DC01 -meteohtml.cgi: page allocation failure

[HWTest]

I noticed this i dmesg on my DC01:

Code: Select all

meteohtml.cgi: page allocation failure. order:1, mode:0x20
[<c002d3f8>] (unwind_backtrace+0x0/0xe0) from [<c007c068>] (__alloc_pages_nodemask+0x460/0x550)
[<c007c068>] (__alloc_pages_nodemask+0x460/0x550) from [<c033b194>] (cache_alloc_refill+0x2cc/0x538)
[<c033b194>] (cache_alloc_refill+0x2cc/0x538) from [<c009adb4>] (kmem_cache_alloc+0x8c/0xb0)
[<c009adb4>] (kmem_cache_alloc+0x8c/0xb0) from [<c0276694>] (sk_prot_alloc.isra.22+0x24/0xec)
[<c0276694>] (sk_prot_alloc.isra.22+0x24/0xec) from [<c02767d8>] (sk_clone+0x14/0x264)
[<c02767d8>] (sk_clone+0x14/0x264) from [<c02a7078>] (inet_csk_clone+0x10/0x84)
[<c02a7078>] (inet_csk_clone+0x10/0x84) from [<c02bcc6c>] (tcp_create_openreq_child+0x14/0x320)
[<c02bcc6c>] (tcp_create_openreq_child+0x14/0x320) from [<c02bac60>] (tcp_v4_syn_recv_sock+0x3c/0x1c4)
[<c02bac60>] (tcp_v4_syn_recv_sock+0x3c/0x1c4) from [<c02bca7c>] (tcp_check_req+0x26c/0x448)
[<c02bca7c>] (tcp_check_req+0x26c/0x448) from [<c02bab20>] (tcp_v4_do_rcv+0x120/0x224)
[<c02bab20>] (tcp_v4_do_rcv+0x120/0x224) from [<c02bc5ac>] (tcp_v4_rcv+0x698/0x7a4)
[<c02bc5ac>] (tcp_v4_rcv+0x698/0x7a4) from [<c029d688>] (ip_local_deliver_finish+0x6c/0x1f4)
[<c029d688>] (ip_local_deliver_finish+0x6c/0x1f4) from [<c029d380>] (ip_rcv_finish+0x114/0x3b0)
[<c029d380>] (ip_rcv_finish+0x114/0x3b0) from [<c02808b4>] (netif_receive_skb+0x290/0x314)
[<c02808b4>] (netif_receive_skb+0x290/0x314) from [<c0281ddc>] (process_backlog+0x84/0xcc)
[<c0281ddc>] (process_backlog+0x84/0xcc) from [<c0281f04>] (net_rx_action+0xb0/0x194)
[<c0281f04>] (net_rx_action+0xb0/0x194) from [<c004d080>] (__do_softirq+0xa8/0x140)
[<c004d080>] (__do_softirq+0xa8/0x140) from [<c004d164>] (do_softirq+0x4c/0x58)
[<c004d164>] (do_softirq+0x4c/0x58) from [<c004d204>] (local_bh_enable_ip+0x94/0xa0)
[<c004d204>] (local_bh_enable_ip+0x94/0xa0) from [<c02c85e8>] (inet_stream_connect+0x16c/0x28c)
[<c02c85e8>] (inet_stream_connect+0x16c/0x28c) from [<c0273b78>] (sys_connect+0x68/0x90)
[<c0273b78>] (sys_connect+0x68/0x90) from [<c00289c0>] (ret_fast_syscall+0x0/0x2c)
Mem-info:
Normal per-cpu:
CPU    0: hi:   42, btch:   7 usd:  13
CPU    1: hi:   42, btch:   7 usd:   2
Active_anon:344 active_file:4217 inactive_anon:1342
 inactive_file:16413 unevictable:403 dirty:22 writeback:0 unstable:0
 free:358 slab:1237 mapped:928 pagetables:162 bounce:0
Normal free:1460kB min:1440kB low:1800kB high:2160kB active_anon:1376kB inactive_anon:5368kB active_file:16868kB inactive_file:65624kB unevictable:1612kB present:130048kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0
Normal: 355*4kB 3*8kB 1*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1460kB
21523 total pagecache pages
169 pages in swap cache
Swap cache stats: add 9640, delete 9471, find 25380/27307
Free swap  = 201260kB
Total swap = 204792kB
32768 pages of RAM
603 free pages
1401 reserved pages
1238 slab pages
5911 pages shared
169 pages swap cached

What does it mean? Is there a problem with my DC01?

[HWTest]

I'm still suffering this problem.
When I ignore this problem, meteohub hangs in a couple of days and I loose data.
I Googled a little bit and found out, that it is a low memory/memory fragmentation problem (http://www.cyberciti.biz/faq/linux-page ... lure-erro/).

