- Arch-general - lists.archlinux.org

[arch-general] Problem with packaging pyxdg and xdg from PyPI
by Ricardo Band 05 Apr '20

05 Apr '20

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Ahoi, I maintain a tool called virtualfish [1] on the AUR. It is writte in python and one dependency of it is called xdg [2]. It is not packaged in archlinux so I would just do that on the AUR. Problem is there already is a package in archlinux extra which is called python-xdg [3]. That package includes a python library called pyxdg [4] which does nearly the same as xdg but is still a different library. In my opinion the arch packages python-xdg and python2-xdg have an incorrect name. They should be called python-pyxdg so the xdg module could be packaged under python-xdg. The naming schema with the python-py prefix looks odd but isn't new. It is already used in the pyserial [5] package. How do we solve this problem? [1]: https://pypi.org/project/virtualfish/ [2]: https://pypi.org/project/xdg/ [3]: https://www.archlinux.org/packages/extra/any/python-xdg/ [4]: https://pypi.org/project/pyxdg/ [5]: https://www.archlinux.org/packages/community/any/python-pyserial/ - -- Greetings Ricardo Band https:// www.ricardo.band mailto:// email(a)ricardo.band xmpp://jabber@ricardo.band -----BEGIN PGP SIGNATURE----- iQIzBAEBCgAdFiEEhyzgFNiV8yZuI1T5Gp1lfQa1gg4FAl6JGRgACgkQGp1lfQa1 gg76dQ/8D+4Fo3AiKoPw+CFcQsATGs7Ry+W5l8awSHzEUjrI7lcMJCFEXS2CjzxO +LBzSpmP1p36LI/Ec/TAAc60nl3i3tmY8SZY4lPhQJDWHSo8F3az3aMCsy4TfBCa YQRIZPzx3iMRKYEiJx8GVHiSW/GebebT4ilp4ssQjgkIrq28HT5G0rK7N7O6/Yp0 gZHo4TlgqpOVZ8HgXfX2lVoO4O1/7isyIKtL5YMCMwbGxsFBaYxkwbahcd1FTd7A AIC6XP3wGNZbpPEtZQGDGFd1soJ0abh1bxUsvwlgPswoEkho/HKq9L3WhbsqFS4x wLt07/7ucpRXZ+16ajmDoLyjjhvLjUfSlO4sXi5sAwrIUYS/rbnB4L7qoUfGkyRs KJ6e07iW+T7ZQRfNpodTs+N/RBDxUzNk5Pj5F5e7VgCWGEsUmePMG4TO5+06skIH OH9EXrATYLUjL31077rO1WzxSKShcoZ17Z0lW8IKYut4RDG1EgUMuIH/hyydKhWj 2VOVMjDY5PxdUtlvphSXHpdpbzGD+9n7TsAEFE3opQigayMbjBnprsUmk/ubX0Og pHXwAs45K5g3Kqa1qY7tow8Rw5a528Tsc3+FZGAEvNj7tRlynrAS370oFnKys9Wc gING0Q/UB2siV1LJRuExBNXLP63yezCqVqdp7LvkvodRdwL06Zg= =4gAQ -----END PGP SIGNATURE-----

3 4

Re: [arch-general] Multi-threaded mkinitpcio
by Amin Vakil 03 Apr '20

03 Apr '20

> > As it's true for pretty much all update process for pretty much every OS > in the world, things can break if the update is interrupted. > I meant as other linux distros keep multiple kernel versions, if something happens during kernel update you can always boot to another kernel which I've done on CentOS servers before when update has broken during kernel update. > > For my particular setup, running `mkinitcpio -P` yields those times: > ------------------------ > real 1m8.320s > user 0m24.742s > sys 0m12.806s > ------------------------ > Half of the time mkinitcpio was waiting for disk I/O. Most likely > `find`, but that?s only mu guess. > As my Arch is on SSD I suppose hard drive is not a bottleneck. > > Even while it's unlikely that only mkinitcpio suffers from a damaged > HDD and even while smartctl not necessarily does show issues of a > already damaged aged HDD, I would run smartctl and take a look, if it > shows something fishy. > Good point, I've checked and my SSD is OK. > > change the COMPRESSION variable to lz4 ou lzop in your /etc/mkinitcpio.conf > : it will considerably reduce the compression time. > Great suggestion, I've compared with gzip and it's really considerably faster. (Four images are built, linux (default, fallback), linux-hardened (default, fallback)) gzip: real 1m42.140s user 1m22.717s sys 0m23.493s lz4: real 1m1.120s user 0m44.310s sys 0m21.785s cat (without compression): real 0m59.477s user 0m43.050s sys 0m21.946s To sum it up I will enable lz4 compression to benefit from both reduced time and saving space. Also I'm planning to read wiki on how to safely disable fallback. I'm using LVM on LUKS method and I suppose fallback image cannot boot my Arch, so it doesn't help in case of an accident. Thank you all

