December 2015 - Arch-general - lists.archlinux.org

[arch-general] ssd trim using fstrim.service and fstrim.timer
by Francis Gerund 29 Dec '15

29 Dec '15

hello. Per the Arch wiki SSD page, I just enabled sysctl fstrim.timer, and then rebooted. I did not "enable" fstab.service. Now fstrim.timer is loaded, and active (but "waiting") and fstrim.service is loaded, but inactive. And the time stamp file the wiki mentions has a "0" size. So, do I have to wait (A WEEK!) to see if it works, or can I somehow now run fstrim.service manually to at least get it done once? Note: I could just add "discard" to /etc/fstab, but wouldn't that wear out the SSD faster than periodic trimming? And yes, my SSD does support trimming.

5 6

[arch-general] Performance with Linux Bridge
by Carlos Ferreira 29 Dec '15

29 Dec '15

Hello to all I'm using Arch Linux in a SuperMicro A1SAi-2750F motherboard and I'm having performance issues while bridging 3 Ethernet Gigabit network interfaces. The 4th network interface of that motherboard, is used to connect to the ISP. This setup is my Router/server and I use the 3 bridged interfaces to connect my home network. But now, I'm noticing severe performance issues. I conducted a small iperf3 test. The scenario was: PC A (100Mbps) <---> Server Ether 1(1Gbps) <---> Linux Bridge (br0) <---> Server Ether 2 (1Gbps) <---> PC B (1Gbps) The result: Connecting to host 192.168.123.136, port 5201 [ 4] local 192.168.123.4 port 55462 connected to 192.168.123.136 port 5201 [ ID] Interval Transfer Bandwidth Retr Cwnd [ 4] 0.00-1.00 sec 1.47 MBytes 12.3 Mbits/sec 56 9.90 KBytes [ 4] 1.00-2.00 sec 1.16 MBytes 9.70 Mbits/sec 50 8.48 KBytes [ 4] 2.00-3.00 sec 1.10 MBytes 9.25 Mbits/sec 30 12.7 KBytes [ 4] 3.00-4.00 sec 1.14 MBytes 9.52 Mbits/sec 40 5.66 KBytes [ 4] 4.00-5.00 sec 1.04 MBytes 8.76 Mbits/sec 37 8.48 KBytes [ 4] 5.00-6.00 sec 1.19 MBytes 9.96 Mbits/sec 40 12.7 KBytes [ 4] 6.00-7.00 sec 1.14 MBytes 9.55 Mbits/sec 46 7.07 KBytes [ 4] 7.00-8.00 sec 1.02 MBytes 8.55 Mbits/sec 39 14.1 KBytes [ 4] 8.00-9.00 sec 1.14 MBytes 9.52 Mbits/sec 47 12.7 KBytes [ 4] 9.00-10.00 sec 1.04 MBytes 8.72 Mbits/sec 35 11.3 KBytes - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 11.4 MBytes 9.58 Mbits/sec 420 sender [ 4] 0.00-10.00 sec 11.4 MBytes 9.57 Mbits/sec receiver Then i conducted another iperf3 test with a different scenario: PC A (100Mbps) <---> Server Ether 1 (1Gbps) <---> Linux Bridge (br0) The result: Connecting to host 192.168.123.1, port 5201 [ 4] local 192.168.123.4 port 58982 connected to 192.168.123.1 port 5201 [ ID] Interval Transfer Bandwidth Retr Cwnd [ 4] 0.00-1.00 sec 11.3 MBytes 94.5 Mbits/sec 0 24.0 KBytes [ 4] 1.00-2.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 2.00-3.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 3.00-4.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 4.00-5.00 sec 11.2 MBytes 94.2 Mbits/sec 0 24.0 KBytes [ 4] 5.00-6.00 sec 11.2 MBytes 94.3 Mbits/sec 0 24.0 KBytes [ 4] 6.00-7.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 7.00-8.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 8.00-9.00 sec 11.2 MBytes 94.1 Mbits/sec 0 24.0 KBytes [ 4] 9.00-10.00 sec 11.2 MBytes 94.2 Mbits/sec 0 24.0 KBytes - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 112 MBytes 94.2 Mbits/sec 0 sender [ 4] 0.00-10.00 sec 112 MBytes 94.1 Mbits/sec receiver iperf Done. Does anyone know if this low performance behaviour by a Linux bridge is "normal" or, if it is a configuration problem? I disabled iptables. Despite gaining a little improvement, I still have severe performance issues. Can anyone provide some light in this case? Thank you! Carlos Miguel Ferreira

2 2

Re: [arch-general] [OT?] Which is most future-proof desktop environment?
by Ralf Mardorf 28 Dec '15

28 Dec '15

On Mon, 28 Dec 2015 00:52:21 +0100, Ivan wrote: >On Sun, 27 Dec 2015 18:43:09 -0500 >Francis Gerund <ranrund(a)gmail.com> wrote: > >> Just a call for opinions: if you use Arch, and you wanted to choose >> and stay with a desktop environment long-term, what would you choose >> - and why? > >Xfce wins for me by far... It's simple, light and has many features ... but within a major release themes got broken, xfce4-terminal doesn't re-wrapp lines, if you resize the window. Thunar, the file manager doesn't allow to split the window and requires Gnome's virtual crap to provide mount functionality and so on and so forth. I recommend to take a look at openbox (jwm is also not bad) and to use it with roxterm, lxpanel or fbpanel, and spacefm, IOW instead of using a desktop environment, just use a window manager. 2 Cents, Ralf

