Swap doesn't work

Hi,

I've made my linux-from-scratch with latest stable (2.4.19) kernel, made swap, turned it on but it doesn't work. It seems it does but when there's not enough memory, the system crashes. Either it kills the application desiring more memory (gcc or something) or crashes the kernel with memory dump. Neither the 2.4.20-pre5-ac3 helped.

Thank you for your help,

Vladimir Trebicky

--
Vladimir Trebicky
[email protected]

Does this occur if you take a kernel build on a standard distribution and boot it on your box rather than one generated by a hand built tool chain ?

Does your swap partition show up in /proc/swaps? It has to contain something like this:

Filename Type Size Used Priority
/dev/hda6 partition 506008 0 -1

Btw, do you see something swap-related in dmesg? Like:

Unable to find swap-space signature
Unable to handle swap header version ...
Swap area shorter than signature indicates
Empty swap-file

And do you actually see something like this:
Adding Swap: 506008k swap-space (priority -1)

How did you initialized the swap partition? Recent kernels support both v1 and v2 swaps, which is can be set for mkswap using -v0 (-v1). Actually i mean did you initialized it at all? 8)

-alex

I have
/dev/hda6 partition 594364 0 -1
in my /proc/swaps

In dmesg I see only this, but some problem with signanture is in syslog (at the
end of this mail)

$ dmesg | grep swap
Starting kswapd
Adding Swap: 594364k swap-space (priority -1)

I just created a partition with fdisk /dev/hda6, done "mkswap /dev/hda6" put the information to /etc/fstab and turned it on with "swapon -a". TOP shows
Swap: 594364K av, 0K used, 594364K free

syslog logs these kinds of kernel messages (those I guess are important):

