Control: found -1 1:9.9.5.dfsg-7
Control: severity grave
Raising to grave because since the latest upgrade of my laptop,
rndc systematically hangs, preventing the machine from reconnecting
automatically after a disconnection. Thus the machine is completely
unusable remotely.
A part of ps output:
UID PID PPID C STIME TTY TIME CMD
root 26300 1 0 2014 ? 00:25:57 /usr/bin/python -O
/usr/share/wicd/daemon/wicd-daemon.py
root 26321 26300 0 2014 ? 00:12:55 [wicd-monitor] <defunct>
root 26365 26300 0 15:50 ? 00:00:00 /bin/sh -c
/etc/network/if-down.d/bind9 > /dev/null 2>&1
root 26366 26365 0 15:50 ? 00:00:00 /bin/sh -e
/etc/network/if-down.d/bind9
root 26367 26366 0 15:50 ? 00:00:00 rndc reconfig
and the backtrace of process 26367:
(gdb) bt full
#0 0x00007f22c23f4482 in do_sigsuspend (set=0x7fff0bd4d220)
at ../sysdeps/unix/sysv/linux/sigsuspend.c:31
resultvar = 18446744073709551102
#1 __GI___sigsuspend (set=0x7fff0bd4d220)
at ../sysdeps/unix/sysv/linux/sigsuspend.c:45
oldtype = 0
#2 0x00007f22c312cb8c in isc.app_ctxrun () from /usr/lib/libisc.so.95
No symbol table info available.
#3 0x00007f22c4cdf1d7 in ?? ()
No symbol table info available.
#4 0x00007f22c23e0b45 in __libc_start_main (main=0x7f22c4cde7d0, argc=2,
argv=0x7fff0bd4d688, init=<optimized out>, fini=<optimized out>,
rtld_fini=<optimized out>, stack_end=0x7fff0bd4d678) at libc-start.c:287
result = <optimized out>
unwind_buf = {cancel_jmp_buf = {{jmp_buf = {0, 111109644434141694,
139787307449364, 140733391885952, 0, 0, -111277123653733890,
-13728647904497154}, mask_was_saved = 0}}, priv = {pad = {0x0,
0x0, 0x7fff0bd4d6a0, 0x7f22c4cdb1a8}, data = {prev = 0x0,
cleanup = 0x0, canceltype = 198497952}}}
not_first_call = <optimized out>
#5 0x00007f22c4cdfc3d in ?? ()
No symbol table info available.
On 2014-09-05 10:24:22 +0100, e-t172 wrote:
> I had an issue on my system that made the named daemon freeze in such
> a way that any rndc command would hang forever. Since rndc is called as
> part of network interface up/down scripts, it made these scripts hang as
> well. As a result, my machine lost its network connection.
>
> Network interface up/down scripts should not have a critical dependency
> on some non-critical part of the system (bind9) behaving correctly.
> Otherwise it makes the network subsystem exceedingly brittle.
>
> To fix the issue, I suggest replacing the line above with the following
> so that the network subsystem will never wait on rndc:
>
> rndc reconfig >/dev/null 2>&1 &
This would fix the hang, but not the real issue.
--
Vincent Lefèvre <[email protected]> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
