> *chromium
[13:04:07] -!- buildmaster has quit [Excess Flood]
[13:04:10] <deep42thought> oops
[13:04:18] -!- buildmaster has joined #archlinux32
[13:04:19] <buildmaster> !rq buildmaster
[13:04:19] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[13:10:03] -!- buildmaster has quit [Excess Flood]
[13:10:18] -!- buildmaster has joined #archlinux32
[13:10:19] <buildmaster> !rq buildmaster
[13:10:19] <deep42thought> *grrrr*
[13:10:20] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[13:21:59] * buildmaster resumes sanity.
[13:32:59] <buildmaster> pentium4/warzone2100 is broken (says nlopc46): https://archlinux32.org
[13:35:29] <buildmaster> i486/k9s are broken (says nlopc46-i486bs0): https://archlinux32.org
[13:37:35] <deep42thought> great, shellcheck just became more restrictive >.<
[13:37:54] <abaumann> and chromium 81.0.4044.122-1.2 works :-)
[13:38:00] <deep42thought> yes, I read :-)
[13:38:05] <abaumann> shellcheck and hledger made it to stable. it seems
[13:38:19] <deep42thought> I'm talking about x86_64's shellcheck
[13:38:23] <abaumann> ah.
[13:38:29] <deep42thought> which currently prevents me from committing ...
[13:38:58] <abaumann> git hook with shellcheck?
[13:39:02] <deep42thought> I would kick this piece of shi...shiny software out - if it had not prevented some errors in the past
[13:39:05] <deep42thought> yes, exactly
[13:41:01] <deep42thought> no, shellcheck, I do *NOT* want to quote every variable!
[13:41:10] <deep42thought> some are without quotes *intentionally*
[13:42:18] * abaumann remembers a saying about the language checking tool and the language.. :->
[13:42:29] <deep42thought> there is even a quote from you about this :-)
[13:42:39] <abaumann> there is? ;-)
[13:42:41] <abaumann> really? ;-)
[13:42:47] <abaumann> :-)
[13:42:50] <deep42thought> you're surprised?
[13:42:55] <abaumann> no really.
[13:42:58] <abaumann> *not
[13:43:18] <abaumann> "the crappier the language, the bigger the lint tool" or so :-)
[13:43:27] <deep42thought> yes
[13:44:07] <abaumann> and now: let's be really brave and try a chromium on an old Pentium III
[13:44:16] <deep42thought> :-D
[13:44:21] <deep42thought> see you tomorrow ;-)
[13:44:31] <abaumann> cu
[13:44:41] <deep42thought> you missed the irony, sry
[13:45:01] <abaumann> oh.
[13:45:03] <abaumann> indeed!
[13:45:04] <abaumann> :-)
[13:45:18] <abaumann> happens
[13:45:24] <deep42thought> I sent my family away, so I have sturmfreie Bude :-)
[13:45:37] <abaumann> ah. that's handy :-)
[13:45:38] <buildmaster> pentium4/k9s are broken (says nlopc46): https://archlinux32.org
[13:47:40] <abaumann> chromium-74.0.3729.169-2.0, last one on i686.. urgh.
[13:47:54] <abaumann> https://bugs.archlinux32.org
[13:47:56] <abaumann> yeah. ok.
[13:48:05] <phrik> Title: FS#70 : chromium not available for i686 (fails in compiler intrinsics) (at bugs.archlinux32.org)
[13:49:43] <deep42thought> I'm not really happy with these mixed-opcode packages on i686
[13:55:49] <buildmaster> i686/warzone2100 is broken (says rechenknecht): https://archlinux32.org
[14:10:42] <deep42thought> why don't we admit, that we need another architecture between i686 and pentium4?
[14:17:01] <abaumann> do we? if i486 would have more packages, we could defacto say: 486 is the one running everywhere.
[14:17:13] <abaumann> pentium4 is SSE2 for modern machines
[14:17:23] <abaumann> i686 is MMX+SSE for mildly older machines
[14:17:47] <abaumann> actually, the original architectures.php was right, just the implmenetation with march was slightly wrong. :-)
[14:20:56] <deep42thought> ok, this is a path, I'm willing to go
[14:21:20] <deep42thought> so we (you? ;-)) try to add as many packages as possible to i486 and then raise the requirements for i686?
[14:22:32] <deep42thought> which reminds me of something ...
[14:22:55] <deep42thought> ah, no, rime on i486 failed yesterday already :-)
[14:36:19] <abaumann> that's a plan, yeah. :-)
[15:33:45] <deep42thought> I reverted architectures.php
[15:34:03] <deep42thought> I think, it's not nice to draw people to i686, who will need to run i486 in the future
[15:35:05] <deep42thought> (I don't think, there are many, but still)
[15:37:09] <abaumann> yep, maybe better.
[15:37:22] <abaumann> just for now we should not chnage SSE/MMX flags yet..
[15:37:33] <abaumann> ..I have some hopes I can bootstrapp some i486 things from i686
[15:37:35] <deep42thought> actually: why?
[15:37:39] <deep42thought> ah, ok
[15:37:51] <deep42thought> you could do so from the archive
[15:37:59] <abaumann> ah.. true again.
[15:38:11] <deep42thought> but I see, that an up-to-date i686 source might be better :-)
[15:38:21] <abaumann> yep :-)
[15:38:28] <deep42thought> anyways: no hurry
[15:38:42] <abaumann> actually. the kernel compiles really fast on a Ryzen 2700 with only minimal configuration ;-)
[15:38:46] <abaumann> *:-)
[15:38:59] * abaumann wonders how you correct smilies in a chat..
[16:23:14] <T`aZ> s/;/:/ ;)
[16:27:06] <abaumann> of course. :-)
[16:33:00] <deep42thought> sed - fixing irc misspells since 1974
[16:43:35] <buildmaster> pentium4/ipv6calc is broken (says nlopc46): https://archlinux32.org
[16:51:13] <abaumann> !grab deep42thought
[16:51:13] <phrik> abaumann: Tada!
[16:57:11] <buildmaster> i486/ipv6calc is broken (says eurobuild6-7-i486): https://archlinux32.org
[17:01:57] <buildmaster> i686/ipv6calc is broken (says nlopc46): https://archlinux32.org
[18:01:47] <abaumann> yeah. i686 don't work on the xo-1, illegal opcodes everywhere.. so.
[18:11:56] -!- abaumann has quit [Quit: leaving]
[18:35:45] <buildmaster> any/bashtop is broken (says eurobuild6-5): https://archlinux32.org
[18:37:01] <buildmaster> i486/bettercap is broken (says nlopc46-i486bs0): https://archlinux32.org
[19:38:46] -!- lodger has joined #archlinux32
[19:44:15] -!- lodger has quit [Quit: Leaving]
[23:14:38] <buildmaster> i486/pari is broken (says nlopc46-i486bs0): https://archlinux32.org
[23:15:12] -!- davor has quit [Ping timeout: 256 seconds]
[23:19:22] -!- davor has joined #archlinux32
[23:56:17] <buildmaster> i686/pari is broken (says nlopc46): https://archlinux32.org