Sep 29 22:04:19 shunka kernel: swap_free: Bad swap offset entry 1b3d0000
...
Sep 29 22:04:19 shunka kernel: swap_free: Bad swap offset entry 1b3d0000
...
Sep 10 10:03:28 shunka2 kernel: swap_dup: Bad swap file entry 00000022
...
Sep  4 21:30:40 shunka kernel: Unable to find swap-space signature        //
!!!!!!!!
...
Oct 26 19:25:29 shunka kernel:  <1>Unable to handle kernel paging request at
virtual address 2064656e
Oct 26 19:25:29 shunka kernel:  printing eip:
Oct 26 19:25:29 shunka kernel: c012f781
Oct 26 19:25:29 shunka kernel: *pde = 00000000
Oct 26 19:25:29 shunka kernel: Oops: 0000
Oct 26 19:25:29 shunka kernel: CPU:    0
Oct 26 19:25:29 shunka kernel: EIP:    0010:[<c012f781>]    Not tainted
Oct 26 19:25:29 shunka kernel: EFLAGS: 00010202
Oct 26 19:25:29 shunka kernel: eax: c026a19c   ebx: c026a19c   ecx: c105f584
edx: 2064656e
Oct 26 19:25:29 shunka kernel: esi: c95472f8   edi: c1d57000   ebp: c95472f8
esp: c459de94
Oct 26 19:25:29 shunka kernel: ds: 0018   es: 0018   ss: 0018
Oct 26 19:25:29 shunka kernel: Process cpp0 (pid: 12431, stackpage=c459d000)
Oct 26 19:25:29 shunka kernel: Stack: c105f584 c012f465 c026a19c 01d56067
c105f584 c012091c cb728da0 080be6e4
Oct 26 19:25:29 shunka kernel:        080be6e4 c3823080 c0120963 cb728da0
c60fd7a0 c95472f8 00000001 080be6e4
Oct 26 19:25:29 shunka kernel:        cb728da0 080be6e4 080be6e4 c3823080
c0120b8b cb728da0 c60fd7a0 080be6e4
Oct 26 19:25:29 shunka kernel: Call Trace:    [<c012f465>] [<c012091c>]
[<c0120963>] [<c0120b8b>] [<c010fe47>]
Oct 26 19:25:29 shunka kernel:   [<c010fd34>] [<c0121f78>] [<c0122067>]
[<c0120f40>] [<c0108844>]
Oct 26 19:25:29 shunka kernel:
Oct 26 19:25:29 shunka kernel: Code: 8b 02 89 83 b4 00 00 00 c7 02 00 00 00
00 89 d0 5b c3 90 56
Oct 26 19:25:29 shunka kernel:  <1>Unable to handle kernel paging request at
virtual address 6164262c
Oct 26 19:25:29 shunka kernel:  printing eip:
Oct 26 19:25:29 shunka kernel: c012f510
Oct 26 19:25:29 shunka kernel: *pde = 00000000
Oct 26 19:25:29 shunka kernel: Oops: 0000
Oct 26 19:25:29 shunka kernel: CPU:    0
Oct 26 19:25:29 shunka kernel: EIP:    0010:[<c012f510>]    Not tainted
Oct 26 19:25:29 shunka kernel: EFLAGS: 00010206
Oct 26 19:25:29 shunka kernel: eax: 61642628   ebx: c1089dfc   ecx: 00000010
edx: c954713c
Oct 26 19:25:29 shunka kernel: esi: c026a19c   edi: c3789760   ebp: 08448000
esp: c459dd10
Oct 26 19:25:29 shunka kernel: ds: 0018   es: 0018   ss: 0018
Oct 26 19:25:29 shunka kernel: Process cpp0 (pid: 12431, stackpage=c459d000)
Oct 26 19:25:29 shunka kernel: Stack: c954713c 00017000 00007000 c011fa1d
c60fd5c0 cb728da0 00017000 08048000
Oct 26 19:25:29 shunka kernel:        08448000 c3823084 c3823084 00000008
00000000 0805f000 c3823080 00000000
Oct 26 19:25:29 shunka kernel:        0805f000 c0122192 cb728da0 08048000
00017000 cb728da0 c459de60 c459c000
Oct 26 19:25:29 shunka kernel: Call Trace:    [<c011fa1d>] [<c0122192>]
[<c0112212>] [<c0116265>] [<c0108d59>]
Oct 26 19:25:29 shunka kernel:   [<c0110037>] [<c010fd34>] [<c012091c>]
[<c0120963>] [<c0120b8b>] [<c010fe47>]
Oct 26 19:25:29 shunka kernel:   [<c015e6c2>] [<c0108844>] [<c012f781>]
[<c012f465>] [<c012091c>] [<c0120963>]
Oct 26 19:25:29 shunka kernel:   [<c0120b8b>] [<c010fe47>] [<c010fd34>]
[<c0121f78>] [<c0122067>] [<c0120f40>]
Oct 26 19:25:29 shunka kernel:   [<c0108844>]
Oct 26 19:25:29 shunka kernel:
Oct 26 19:25:29 shunka kernel: Code: 39 50 04 75 0e 56 53 57 50 e8 76 02 00
00 83 c4 10 eb 08 89
Oct 26 19:25:29 shunka kernel:  <1>Unable to handle kernel paging request at
virtual address 2064656e
Oct 26 19:25:29 shunka kernel:  printing eip:
Oct 26 19:25:29 shunka kernel: c012f781
Oct 26 19:25:29 shunka kernel: *pde = 00000000
Oct 26 19:25:29 shunka kernel: Oops: 0000
Oct 26 19:25:29 shunka kernel: CPU:    0
Oct 26 19:25:29 shunka kernel: EIP:    0010:[<c012f781>]    Not tainted
Oct 26 19:25:29 shunka kernel: EFLAGS: 00010202
Oct 26 19:25:29 shunka kernel: eax: c026a19c   ebx: c026a19c   ecx: c97fe744
edx: 2064656e
Oct 26 19:25:29 shunka kernel: esi: c97fe744   edi: 081d1e42   ebp: c39f4080
esp: c6657ebc
Oct 26 19:25:29 shunka kernel: ds: 0018   es: 0018   ss: 0018
Oct 26 19:25:29 shunka kernel: Process cc1 (pid: 12432, stackpage=c6657000)
Oct 26 19:25:29 shunka kernel: Stack: c110d17c c012f465 c026a19c c110d17c
081d1e42 c0120aa1 cb7289e0 081d1e42
Oct 26 19:25:29 shunka kernel:        081d1e42 c39f4080 c0120b8b cb7289e0
c493ed40 081d1e42 00000000 c97fe744
Oct 26 19:25:29 shunka kernel:        cb7289e0 081d1e42 cb7289fc c493ed40
c010fe47 cb7289e0 c493ed40 081d1e42
Oct 26 19:25:29 shunka kernel: Call Trace:    [<c012f465>] [<c0120aa1>]
[<c0120b8b>] [<c010fe47>] [<c010fd34>]
Oct 26 19:25:29 shunka kernel:   [<c01e0e5d>] [<c01e0f5e>] [<c01173ea>]
[<c0109c0d>] [<c0108844>]
Oct 26 19:25:29 shunka kernel:
Oct 26 19:25:29 shunka kernel: Code: 8b 02 89 83 b4 00 00 00 c7 02 00 00 00
00 89 d0 5b c3 90 56
...
Oct 26 19:20:52 shunka kernel: kernel BUG at page_alloc.c:117!
Oct 26 19:20:52 shunka kernel: invalid operand: 0000
Oct 26 19:20:52 shunka kernel: CPU:    0
Oct 26 19:20:52 shunka kernel: EIP:    0010:[<c012ab39>]    Not tainted
Oct 26 19:20:52 shunka kernel: EFLAGS: 00010282
Oct 26 19:20:52 shunka kernel: eax: 01000010   ebx: c1111d40   ecx: c026a19c
edx: c10c42a4
Oct 26 19:20:52 shunka kernel: esi: 00000000   edi: 00148000   ebp: 085e5000
esp: c8fb1e30
Oct 26 19:20:52 shunka kernel: ds: 0018   es: 0018   ss: 0018
Oct 26 19:20:52 shunka kernel: Process cc1 (pid: 12012, stackpage=c8fb1000)
Oct 26 19:20:52 shunka kernel: Stack: c1111d40 001e9000 00148000 085e5000
c01173ea 00000046 00000000 c026a19c
Oct 26 19:20:52 shunka kernel:        c103400c c026a1d8 00000216 ffffffff
00003321 c012b389 c012b807 c1111d40
Oct 26 19:20:52 shunka kernel:        c011f5f9 c1111d40 c3af5cb4 c011fa2b
05441067 c493ec80 cb728bc0 001e9000
Oct 26 19:20:52 shunka kernel: Call Trace:    [<c01173ea>] [<c012b389>]
[<c012b807>] [<c011f5f9>] [<c011fa2b>]
Oct 26 19:20:52 shunka kernel:   [<c0122192>] [<c0112212>] [<c0116265>]
[<c011b21a>] [<c01085cf>] [<c010fd34>]
Oct 26 19:20:52 shunka kernel:   [<c0122067>] [<c01173ea>] [<c0120f40>]
[<c0108844>] [<c0108774>]
Oct 26 19:20:52 shunka kernel:
Oct 26 19:20:52 shunka kernel: Code: 0f 0b 75 00 2c c1 22 c0 8b 43 18 24 eb

