Deadlock with 32GB swap-backed md

From: Kris Kennaway <kris_at_obsecurity.org>
Date: Wed, 5 Jan 2005 18:50:42 -0800
In my quest for ever-more-insane things to do to panic package build
machines, I switched them over to use a 32GB swap partition as a
swap-backed md, in the hopes of improving build times by cutting back
on disk accesses.  This seemed to work (I wasn't sure you could use
this much on i386, but the system let me configure as much as 32GB),
except two of them have locked up.  ps shows

17864 c22f1dc8    0   594   594 0000000 new [INACTIVE] inetd
17863 c24085e8    0 17861 11791 0004000 [SLPQ biord 0xc6618460][SLP] tar
17862 c230a1f8    0 17861 11791 0004000 [SLPQ pipdwt 0xc16ff000][SLP] tar
17861 c22ec9d8    0 17535 11791 0004000 [SLPQ wait 0xc22ec9d8][SLP] sh
17535 c1a9ddc8    0 15940 11791 0004000 [SLPQ wait 0xc1a9ddc8][SLP] pkg_add
17024 c22c91f8    0 17022 15818 0004000 [SLPQ piperd 0xc1700300][SLP] tar
17023 c22c93f0    0 17022 15818 0004000 [SLPQ biord 0xc6602030][SLP] bzip2
17022 c23087e0    0 17021 15818 0000000 [SLPQ wait 0xc23087e0][SLP] sh
17021 c2308dc8    0 16772 15818 0004000 [SLPQ wait 0xc2308dc8][SLP] sh
16772 c1a9dbd0    0 16771 15818 0004000 [SLPQ wait 0xc1a9dbd0][SLP] make
16771 c22ca3f0    0 16639 15818 0004000 [SLPQ wait 0xc22ca3f0][SLP] sh
16640 c22c9dc8    0 16638 15818 0000000 [SLPQ pfault 0xc07e9018][SLP] pnohang
16639 c2308000    0 16638 15818 0004000 [SLPQ wait 0xc2308000][SLP] make
16638 c230a000    0 16620 15818 0004000 [SLPQ wait 0xc230a000][SLP] pnohang
16620 c22ca1f8    0 15819 15818 0004000 [SLPQ wait 0xc22ca1f8][SLP] sh
15940 c22f17e0    0 11792 11791 0004000 [SLPQ wait 0xc22f17e0][SLP] sh
15819 c1a9c7e0    0 15818 15818 0004000 [SLPQ wait 0xc1a9c7e0][SLP] sh
15818 c2308bd0    0 15816 15818 0004000 [SLPQ pause 0xc2308c04][SLP] csh
15816 c22cadc8    0   370 15816 0004100 [SLPQ select 0xc07e0464][SLP] sshd
11792 c22ca000    0 11791 11791 0004000 [SLPQ wait 0xc22ca000][SLP] sh
11791 c22c9bd0    0 11789 11791 0004000 [SLPQ pause 0xc22c9c04][SLP] csh
11789 c24087e0    0   370 11789 0004100 [SLPQ select 0xc07e0464][SLP] sshd
  614 c156fdc8    0     1   614 0004002 [SWAP] getty
  613 c1a9d3f0    0     1   613 0004002 [SLPQ ttyin 0xc15e4010][SLP][SWAP] getty
  612 c1a9c1f8    0     1   612 0004002 [SLPQ ttyin 0xc15e4410][SLP][SWAP] getty
  611 c1a9c3f0    0     1   611 0004002 [SLPQ ttyin 0xc15e4810][SLP][SWAP] getty
  610 c1a9d7e0    0     1   610 0004002 [SLPQ ttyin 0xc15e4c10][SLP][SWAP] getty
  609 c1a9c5e8    0     1   609 0004002 [SLPQ ttyin 0xc15e5010][SLP][SWAP] getty
  608 c1a9d5e8    0     1   608 0004002 [SLPQ ttyin 0xc15e5410][SLP][SWAP] getty
  607 c16b89d8    0     1   607 0004002 [SLPQ ttyin 0xc15e5810][SLP][SWAP] getty
  606 c1a9d1f8    0     1   606 0004002 [SLPQ ttyin 0xc15e5c10][SLP][SWAP] getty
  594 c1a9c9d8    0     1   594 0000000 [SLPQ vmwait 0xc07e9018][SLP] inetd
  573 c1a9d000  100   557   573 0004000 [SLPQ piperd 0xc16ff480][SLP] unlinkd
  557 c1a9c000  100   555   555 0004000 [SLPQ pfault 0xc07e9018][SLP] squid
  555 c1a9cdc8  100     1   555 0000000 [SLPQ wait 0xc1a9cdc8][SLP][SWAP] squid
  414 c1a9cbd0    0     0     0 0000204 [SLPQ vmwait 0xc07e9018][SLP] md3
  385 c16b85e8    0     1   385 0000000 [SLPQ pfault 0xc07e9018][SLP] cron
  370 c16b8000    0     1   370 0000100 [SLPQ pfault 0xc07e9018][SLP] sshd
  355 c156f7e0    0     1   355 0000000 [SLPQ pfault 0xc07e9018][SLP] ntpd
