The profile for qemu uses:
#include <abstractions/consoles>
And that has:
/dev/pts/[0-9]* rw,
But the strace doesn't show denies on pts, so is this a red herring?
0.000034 stat("/dev/pts/0", {st_mode=S_IFCHR|0600, st_rdev=makedev(136,
0), ...}) = 0 <0.000017>
0.000052 openat(AT_FDCWD, "/dev/pts/0", O_RDWR|O_NOCTTY) = 11 <0.000019>
0.000330 ioctl(11, TCGETS, {B38400 opost isig icanon echo ...}) = 0
<0.000105>
0.000139 ioctl(11, TCGETS, {B38400 opost isig icanon echo ...}) = 0
<0.000010>
0.000034 ioctl(11, SNDCTL_TMR_CONTINUE or TCSETSF, {B38400 -opost -isig
-icanon -echo ...}) = 0 <0.000013>
0.000037 ioctl(11, TCGETS, {B38400 -opost -isig -icanon -echo ...}) = 0
<0.000010>
0.000034 ioctl(10, TCGETS, {B38400 -opost -isig -icanon -echo ...}) = 0
<0.000011>
0.000033 ioctl(10, TIOCGPTN, [0]) = 0 <0.000010>
0.000033 stat("/dev/pts/0", {st_mode=S_IFCHR|0600, st_rdev=makedev(136,
0), ...}) = 0 <0.000016>
0.000045 close(11) = 0 <0.000013>
The only Permission denied thou (at all) is on /dev/pts/0 with this call:
0.000055 ioctl(10, TIOCGPTPEER, 0x102) = -1 EACCES (Permission denied)
<0.000025>
Maybe try to get gdb on it, and break on syscall open?
That would give the BT as well, but I think we are not fast enough to catch it
in time.
I tried I could not catch it fast enough in a few tries (directly hooked gdb in
a pidof loop).
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1764373
Title:
copy-storage migrations fail on nbd-server-add
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1764373/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