Increasing the amount of memory is obviously not possible on the DC01.

Should I try to change the values of vm.lower_zone_protection and vm.vfs_cache_pressure as recommended in the linked article?
Any recommended values?

Now I know, why I didn't encounter this problem on Sheeva - 512MB RAM vs 125MB on the DC01.

Here is the log from toady if it helps ...?

Code: Select all

meteohtml.cgi: page allocation failure. order:1, mode:0x20
[<c002d3f8>] (unwind_backtrace+0x0/0xe0) from [<c007c068>] (__alloc_pages_nodemask+0x460/0x550)
[<c007c068>] (__alloc_pages_nodemask+0x460/0x550) from [<c033b194>] (cache_alloc_refill+0x2cc/0x538)
[<c033b194>] (cache_alloc_refill+0x2cc/0x538) from [<c009adb4>] (kmem_cache_alloc+0x8c/0xb0)
[<c009adb4>] (kmem_cache_alloc+0x8c/0xb0) from [<c0276694>] (sk_prot_alloc.isra.22+0x24/0xec)
[<c0276694>] (sk_prot_alloc.isra.22+0x24/0xec) from [<c02767d8>] (sk_clone+0x14/0x264)
[<c02767d8>] (sk_clone+0x14/0x264) from [<c02a7078>] (inet_csk_clone+0x10/0x84)
[<c02a7078>] (inet_csk_clone+0x10/0x84) from [<c02bcc6c>] (tcp_create_openreq_child+0x14/0x320)
[<c02bcc6c>] (tcp_create_openreq_child+0x14/0x320) from [<c02bac60>] (tcp_v4_syn_recv_sock+0x3c/0x1c4)
[<c02bac60>] (tcp_v4_syn_recv_sock+0x3c/0x1c4) from [<c02bca7c>] (tcp_check_req+0x26c/0x448)
[<c02bca7c>] (tcp_check_req+0x26c/0x448) from [<c02bab20>] (tcp_v4_do_rcv+0x120/0x224)
[<c02bab20>] (tcp_v4_do_rcv+0x120/0x224) from [<c02bc5ac>] (tcp_v4_rcv+0x698/0x7a4)
[<c02bc5ac>] (tcp_v4_rcv+0x698/0x7a4) from [<c029d688>] (ip_local_deliver_finish+0x6c/0x1f4)
[<c029d688>] (ip_local_deliver_finish+0x6c/0x1f4) from [<c029d380>] (ip_rcv_finish+0x114/0x3b0)
[<c029d380>] (ip_rcv_finish+0x114/0x3b0) from [<c02808b4>] (netif_receive_skb+0x290/0x314)
[<c02808b4>] (netif_receive_skb+0x290/0x314) from [<c0281ddc>] (process_backlog+0x84/0xcc)
[<c0281ddc>] (process_backlog+0x84/0xcc) from [<c0281f04>] (net_rx_action+0xb0/0x194)
[<c0281f04>] (net_rx_action+0xb0/0x194) from [<c004d080>] (__do_softirq+0xa8/0x140)
[<c004d080>] (__do_softirq+0xa8/0x140) from [<c004d164>] (do_softirq+0x4c/0x58)
[<c004d164>] (do_softirq+0x4c/0x58) from [<c004d204>] (local_bh_enable_ip+0x94/0xa0)
[<c004d204>] (local_bh_enable_ip+0x94/0xa0) from [<c02c85e8>] (inet_stream_connect+0x16c/0x28c)
[<c02c85e8>] (inet_stream_connect+0x16c/0x28c) from [<c0273b78>] (sys_connect+0x68/0x90)
[<c0273b78>] (sys_connect+0x68/0x90) from [<c00289c0>] (ret_fast_syscall+0x0/0x2c)
Mem-info:
Normal per-cpu:
CPU    0: hi:   42, btch:   7 usd:  25
CPU    1: hi:   42, btch:   7 usd:   1
Active_anon:1062 active_file:9509 inactive_anon:1524
 inactive_file:10098 unevictable:403 dirty:1 writeback:0 unstable:0
 free:395 slab:1220 mapped:890 pagetables:208 bounce:0
Normal free:1580kB min:1440kB low:1800kB high:2160kB active_anon:4248kB inactive_anon:6096kB active_file:38036kB inactive_file:40392kB unevictable:1612kB present:130048kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0
Normal: 327*4kB 6*8kB 14*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1580kB
20756 total pagecache pages
369 pages in swap cache
Swap cache stats: add 84633, delete 84264, find 237685/256154
Free swap  = 200724kB
Total swap = 204792kB
32768 pages of RAM
662 free pages
1400 reserved pages
1220 slab pages
7479 pages shared
369 pages swap cached