tags #416907 + moreinfo
thanks
Alexander Lazarevic wrote:
> Package: imapproxy
> Version: 1.2.4-10
> Severity: grave
> Justification: renders package unusable
>
>
> imapproxy runs and gets used a few days non-stop without any problems. Then I
> get out of memory errors and I can't connect to the machine anymore. Here is
> part of the syslog:
>
> Mar 31 03:21:26 renaud kernel: oom-killer: gfp_mask=0x200d2, order=0
> Mar 31 03:21:26 renaud kernel: [<c0144b2d>] out_of_memory+0x25/0x144
> Mar 31 03:21:26 renaud kernel: [<c0145fc3>] __alloc_pages+0x1f5/0x275
> Mar 31 03:21:26 renaud kernel: [<c0152155>] read_swap_cache_async+0x2f/0xb2
> Mar 31 03:21:26 renaud kernel: [<c014b616>] swapin_readahead+0x3a/0x58
> Mar 31 03:21:26 renaud kernel: [<c014c9a9>] __handle_mm_fault+0x4aa/0x705
> Mar 31 03:21:26 renaud kernel: [<c011554e>] do_page_fault+0x18a/0x46c
> Mar 31 03:21:27 renaud kernel: [<c01153c4>] do_page_fault+0x0/0x46c
> Mar 31 03:21:27 renaud kernel: [<c01037d5>] error_code+0x39/0x40
> Mar 31 03:21:27 renaud kernel: Mem-info:
> Mar 31 03:21:27 renaud kernel: DMA per-cpu:
> Mar 31 03:21:27 renaud kernel: cpu 0 hot: high 0, batch 1 used:0
> Mar 31 03:21:27 renaud kernel: cpu 0 cold: high 0, batch 1 used:0
> Mar 31 03:21:27 renaud kernel: DMA32 per-cpu: empty
> Mar 31 03:21:27 renaud kernel: Normal per-cpu:
> Mar 31 03:21:27 renaud kernel: cpu 0 hot: high 186, batch 31 used:5
> Mar 31 03:21:27 renaud kernel: cpu 0 cold: high 62, batch 15 used:51
> Mar 31 03:21:27 renaud kernel: HighMem per-cpu: empty
> Mar 31 03:21:27 renaud kernel: Free pages: 4532kB (0kB HighMem)
> Mar 31 03:21:27 renaud kernel: Active:761 inactive:1211 dirty:1 writeback:20
> unstable:0 free:1133 slab:91922 mapped:384 pagetables:312
> Mar 31 03:21:27 renaud kernel: DMA free:1596kB min:104kB low:128kB high:156kB
> active:316kB inactive:552kB present:16384kB pages_scanned:263
> all_unreclaimable? no
> Mar 31 03:21:27 renaud kernel: lowmem_reserve[]: 0 0 367 367
> Mar 31 03:21:27 renaud kernel: DMA32 free:0kB min:0kB low:0kB high:0kB
> active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
> Mar 31 03:21:27 renaud kernel: lowmem_reserve[]: 0 0 367 367
> Mar 31 03:21:27 renaud kernel: Normal free:2936kB min:2400kB low:3000kB
> high:3600kB active:2728kB inactive:4292kB present:376752kB pages_scanned:0
> all_unreclaimable? no
> Mar 31 03:21:27 renaud kernel: lowmem_reserve[]: 0 0 0 0
> Mar 31 03:21:27 renaud kernel: HighMem free:0kB min:128kB low:128kB
> high:128kB active:0kB inactive:0kB present:0kB pages_scanned:0
> all_unreclaimable? no
> Mar 31 03:21:27 renaud kernel: lowmem_reserve[]: 0 0 0 0
> Mar 31 03:21:27 renaud kernel: DMA: 5*4kB 3*8kB 5*16kB 0*32kB 1*64kB 1*128kB
> 1*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1596kB
> Mar 31 03:21:27 renaud kernel: DMA32: empty
> Mar 31 03:21:27 renaud kernel: Normal: 164*4kB 5*8kB 12*16kB 8*32kB 0*64kB
> 0*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2936kB
> Mar 31 03:21:27 renaud kernel: HighMem: empty
> Mar 31 03:21:27 renaud kernel: Swap cache: add 29672413, delete 29670997,
> find 2875197860/2883357192, race 5+764
> Mar 31 03:21:27 renaud kernel: Free swap = 918200kB
> Mar 31 03:21:27 renaud kernel: Total swap = 979956kB
> Mar 31 03:21:27 renaud kernel: Free swap: 918200kB
> Mar 31 03:21:27 renaud kernel: 98284 pages of RAM
> Mar 31 03:21:27 renaud kernel: 0 pages of HIGHMEM
> Mar 31 03:21:27 renaud kernel: 1604 reserved pages
> Mar 31 03:21:27 renaud kernel: 1069 pages shared
> Mar 31 03:21:27 renaud kernel: 1416 pages swap cached
> Mar 31 03:21:27 renaud kernel: 1 pages dirty
> Mar 31 03:21:27 renaud kernel: 0 pages writeback
> Mar 31 03:21:27 renaud kernel: 384 pages mapped
> Mar 31 03:21:27 renaud kernel: 91922 pages slab
> Mar 31 03:21:27 renaud kernel: 312 pages pagetables
> Mar 31 03:21:27 renaud kernel: Out of Memory: Kill process 21019 (imapproxyd)
> score 2872 and children.
> Mar 31 03:21:27 renaud kernel: Out of memory: Killed process 21019
> (imapproxyd).
>
I am not that familiar with OOM logs, but... are we sure that it was
imapproxy who was using that much memory ?
Besides, swap is almost not being used -- there is nothing within
imapproxy locking its pages in main memory, so I suspect this is not the
culprit :-S
However, any additional details will be welcome.
There are no substantial changes in imapproxy's code since 1.2.4-7 or so
(when IPv6 was added). This was several months ago, and I have been
using it in production without seeing any adverse effects.
J.L.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
