标题: 无法回收的僵尸进程

https://scz.617.cn/unix/201511041359.txt

zyh 2015-11-04 13:59

通常情况下父进程结束后，僵尸进程会被init回收。但是可能出现以下情况:

# ps -elf
F S UID        PID  PPID  C PRI  NI ADDR SZ WCHAN  STIME TTY          TIME CMD
...
0 Z test      2885     1  0  80   0 -     0 exit   Aug11 ?        00:10:38 [adb] <defunct>
...

2885的父进程已经是init，但它仍保持为僵尸进程。这是因为该进程还有一个内核态
的线程:

# ps aumx
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
...
test      2885  0.0  0.0      0     0 ?        -    Aug11  10:38 [adb] <defunct>
test         -  0.0    -      -     - -        Zl   Aug11   0:00 -
test         -  0.0    -      -     - -        Dl   Aug11  10:38 -
...

STAT列的Dl表示"uninterruptible sleep state"，这个线程等待在内核态，并且不
接受任何中断，包括SIGKILL。

用"magic SysRq key"可以看到线程的调用栈:

# echo "1" > /proc/sys/kernel/sysrq
# echo 'w' > /proc/sysrq-trigger
# dmesg
...
[7369283.509835] Call Trace:
[7369283.509837]  [<ffffffff8165fe7f>] schedule+0x3f/0x60
[7369283.509840]  [<ffffffff8140432e>] rpm_resume+0x17e/0x690
[7369283.509842]  [<ffffffff8108c0e0>] ? add_wait_queue+0x60/0x60
[7369283.509844]  [<ffffffff814053ac>] __pm_runtime_resume+0x5c/0x90
[7369283.509847]  [<ffffffff8149d749>] usb_autoresume_device+0x29/0x60
[7369283.509849]  [<ffffffff814a202b>] usbdev_open+0x11b/0x250
[7369283.509852]  [<ffffffff8117eae9>] chrdev_open+0xd9/0x250
[7369283.509854]  [<ffffffff81178200>] __dentry_open+0x2b0/0x380
[7369283.509856]  [<ffffffff8117ea10>] ? cdev_put+0x30/0x30
[7369283.509859]  [<ffffffff812a05cc>] ? security_inode_permission+0x1c/0x30
[7369283.509862]  [<ffffffff81185fba>] ? inode_permission+0x4a/0x110
[7369283.509863]  [<ffffffff8117887d>] vfs_open+0x3d/0x40
[7369283.509865]  [<ffffffff81179780>] nameidata_to_filp+0x40/0x50
[7369283.509867]  [<ffffffff811887f8>] do_last+0x3f8/0x730
[7369283.509869]  [<ffffffff81189ed1>] path_openat+0xd1/0x3f0
[7369283.509871]  [<ffffffff8108fca0>] ? lock_hrtimer_base.isra.24+0x30/0x60
[7369283.509873]  [<ffffffff8108ff60>] ? hrtimer_try_to_cancel+0x50/0xc0
[7369283.509876]  [<ffffffff8118a312>] do_filp_open+0x42/0xa0
[7369283.509878]  [<ffffffff8131cee1>] ? strncpy_from_user+0x31/0x40
[7369283.509880]  [<ffffffff8118565a>] ? do_getname+0x10a/0x180
[7369283.509882]  [<ffffffff81661d8e>] ? _raw_spin_lock+0xe/0x20
[7369283.509885]  [<ffffffff81197637>] ? alloc_fd+0xf7/0x150
[7369283.509886]  [<ffffffff81179888>] do_sys_open+0xf8/0x240
[7369283.509889]  [<ffffffff811c914a>] compat_sys_open+0x1a/0x20
[7369283.509891]  [<ffffffff8166c870>] sysenter_dispatch+0x7/0x2e
...

这里似乎在等待某个USB设备上的文件，然而本机并未接入USB存储设备。所以，能做
的只有重启了。