  280 c156fbd0    0     0     0 0000204 [SLPQ mdwait 0xc1a21000][SLP] md2
  247 c156f3f0    0     1   247 0000000 [SLPQ pfault 0xc07e9018][SLP] syslogd
  227 c156f5e8    0     1   227 0000000 [SLPQ select 0xc07e0464][SLP][SWAP] devd
  115 c16b87e0    0     0     0 0000204 [SLPQ mdwait 0xc1818800][SLP] md1
   73 c156f9d8    0     0     0 0000204 [SLPQ mdwait 0xc15ea000][SLP] md0
   43 c16b8bd0    0     0     0 0000204 [SLPQ - 0xcc8b3d00][SLP] schedcpu
   42 c1510bd0    0     0     0 0000204 [SLPQ - 0xc07e7f2c][SLP] nfsiod 3
   41 c1510dc8    0     0     0 0000204 [SLPQ - 0xc07e7f28][SLP] nfsiod 2
   40 c156c000    0     0     0 0000204 [SLPQ - 0xc07e7f24][SLP] nfsiod 1
   39 c156c1f8    0     0     0 0000204 [SLPQ - 0xc07e7f20][SLP] nfsiod 0
   38 c156c3f0    0     0     0 0000204 [SLPQ biord 0xc661ec28][SLP] syncer
   37 c156c5e8    0     0     0 0000204 [SLPQ vlruwt 0xc156c5e8][SLP] vnlru
   36 c156c7e0    0     0     0 0000204 [SLPQ psleep 0xc07e09cc][SLP] bufdaemon
    9 c156c9d8    0     0     0 000020c [SLPQ pgzero 0xc07e9714][SLP] pagezero
    8 c156cbd0    0     0     0 0000204 [SLPQ psleep 0xc07e9768][SLP] vmdaemon
    7 c156cdc8    0     0     0 0000204 [SLPQ VMWait 0xc07e9728][SLP] pagedaemon
   35 c156f000    0     0     0 0000204 [IWAIT] swi0: sio
   34 c156f1f8    0     0     0 0000204 [IWAIT] swi6:+
    6 c15075e8    0     0     0 0000204 [SLPQ - 0xc155b040][SLP] thread taskq
   33 c15077e0    0     0     0 0000204 [IWAIT] swi5:+
    5 c15079d8    0     0     0 0000204 [SLPQ - 0xc155b100][SLP] kqueue taskq
   32 c1507bd0    0     0     0 0000204 [IWAIT] swi2: cambio
   31 c1507dc8    0     0     0 0000204 [IWAIT] swi6: task queue
   30 c1510000    0     0     0 0000204 [SLPQ - 0xc07841c0][SLP] yarrow
    4 c15101f8    0     0     0 0000204 [SLPQ - 0xc0786dc8][SLP] g_down
    3 c15103f0    0     0     0 0000204 [SLPQ - 0xc0786dc4][SLP] g_up
    2 c15105e8    0     0     0 0000204 [SLPQ - 0xc0786dbc][SLP] g_event
   29 c15107e0    0     0     0 0000204 [IWAIT] swi1: net
   28 c15109d8    0     0     0 0000204 [IWAIT] swi3: vm
   27 c14e21f8    0     0     0 000020c [IWAIT] swi4: clock sio
   26 c14e23f0    0     0     0 0000204 [IWAIT] irq15: ata1
   25 c14e25e8    0     0     0 0000204 [IWAIT] irq14: ata0
   24 c14e27e0    0     0     0 0000204 [IWAIT] irq13:
   23 c14e29d8    0     0     0 0000204 [IWAIT] irq12:
   22 c14e2bd0    0     0     0 0000204 [IWAIT] irq11:
   21 c14e2dc8    0     0     0 0000204 [IWAIT] irq10: fxp0
   20 c1507000    0     0     0 0000204 [IWAIT] irq9:
   19 c15071f8    0     0     0 0000204 [IWAIT] irq8: rtc
   18 c15073f0    0     0     0 0000204 [IWAIT] irq7: ppc0
   17 c14db000    0     0     0 0000204 [IWAIT] irq6:
   16 c14db1f8    0     0     0 0000204 [IWAIT] irq5:
   15 c14db3f0    0     0     0 0000204 [IWAIT] irq4: sio0
   14 c14db5e8    0     0     0 0000204 [IWAIT] irq3: sio1
   13 c14db7e0    0     0     0 0000204 [IWAIT] irq1: atkbd0
   12 c14db9d8    0     0     0 0000204 [IWAIT] irq0: clk
   11 c14dbbd0    0     0     0 000020c [CPU 0] idle
    1 c14dbdc8    0     0     1 0004200 [SLPQ wait 0xc14dbdc8][SLP] init
   10 c14e2000    0     0     0 0000204 [SLPQ ktrace 0xc0793998][SLP] ktrace
    0 c0786ec0    0     0     0 0000200 [SLPQ vmwait 0xc07e9018][SLP] swapper

Any ideas?

Kris
Received on Thu Jan 06 2005 - 01:50:10 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:25 UTC