3 3

[arch-general] Problems with files creation / copy with Samba 4.12.x and SMB1 only device.
by fredbezies 03 Apr '20

03 Apr '20

Hello. Before all, thanks to David C. Radkin for the tweak related to SMB1 activation. After installing Samba 4.12.x, I added only this in [global] section of /etc/samba/smb.conf to get access to one of my old SMB1 only network based HDD: server min protocol = CORE client min protocol = CORE After a reboot, I got access to my share, but every single time I want to copy / create a new file, a folder is created! Of course, I cannot delete this folder. Here is an error log I had in /var/log/everything.log: Mar 31 23:28:40 fredo-arch-mate gvfsd[1096]: Kerberos auth with 'fred@WORKGROUP' (WORKGROUP\fred) to access '192.168.1.254' not possible Any idea? Thanks! -- Frederic Bezies fredbezies(a)gmail.com

2 1

[arch-general] TexLive 2020 packages
by Rémy Oudompheng 02 Apr '20

02 Apr '20

Hello, TexLive 2020 is frozen upstream and will be officially released in the coming days. Packages have been uploaded to [testing] and [community-testing] (for biber). Please let me know if you encounter issues with these. Regards, Rémy.

1 0

[arch-general] nvidia-dkms pkgrel increase after every linux upgrade
by Amin Vakil 31 Mar '20

31 Mar '20

I have recently installed nvidia-dkms as I use linux-hardened, although I have installed linux package as well, but I always boot to linux-hardened. Anyway I've noticed that nvidia-dkms increases pkgrel everytime linux package gets upgraded, why is that so? Taken from NVIDIA wiki: Custom kernel If you are using a custom kernel, compilation of the Nvidia kernel modules can be automated with DKMS. Install the nvidia-dkms package (or a specific branch). The Nvidia module will be rebuilt after every Nvidia or kernel update thanks to the DKMS pacman hook. Correct me if I'm wrong but it doesn't mean that nvidia module will be rebuilt after kernel update thanks to the DKMS pacman hook? So why should nvidia-dkms upgrades?

4 4

[arch-general] Apache Virtual Host Connection Refused Error
by Matt Schwartz 31 Mar '20

31 Mar '20

Hello list, I have a vexing problem with an Apache Virtual Host setup. In a nutshell, I have verified that apache is listening on port 80 and I have double checked my virtual host setup. Indeed, I can reach by IP address and the connection is logged under the virtual host. However, when I try to use the domain name, I get a connection refused error. The httpd.conf file is pretty much stock with the exception of uncommenting the userdir stuff and the vhosts line at the bottom. I pretty much followed the article on the wiki. I am leaning towards a DNS issue but my domain properly resolves to the correct IP address. The IP address is 149.28.50.65. What could I have missed? # /etc/httpd/conf/extra/httpd-vhosts.conf <VirtualHost *:80> ServerAdmin webmaster(a)goblackcat.com.com DocumentRoot "/home/user/public_html" ServerName "goblackcat.com" ServerAlias "www.goblackcat.com" ErrorLog "/var/log/httpd/goblackcat.com-error_log" CustomLog "/var/log/httpd/goblackcat.com-access_log" common <Directory "/home/user/public_html"> Require all granted </Directory> </VirtualHost> #/etc/hosts 127.0.0.1 localhost ::1 localhost Thanks! Matt

4 6

Re: [arch-general] Samba is broken on stable repositories since March 28th, 2020.
by fredbezies 30 Mar '20

