DragonFlyBSD Kernel Audit
DF-0079 / serial_wedge_capture.txt
← back to finding ↓ download raw
##### WATCHER2_START Wed Jul  1 22:13:05 UTC 2026 #####

##### baseline top: #####
load averages:  0.19,  0.06,  0.02;               up 0+00:01:13        22:13:06
21 processes: 2 running, 19 sleeping
CPU states:  1.8% user,  0.0% nice, 27.0% system,  0.4% interrupt, 70.7% idle
Memory: 86M Active, 6460K Inact, 349M Wired, 50M Cache, 85M Buf, 1522M Free
Swap: 4096M Total, 4096M Free

   PID USERNAME   NICE  SIZE    RES    STATE   C   TIME   CTIME    CPU COMMAND
   699 root         0    25M  2456K   kqread   0   0:00    0:00  0.88% sshd

##### WEDGE pid=852 iter=0 t=22:13:06 #####
   PID   PPID STAT     UID %CPU      TIME UCOMM            COMMAND
   852      1 R0      1001  0.0   0:00.50 df0079           ./df0079
load averages:  0.19,  0.06,  0.02;               up 0+00:01:14        22:13:07
21 processes: 2 running, 19 sleeping
CPU states:  0.0% user,  0.0% nice, 50.0% system,  0.0% interrupt, 50.0% idle
Memory: 87M Active, 6432K Inact, 349M Wired, 50M Cache, 85M Buf, 1522M Free
Swap: 4096M Total, 4096M Free

##### WEDGE pid=852 iter=1 t=22:13:07 #####
   PID   PPID STAT     UID %CPU      TIME UCOMM            COMMAND
   852      1 R0      1001  0.0   0:01.68 df0079           ./df0079
load averages:  0.19,  0.06,  0.02;               up 0+00:01:15        22:13:08
21 processes: 2 running, 19 sleeping
CPU states:  0.0% user,  0.0% nice, 50.0% system,  0.0% interrupt, 50.0% idle

##### WEDGE pid=852 iter=2 t=22:13:08 #####
   PID   PPID STAT     UID %CPU      TIME UCOMM            COMMAND
   852      1 R0      1001  0.0   0:02.86 df0079           ./df0079
CPU states:  0.0% user,  0.0% nice, 50.0% system,  0.0% interrupt, 50.0% idle

##### WEDGE pid=852 iter=3 t=22:13:09 #####
   852      1 R0      1001  0.0   0:04.04 df0079           ./df0079

##### WEDGE pid=852 iter=4 t=22:13:10 #####
   852      1 R0      1001  0.0   0:05.22 df0079           ./df0079

##### WEDGE pid=852 iter=5 t=22:13:11 #####
   852      1 R0      1001  0.0   0:06.40 df0079           ./df0079

##### WEDGE pid=852 iter=6 t=22:13:13 #####
   852      1 R0      1001  0.0   0:07.58 df0079           ./df0079

##### WEDGE pid=852 iter=7 t=22:13:14 #####
   852      1 R0      1001  0.0   0:08.76 df0079           ./df0079

##### WEDGE pid=852 iter=8 t=22:13:15 #####
   852      1 R0      1001  0.0   0:09.94 df0079           ./df0079

##### WEDGE pid=852 iter=9 t=22:13:16 #####
   852      1 R0      1001  0.0   0:11.11 df0079           ./df0079

##### WEDGE pid=852 iter=10 t=22:13:17 #####
   852      1 R0      1001  0.0   0:12.30 df0079           ./df0079

##### WEDGE pid=852 iter=11 t=22:13:19 #####
   852      1 R0      1001  0.0   0:13.47 df0079           ./df0079

##### WEDGE pid=852 iter=12 t=22:13:20 #####
   852      1 R0      1001  0.0   0:14.66 df0079           ./df0079

##### WEDGE pid=852 iter=13 t=22:13:21 #####
   852      1 R0      1001  0.0   0:15.84 df0079           ./df0079

##### WEDGE pid=852 iter=14 t=22:13:22 #####
   852      1 R0      1001  0.0   0:17.02 df0079           ./df0079

##### WEDGE pid=852 iter=15 t=22:13:23 #####
   852      1 R0      1001  0.0   0:18.20 df0079           ./df0079

##### WEDGE pid=852 iter=16 t=22:13:24 #####
   852      1 R0      1001  0.0   0:19.39 df0079           ./df0079

##### WEDGE pid=852 iter=17 t=22:13:26 #####
   852      1 R0      1001  0.0   0:20.56 df0079           ./df0079

################################################################################
# SUMMARY (extracted from the per-iteration snapshots above)
################################################################################
# Process: pid=852, PPID=1 (orphaned, parent exited), UID=1001 (maxx, UNPRIVILEGED)
# STAT:    R0  = Running on CPU, no lock held, NOT sleeping/blocked -- i.e. the
#          kernel thread is spinning in on-CPU kernel context and never blocks.
# COMMAND: ./df0079   (the PoC: write(/dev/null, (void*)0x1, 0x100000000))
# CPUTIME: monotonically increasing with NO progress in residual:
#            iter 0 (t+0.0s): 0:00.50
#            iter 1 (t+0.7s): 0:01.68   (+1.18s)
#            iter 2 (t+1.4s): 0:02.86   (+1.18s)
#            ...
#            iter17 (t+19.5s):0:20.56   (+1.18s)
#          => ~1.18s of CPU per ~1.0s wall = 100% of one core consumed forever.
# CPU states (top): "50.0% system, 50.0% idle" on a 2-CPU guest => exactly one
#          CPU fully in kernel `sys` (the wedged mmrw loop); the other idle/used
#          by the watcher. With N copies (fork variant), N CPUs wedge => 100%.
# RECOVERY: the process is unkillable from userspace (stuck in an unyielding
#          kernel loop with no signal-check point). Only a hard reset recovers.
################################################################################