NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: kern/52783: parallel fsck hangs during boot of 8.99.[5678]



The following reply was made to PR kern/52783; it has been noted by GNATS.

From: Onno van der Linden <o.vd.linden%quicknet.nl@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: kern/52783: parallel fsck hangs during boot of 8.99.[5678]
Date: Sat, 14 Jul 2018 20:24:36 +0200

 Some additional debug info.
 
 My system has 3 drives
 wd0 (partitions wd0[abefghijk])
 wd1 (partitions wd1[abefgh])
 dk (partitions dk[0123])
 
 In my fstab wd0a has passno 1, the rest has 2.
 
 Recent kernel (8.99.21 sources from July 13th 2018)
 
 Parallel fsck starts, does
 rdk0
 rwd1a
 rwd0e
 rdk1
 
 and hangs.
 Into ddb and ps says
 
 PID FLAGS LWP       NAME     WAIT
  58   0   c3213d20  fsck     biolock
  57   0   c31a9020  fsck_ffs biolock
  54   0   c31a92c0  fsck_ffs tstile
  50   0   c31a9560  fsck     wait
  46  80   c31a9aa0  sh       wait
  17   0   c31a9d40  sh       tstile
  16   0             sh       wait
  11   0             sh       pipe_rd
  10   0             sh       wait
   9   0             sh       wait
   8   0             sh       wait
   2   0             sh       wait
   1                 init
 
 tr /a c3213d20         tr /a c31a9020
 pid 58                 pid 57
 sleepq_block           sleepq_block
 cv_timedwait           cv_timedwait
 bbusy                  bbusy
 getblk                 getblk
 bio_doread             bio_doread
 bread                  bread
 ffs_initvnode          ffs_update.part.3
 ffs_loadvnode          ffs_full_fsync
 vcache_get             VOP_FSYNC
 ufs_lookup             ffs_sync
 VOP_LOOOKUP            VFS_SYNC
 lookup_once            do_sys_sync
 namei_tryemulroot      sys_sync
 namei                  syscall
 namei_at_simple_user
 do_sys_pstatvfs
 sys_statvfs1
 syscall
 
 tr /a c31a92c0         tr /a c31a9560
 pid 54                 pid 50
 sleepq_block           sleepq_blcok
 turnstile_block        cv_wait
 mutex_vector_enter     fork1
 do_sys_sync            sys___vfork14
 sys_sync               syscall
 syscall
 
 
 8.99.21 kernel (same sources as earlier) compiled with
 LOCKDEBUG. And after 5 reboot attempts .......
 
 Parallel fsck starts, does
 dk0
 wd0e
 wd1a
 wd0f
 dk1
 wd1b
 wd0g
 dk2
 wd1e
 dk3
 
 and hangs.
 Into ddb and ps says
 
 PID FLAGS LWP       NAME     WAIT
  69   0   c31fa800  sh       tstile
  59   0   c31fa020  fsck_ffs biolock
  61   4   c31fa2c0  fsck_ffs physio
  50   0   c31aa560  fsck     wait
  46  80             sh       wait
  17   0             sh       tstile
  16   0             sh       wait
  11   0             sh       pipe_rd
  10   0             sh       wait
   9   0             sh       wait
   8   0             sh       wait
   2   0             sh       wait
   1                 init
 
 tr /a c31fa2c0         tr /a c31fa020
 pid 61                 pid 59
 sleepq_block           sleepq_block
 cv_wait                cv_timedwait
 physio                 bbusy
 wdread                 getblk
 cdev_read              bio_doread
 spec_read              bread
 VOP_READ               ffs_update.part.3
 vn_read                ffs_full_fsync
 dofileread             VOP_FSYNC
 sys_pread              ffs_sync
 syscall                VFS_SYNC
                        do_sys_sync
                        sys_sync
                        syscall
 
 
 Onno
 


Home | Main Index | Thread Index | Old Index