如何分析D状态进程-白红宇

在使用top查看进程状态时，我们有时候会看到D状态的进程。

w: S  --  Process Status          The status of the task which can be one of:             ’D’ = uninterruptible sleep             ’R’ = running             ’S’ = sleeping             ’T’ = traced or stopped             ’Z’ = zombie

D是一种不可中断的sleep，如果你发现大量的D状态的进程，这个时候这些进程实际上是没有在处理业务逻辑的。

例如使用PostgreSQL时，批量的往数据库导入数据，如果导入的数据量大到OS脏页回写的速度赶不上写入的速度时，并且用户刷dirty page的阈值到达，用户进程会需要主动刷脏页。

vm.dirty_background_ratio = 10vm.dirty_background_bytes = 0vm.dirty_ratio = 20vm.dirty_bytes = 0vm.dirty_writeback_centisecs = 50vm.dirty_expire_centisecs = 6000

例如以上配置，OS脏页超过20%时，用户调write也需要主动的刷脏页，就会看到进程处于D状态，直到脏页水位下降到10%以下。

当然还有其他的原因会导致进程进入D状态，我们需要观察进程的stack，看看它处于什么状态。

例如处于R状态的COPY PostgreSQL进程，它的stack是什么样的？

cat /proc/17944/status ; echo -e "\n"; cat /proc/17944/stackName:   postgresState:  R (running)Tgid:   17944Pid:    17944PPid:   57925TracerPid:      0Uid:    123293  123293  123293  123293Gid:    100     100     100     100Utrace: 0FDSize: 64Groups: 100 19001 VmPeak: 272294920 kBVmSize:   119788 kBVmLck:         0 kBVmHWM:      3244 kBVmRSS:      2812 kBVmData:     2140 kBVmStk:       152 kBVmExe:      5852 kBVmLib:      2400 kBVmPTE:        64 kBVmSwap:        0 kBThreads:        1SigQ:   0/4131614SigPnd: 0000000000000000ShdPnd: 0000000000000000SigBlk: 0000000000000000SigIgn: 0000000001301800SigCgt: 0000000180006287CapInh: 0000000000000000CapPrm: 0000000000000000CapEff: 0000000000000000CapBnd: ffffffffffffffffCpus_allowed:   ffffffff,ffffffffCpus_allowed_list:      0-63Mems_allowed:   00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000001Mems_allowed_list:      0voluntary_ctxt_switches:        55758nonvoluntary_ctxt_switches:     103995[
    
     ] sys_semtimedop+0x81a/0x840[
     
      ] 0xffffffffffffffff

例如处于D状态的COPY PostgreSQL进程，它的stack是什么样的？

可以看到它处于刷脏页速度受限的状态，与ext4内核有关。

cat /proc/17944/status ; echo -e "\n"; cat /proc/17944/stackName:   postgresState:  D (disk sleep)Tgid:   17944Pid:    17944PPid:   57925TracerPid:      0Uid:    123293  123293  123293  123293Gid:    100     100     100     100Utrace: 0FDSize: 64Groups: 100 19001 VmPeak: 272294920 kBVmSize:   119788 kBVmLck:         0 kBVmHWM:      3244 kBVmRSS:      2812 kBVmData:     2140 kBVmStk:       152 kBVmExe:      5852 kBVmLib:      2400 kBVmPTE:        64 kBVmSwap:        0 kBThreads:        1SigQ:   0/4131614SigPnd: 0000000000000000ShdPnd: 0000000000000000SigBlk: 0000000000000000SigIgn: 0000000001301800SigCgt: 0000000180006287CapInh: 0000000000000000CapPrm: 0000000000000000CapEff: 0000000000000000CapBnd: ffffffffffffffffCpus_allowed:   ffffffff,ffffffffCpus_allowed_list:      0-63Mems_allowed:   00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000001Mems_allowed_list:      0voluntary_ctxt_switches:        55922nonvoluntary_ctxt_switches:     104189[
    
     ] balance_dirty_pages_ratelimited_nr+0x2d0/0x9a0[
     
      ] generic_file_buffered_write+0x1da/0x2e0[
      
       ] __generic_file_aio_write+0x260/0x490[
       
        ] generic_file_aio_write+0x88/0x100[
        
         ] ext4_file_write+0x43/0xe0 [ext4][
         
          ] do_sync_write+0xfa/0x140[
          
           ] vfs_write+0xb8/0x1a0[
           
            ] sys_write+0x51/0x90[
            
             ] system_call_fastpath+0x16/0x1b[
             
              ] 0xffffffffffffffff