urxvt + ncurses causes machine to hang

JohnCC · 10-23-09, 06:38 AM

I have a problem in Archlinux where using ncurses apps like alsamixer or vitetris in urxvt cause the machine to hang. The hang is total - no ping, no magic sysrq. I don't get the same problem in other terminals like xterm or xfce4's terminal.

I've tried using strace/ltrace after remounting the root FS sync, but there's nothing obvious. The crash seems to occur while writing escape codes to the terminal AFAICT. I also tried using netconsole, but I can't get anything from the kernel before the crash at all.

Since I've also had a couple of occasions lately where just typing in urxvt has crashed X, it makes me wonder if this freeze is related to Xorg/Nvidia drivers/GPU, especially as the kernel seems to know nothing about it. Also, changing the xorg driver to "vesa" stops the crash happening.

Could such a specific combination really cause the driver to crash? Any idea how I can possibly debug this? There is nothing in the Xorg logs.

rxvt-unicode 9.06-3
kernel26 2.6.31.4-1
xorg-server 1.6.3.901-1
nvidia-beta 190.40-1

[Note, I was running the current stable release, nvidia 185.18.36-2 which showed the same problem, before someone suggested I try the RC]

10-23-09, 06:38 AM	#1
JohnCC Registered User Join Date: Oct 2007 Location: Dorset, UK Posts: 2	urxvt + ncurses causes machine to hang I have a problem in Archlinux where using ncurses apps like alsamixer or vitetris in urxvt cause the machine to hang. The hang is total - no ping, no magic sysrq. I don't get the same problem in other terminals like xterm or xfce4's terminal. I've tried using strace/ltrace after remounting the root FS sync, but there's nothing obvious. The crash seems to occur while writing escape codes to the terminal AFAICT. I also tried using netconsole, but I can't get anything from the kernel before the crash at all. Since I've also had a couple of occasions lately where just typing in urxvt has crashed X, it makes me wonder if this freeze is related to Xorg/Nvidia drivers/GPU, especially as the kernel seems to know nothing about it. Also, changing the xorg driver to "vesa" stops the crash happening. Could such a specific combination really cause the driver to crash? Any idea how I can possibly debug this? There is nothing in the Xorg logs. rxvt-unicode 9.06-3 kernel26 2.6.31.4-1 xorg-server 1.6.3.901-1 nvidia-beta 190.40-1 [Note, I was running the current stable release, nvidia 185.18.36-2 which showed the same problem, before someone suggested I try the RC]