You are not logged in.
EDIT: I meant to post this in Applications & Desktop Environments, but don't see how to delete / move it myself.
I found an existing topic about this that was closed because the OP wasn't using a plain arch system: https://bbs.archlinux.org/viewtopic.php?pid=1702631
In addition to nautilus, epiphany and gnome-character-map, I can also reproduce the problem with almost all gnome programs (gedit, builder, gnome-weather, ...) and a few non-gnome gtk3 programs (transmission-gtk, transmageddon); other gtk3 applications such as Quod Libet and Glade work fine though.
Seen as the first reply there asked for the output of hostnamectl status configuration, here it is for my setup:
Static hostname: jp-tuxedo
Icon name: computer-laptop
Chassis: laptop
Machine ID: d1500bbead4b4ccea55787846f4e9391
Boot ID: 2b4785d29dd54ade896dbbba7b9d2961
Operating System: Arch Linux
Kernel: Linux 4.10.8-1-ARCH
Architecture: x86-64I've also used strace -c to check if some syscall is blocking, but that doesn't seem to be the problem as the total time spent in syscalls is never more than a fraction of a second.
Last edited by jonessen (2017-04-18 10:52:42)
Offline
Mod note: Moving to Apps & DEs.
For future reference, you can use the report button to request topic moves.
Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD
Making lemonade from lemons since 2015.
Offline
I have/had the same problem some times now when updating the gnome major version. I do not exactly know where the problem resides. But starting with a fresh config always helps. As for now I've been too lazy to search around which exactly part in the config is causing the trouble.
Offline
What do you mean by "starting with a fresh config"? Deleting your .config?!
Meanwhile, I've found a very weird workaroud: Start gnome-shell on a separate tty, and launch an application from there. The first time I did it it still took quite a while, but from the point where I launched a second program in the gnome shell on the problem was gone until reboot. Why I consider this very weird is that I could even kill the gnome shell and the issue still didn't come back.
Also, it seems like the "assertion 'G_IS_DBUS_PROXY (proxy)' failed" in the terminal output is significant, since it disappeared with the main issue. (I did not assume that initially because I feel like I always get a bunch of warnings and errors in the terminal output of gtk programs whenever I look at it)
Last edited by jonessen (2017-04-14 09:08:03)
Offline
Things like this might be network related, especially when the loopback device is somehow misconfigured. You must be able to ping localhost and 127.0.0.1 as well as your ip address and hostname (see the output of hostname) in all circumstances (wether the internet is connected or not). If you don't use it disable ipv6 (pass ipv6.disable=1 as a kernel option). So check /etc/hosts and the the output of the above ping commands.
Offline
Huh, well I guess I found something out of the ordinary, at least on my laptop (I have this problem on both my laptop and my desktop). And of course it's completely my own fault :S
So not so long ago, I read on some blog that 127.0.0.1 isn't actually that special and you can use everything in 127.0.0.0/8 if you want. Naturally I played around with that, removing 127.0.0.1 from lo with ip link, and adding 127.0.0.2... That configuration was obviously not meant to be persistent, and when I check ip addr now, it shows me 127.0.0.1 and ::1 for lo. However... When I am not connected to a router (so the hostname is not looked up via DNS, which seems to be the case otherwise), ping -4 connects to 127.0.0.2 for some reason. I still get a reply just as fast as when using localhost, but I guess this might be problematic? pinging 127.0.0.1 or ::1, as well as localhost or my hostname (both -4 and -6) always gets me a reply though, so I'm not sure if this is the actual problem...
This is my hosts file (I omitted an alias to my small webserver):
#
# /etc/hosts: static lookup table for host names
#
#<ip-address> <hostname.domain.org> <hostname>
127.0.0.1 localhost.localdomain localhost
::1 localhost.localdomain localhostOffline
Put your hostname like this:
127.0.0.1 localhost.localdomain localhost yourhostnameAnd disable ipv6 unless you use it. 127.0.0.1 is and should be the local machine, you can't use it for something else.
Offline
Okay, I added my hostname to both the localhost lines in /etc/hosts and now my hostname is resolved correctly:
jplatte@jp-tuxedo ~ % ping -4 jp-tuxedo
PING localhost.localdomain (127.0.0.1) 56(84) bytes of data.
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=1 ttl=64 time=0.024 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=2 ttl=64 time=0.039 ms
^C
--- localhost.localdomain ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1008ms
rtt min/avg/max/mdev = 0.024/0.031/0.039/0.009 msThe program startup issue remains, though; even after a reboot.
Why would I disable ipv6? I want to use it where it's available. Or do you just mean disable it temporarily to make troubleshooting easier?
Offline
Why would I disable ipv6? I want to use it where it's available. Or do you just mean disable it temporarily to make troubleshooting easier?
If there is ipv6, it must work correctly too. With your "ping -4"; you only test ipv4. In the past, some router/adsl modem were not able to handle ipv6 and sometimes the system gets no answer. The hang due to a misconfigured network is due to some network connection that doesn't respond. The system waiting for 20s for an answer before deciding it has failed. If ipv6 is completely disabled, any ipv6 things will immediately fail and no hang will happen. So if you use ipv6 and it is working properly, you can use it (but test then that you can ping localhost and your hostname in ipv6 too). If you do not, disable it in order to be sure in order to be sure that bad configuration/router do not cause the previously mentioned hangs.
Offline
Ah, okay. Well with the current hosts configuration, resolving my own hostname should never hit the network though, right?
In any case, as I wrote in #6 pinging localhost or my own hostname via ipv6 (which ping seems to default to) worked all along. I just showed the output of -4 because I wanted to show ipv4 is working now too.
Last edited by jonessen (2017-04-14 16:45:16)
Offline
What do you mean by "starting with a fresh config"? Deleting your .config?!
Exactly. I moved all .$dirs to a backup folder and moved back obviously non related ones like .mozilla or .ssh and so on. And the problem is permanently gone ... Maybe if its raining tomorrow I'll have a closer look.
Offline
Okay, so now I've tried this whole thing with a new user (=> fresh config). Still the same problem. Starting i3 ("exec i3" in .xinitrc, then run startx) and executing nautilus or gedit leaves me waiting for around 25 seconds. Doing the same in gnome-shell ("gnome-shell --wayland") starts the program with less than a second of delay.
Offline
Can you post your .xinitrc? The exec i3 is after the essential includes that are part of the default .xinitrc I hope?
Offline
Uhhh, no. That's all that is in there, and that is how it has been for quite a while. I didn't know I was supposed to have anything else in there, but I'm reading the xinit wiki article now...
Offline
Well then... I guess not running /etc/X11/xinit/xinitrc.d/50-systemd-user.sh was what caused this. I now have all the stuff from /etc/X11/xinit/xinitrc in my own .xinitrc and the issue is fixed. Thanks a lot!
Last edited by jonessen (2017-04-18 10:53:06)
Offline