2 3

[arch-general] virt-manager empty package?
by Damjan Georgievski 27 Dec '15

27 Dec '15

Does anyone know why virt-manager is an empty package? The package depends on virt-install which contains the virt-manager program, and just adds its dependencies. Is this intentional? The PKGBUILD suggests it tries to do something. -- damjan

5 6

Re: [arch-general] [arch-dev-public] Signoff report for [testing]
by Jens Adam 23 Dec '15

23 Dec '15

> * linux-4.3.3-1 (i686) > 0/1 signoffs Can we get 4.3.x into [core] before two months are over (or before 4.4 is out)? Consider this an user signoff; Arch [testing], default kernel (plus tp_smapi) on a Thinkpad T60p (i686-only) with xf86-video-ati and Xfce. --byte

2 1

[arch-general] Note: /usr/share/pear/test/Console_Getopt/tests/... conflicts on pear upgrade.
by David C. Rankin 21 Dec '15

21 Dec '15

Just a note: Not sure whether it is worth a bug or install note, but when upgrading pear the following conflicting files are noted: ... error: failed to commit transaction (conflicting files) php-pear: /usr/share/pear/test/Console_Getopt/tests/001-getopt.phpt exists in filesystem php-pear: /usr/share/pear/test/Console_Getopt/tests/bug10557.phpt exists in filesystem php-pear: /usr/share/pear/test/Console_Getopt/tests/bug11068.phpt exists in filesystem php-pear: /usr/share/pear/test/Console_Getopt/tests/bug13140.phpt exists in filesystem Errors occurred, no packages were upgraded. Just thought I would pass it along. Let me know if you think it needs a report and I'll file if so. -- David C. Rankin, J.D.,P.E.

3 5

[arch-general] Samba Clustering Support
by Carsten Feuls 20 Dec '15

20 Dec '15

Is it possible to compile samba with clustering support in arch. I've tested it with the PKGBUILD for samba. And I only need to add --with-cluster-support in the ./configure part. And the config files need to be moved to /etc. Whould it make sense for arch to activate it? I could not find any regressions with it! Sincerely yours Carsten Feuls

2 1

Re: [arch-general] Community and TU questions
by Ludwig Zins 19 Dec '15

19 Dec '15

On 09/12/15 um 18:00, Ivan wrote: > On Wed, 9 Dec 2015 17:46:58 +0100 > Sebastiaan Lokhorst <sebastiaanlokhorst(a)gmail.com> wrote: > > > You can easily search the bug tracker[1] yourself, and help out with > > bugs in packages you are familiar with. > > Maintaining packages is usually not very time-intensive for the > > maintainers. So I think that there is little to offer there. > > Besides that, you can have a look at this page[2] for more things to > > do. > > > > Basically, you don't have to be a TU to help out on most tasks. > > > > Good luck and thanks in advance, > > Sebastiaan > > > > [1] https://bugs.archlinux.org/ > > [2] https://wiki.archlinux.org/index.php/Getting_involved > > Thank you for your answers. This will do good :) > I hope I will too haha > > Cheers! I encountered the same bug! If you downgrade wpa_supplicant to 2.3.1 evrything should work again. --

1 0

Re: [arch-general] [arch-dev-public] [RFC] Filesystem package update
by Damjan Georgievski 18 Dec '15

18 Dec '15

> > I assume this is a copy paste error... Seems like you're suggesting we > use > > resolved by default? > Yes it is, the new hosts line will be: > hosts: files resolve mymachines myhostname > > Yes too, I'm suggesting to use nss-resolve by default, as it switch to > nss-dns > for configuration with systemd-resolved disabled. > I'd advice against that at this time, nss-resolve has some strange issues now, and also the fallback is not ideal breaks some apps: https://github.com/systemd/systemd/issues/1965 and doesn't fallback when dbus is not available https://github.com/systemd/systemd/issues/1692

1 0

[arch-general] gnupg, dirmngr and sockets
by arnaud gaboury 18 Dec '15

18 Dec '15

As I was building an AUR package[0], I had an issue with validating the key. I run: $ gpg --rec-keys PkgKey only to get an error about dirmngr not being able to connect to ~/.config/gnupg/S.dirmngr $GNUPGHOME is set in my env. After hours of reading similar issues, but usually running such command as root, I couldn't find any clear answer. I tried to understand how dirmngr worked, the sockets, and the whole process. I got the answer running on one shell $ dirmngr --debug-level expert --server and the above command in another one. The culprit was in fact my gnupd home dire files set as mod 600. Settings the files as 700 solved the issue. Now I am wondering a few things: - man dirmngr says running in daemon mode is depreciated. But grepping the process, I see: 289:root 27509 1 0 09:12 ? 00:00:01 dirmngr --daemon --homedir /etc/pacman.d/gnupg What starts this daemon (a previous gpg command ran as root?)? How to change the way it is run from daemon to server mode? - the sockets directory (root one in /root/.gnupg/ and user one in ~./config/gnupg/) do not seem to me the best place. Something under /run would be more appropriate. I can not find any doc on how to set the socket in another place. Same can be said for socket S.pgp-agent. In fact, for root, this socket is in two places: /etc/pacman.d/gnupg and /root/.gnupg. I am not sure having /root/.gnupg is a good option. Is there any way to change all these behaviors via any config files ? If no, I am thinking of writing systemd service + socket files. Thank you for any tips and thoughts. [0]https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=lib32-nettle4 -- google.com/+arnaudgabourygabx

2 3