Hi!
Just a small update: midcom-cron is now disabled on the server for two
weeks, but the apache problems still occur, albeit less regular thatn
they used to. For example, today I had three or for entries like this in
the syslog:
Jul 3 17:46:25 h9142 kernel: printk: 1376 messages suppressed.
Jul 3 17:46:27 h9142 kernel: oom-killer: gfp_mask=0x201d2, order=0
Jul 3 17:46:30 h9142 kernel: [<c014457b>] out_of_memory+0x25/0x13e
Jul 3 17:46:31 h9142 kernel: [<c01459e8>] __alloc_pages+0x1f5/0x275
Jul 3 17:46:31 h9142 kernel: [<c0146b97>]
__do_page_cache_readahead+0xc8/0x1e8
Jul 3 17:46:31 h9142 kernel: [<c0280244>] io_schedule+0x26/0x30
Jul 3 17:46:31 h9142 kernel: [<c0280460>] __wait_on_bit_lock+0x4b/0x52
Jul 3 17:46:31 h9142 kernel: [<c014177c>] __lock_page+0x51/0x57
Jul 3 17:46:31 h9142 kernel: [<c0143c8c>] filemap_nopage+0x15b/0x313
Jul 3 17:46:31 h9142 kernel: [<c014c154>] __handle_mm_fault+0x26e/0x740
Jul 3 17:46:31 h9142 kernel: [<c01154b6>] do_page_fault+0x18a/0x481
Jul 3 17:46:31 h9142 kernel: [<c011532c>] do_page_fault+0x0/0x481
Jul 3 17:46:31 h9142 kernel: [<c01037f9>] error_code+0x39/0x40
Jul 3 17:46:31 h9142 kernel: Mem-info:
Jul 3 17:46:31 h9142 kernel: DMA per-cpu:
Jul 3 17:46:31 h9142 kernel: cpu 0 hot: high 0, batch 1 used:0
Jul 3 17:46:31 h9142 kernel: cpu 0 cold: high 0, batch 1 used:0
Jul 3 17:46:31 h9142 kernel: DMA32 per-cpu: empty
Jul 3 17:46:31 h9142 kernel: Normal per-cpu:
Jul 3 17:46:31 h9142 kernel: cpu 0 hot: high 186, batch 31 used:79
Jul 3 17:46:31 h9142 kernel: cpu 0 cold: high 62, batch 15 used:12
Jul 3 17:46:31 h9142 kernel: HighMem per-cpu: empty
Jul 3 17:46:31 h9142 kernel: Free pages: 5736kB (0kB HighMem)
Jul 3 17:46:31 h9142 kernel: Active:62055 inactive:61835 dirty:0
writeback:0 unstable:0 free:1434 slab:1870 mapped:12 pagetables:823
Jul 3 17:46:31 h9142 kernel: DMA free:2068kB min:88kB low:108kB
high:132kB active:6080kB inactive:5240kB present:16384kB
pages_scanned:12782 all_unreclaimable? y
es
Jul 3 17:46:31 h9142 kernel: lowmem_reserve[]: 0 0 495 495
Jul 3 17:46:31 h9142 kernel: DMA32 free:0kB min:0kB low:0kB high:0kB
active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
Jul 3 17:46:31 h9142 kernel: lowmem_reserve[]: 0 0 495 495
Jul 3 17:46:31 h9142 kernel: Normal free:3668kB min:2804kB low:3504kB
high:4204kB active:242140kB inactive:242100kB present:507840kB
pages_scanned:410287 all_unr
eclaimable? no
Jul 3 17:46:31 h9142 kernel: lowmem_reserve[]: 0 0 0 0
Jul 3 17:46:31 h9142 kernel: HighMem free:0kB min:128kB low:128kB
high:128kB active:0kB inactive:0kB present:0kB pages_scanned:0
all_unreclaimable? no
Jul 3 17:46:31 h9142 kernel: lowmem_reserve[]: 0 0 0 0
Jul 3 17:46:31 h9142 kernel: DMA: 1*4kB 2*8kB 0*16kB 0*32kB 2*64kB
1*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2068kB
Jul 3 17:46:31 h9142 kernel: DMA32: empty
Jul 3 17:46:31 h9142 kernel: Normal: 255*4kB 1*8kB 3*16kB 1*32kB 0*64kB
6*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 3668kB
Jul 3 17:46:31 h9142 kernel: HighMem: empty
Jul 3 17:46:31 h9142 kernel: Swap cache: add 330114742, delete
330114742, find 38416427/89413861, race 1105+5119
Jul 3 17:46:31 h9142 kernel: Free swap = 0kB
Jul 3 17:46:31 h9142 kernel: Total swap = 514072kB
Jul 3 17:46:31 h9142 kernel: Free swap: 0kB
Jul 3 17:46:31 h9142 kernel: 131056 pages of RAM
Jul 3 17:46:31 h9142 kernel: 0 pages of HIGHMEM
Jul 3 17:46:31 h9142 kernel: 1839 reserved pages
Jul 3 17:46:31 h9142 kernel: 1429 pages shared
Jul 3 17:46:31 h9142 kernel: 0 pages swap cached
Jul 3 17:46:31 h9142 kernel: 0 pages dirty
Jul 3 17:46:31 h9142 kernel: 0 pages writeback
Jul 3 17:46:31 h9142 kernel: 12 pages mapped
Jul 3 17:46:31 h9142 kernel: 1870 pages slab
Jul 3 17:46:31 h9142 kernel: 823 pages pagetables
Jul 3 17:46:31 h9142 kernel: Out of Memory: Kill process 28244
(apache2) score 16676 and children.
Jul 3 17:46:31 h9142 kernel: Out of memory: Killed process 28244 (apache2).
15 Minutes later, I couldn't even reach the machine with SSH and had to
do a hard reset. Is there anything I could do to track this problem
down? Like I said, with 1.8.2.2, I never had any problems, but with the
upgrade, it started.
Bye,
Andreas
Andreas Flack schrieb:
Hi!
I have strange performance problems since updating from 1.8.2.2 to 1.8.3. on a debian stable server. When I request a normal PHP page, everything works fine, but when I request a Midgard page, it takes something like 30 seconds to load and the system load as reported by uptime jumps from 0.01 to something like 5.00.
I'm not really sure how to debug this, but it really renders the server more or
less useless. Is there anything i could do about it or schould I just downgrade
to 1.8.2.2 again?
Bye,
Andreas
_______________________________________________
user mailing list
[email protected]
http://lists.midgard-project.org/mailman/listinfo/user
begin:vcard
fn:Andreas Flack
n:Flack;Andreas
org:CONTENT CONTROL Berlin GbR
adr;dom:;;Ebertystr. 20;Berlin;Berlin;10249
email;internet:[EMAIL PROTECTED]
tel;work:+49 (0) 177 565 19 13
tel;fax:+49 (0) 30 639 018 64
x-mozilla-html:FALSE
url:http://www.contentcontrol-berlin.de
version:2.1
end:vcard
_______________________________________________
user mailing list
[email protected]
http://lists.midgard-project.org/mailman/listinfo/user