89 43 18 c6 43 24 05

Thanks,
Vladimir Trebicky

--
Vladimir Trebicky
[email protected]

Vladim?r Trebick?, Sun, Oct 27, 2002 09:51:17 +0100:

looks ok.

May i assume you did swapoff before?

You change hostname inbetween or this is just a typo?

Wow. Any of the errors above prevents swap partition from being used. How did you manage to see anything in /proc/swaps?
I suggest you do:
swapoff /dev/hda6
badblocks /dev/hda6

Alternatively, you can try

dd if=/dev/zero of=/dev/hda6; mkswap /dev/hda6

Look for "SWAP-SPACE" (old swap) or "SWAPSPACE2" (the new one). Just to make sure you've initialized the partition properly. Than turn it on: swapon /dev/hda6; tail /var/log/syslog

Oops, you've sent, is pretty useless without decoding. Read Documentation/oops-tracing.txt from the kernel source tree.

-alex

Am Sonntag, 27. Oktober 2002 09:51 schrieb Vladimír Trebický:

Just one hint, maybe this works: dump your swap partition witz zeros, i.e. dd if=/dev/zero of=/dev/hda6
Keep an eye on the logs (Any IO-errors here regarding /dev/hda6?) re-create your swap partition with mkswap

Of course, you should turn off swapping before starting this ;-)

Greetings,
Alex

Alexander Puchmayr            Systemadministrator for Theoretical Physics
University Linz, Austria      e-mail: [email protected]
Altenbergerstrasse 69         phone: +43/732/2468-8633
A-4040 Linz-Auhof             FAX:   +43/732/2468-8585