30 Mar '20

>On 03/29/2020 11:39 AM, fredbezies via arch-general wrote: >> Hello. >> >> Last update of samba, pushed on extra - instead of testing - is breaking >> share access. >> >> Commit: >> https://lists.archlinux.org/pipermail/arch-commits/2020-March/727612.html >> >> See this thread: https://bbs.archlinux.org/viewtopic.php?id=254083 >> >> Only fix for now? Downgrading samba. >> >This is one of those rare occurrences when an update breaks critical server >functionality for some older environments. Samba 4.11 disables SMB1 by >default for the first time. I have old clients in backoffice roles that still use it. I don't know if my shares are using SMB1 protocol, but I got python errors like in this bug: https://bugs.archlinux.org/task/64537 Only real fix is to upgrade to samba 4.12.x. > >Edit /etc/samba/smb.conf and add > >client min protocol = NT1 >server min protocol = NT1 > >to the global section to restore workgroups and SMB1 No luck here: $ ls /etc/samba/ 14:03:08 private There is no /etc/smb.conf file. Archlinux needs to push Samba 4.12.x. Nothing less, nothing more. Pushing a 4 months old development version is weird. -- Frederic Bezies fredbezies(a)gmail.com

1 0

[arch-general] Brightness keys not working on laptop
by David Rosenstrauch 30 Mar '20

30 Mar '20

I can't seem to get the screen brightness up and down keys working on my laptop. (A new HP EliteBook x360 1030 G4.) Controlling the screen brightness programmatically does work. I.e., I'm able to set up mappings in my DE (XFCE) for specific key combos to call out to the "light" application and increase/decrease brightness that way. But the built-in brightness function keys (F3 and F4, without pressing fn) don't work at all. xev doesn't register any keypress when I hit those keys, and evtest always reports an unknown/wrong scan and key code: Event: time 1583360957.168127, -------------- SYN_REPORT ------------ Event: time 1583360959.602902, type 4 (EV_MSC), code 4 (MSC_SCAN), value 81 Event: time 1583360959.602902, type 1 (EV_KEY), code 465 (KEY_FN_ESC), value 1 I've been trying to debug this for several weeks now, but without success. I'm also pretty sure that I saw these keys working about 2 weeks ago. But although I've tried rolling back to versions of some key packages (e.g., kernel, libinput, and systemd) from that date range, I've not been able to get them working again. I've searched for bugs and forum posts on the issue, but haven't found anything that fixes it. I've also tried mucking with some of the suggestions at https://wiki.archlinux.org/index.php/Map_scancodes_to_keycodes but no joy there either. I'd file a bug on this, but I have no idea which package is causing the issue. I'm fairly certain these hotkeys are controlled by the kernel hp_wmi module, and I do see some error message lines in dmesg about that: [ 5.683232] hp_wmi: query 0x4 returned error 0x5 [ 5.697112] hp_wmi: query 0x4 returned error 0x5 [ 5.704773] hp_wmi: query 0xd returned error 0x5 [ 5.730490] hp_wmi: query 0x1b returned error 0x5 But I'm not certain that those error messages are actually related to my issue, and googling on them didn't turn up anything of much relevance. Anyone have any suggestions/pointers on what might be broken here and/or how to fix? Thanks, DR

2 2

[arch-general] Samba is broken on stable repositories since March 28th, 2020.
by fredbezies 29 Mar '20

29 Mar '20

Hello. Last update of samba, pushed on extra - instead of testing - is breaking share access. Commit: https://lists.archlinux.org/pipermail/arch-commits/2020-March/727612.html See this thread: https://bbs.archlinux.org/viewtopic.php?id=254083 Only fix for now? Downgrading samba. -- Frederic Bezies fredbezies(a)gmail.com

2 1

[arch-general] latest kernel update surprise
by Jude DaShiell 26 Mar '20

26 Mar '20

5.59-10 on the machine I use. I'm using a different version of linux on another disk to write this message. Strangely, both speaker-test and espeakup no longer work. The speaker-test failure would of course cover espeakup since espeakup uses sound card resources to do screen reading. Was anything done to the kernel to cause these failures? --

9 16