这是Linux内核或Sys Admin Guru的问题。
我从Qemu获得了此错误，试图使用3GB内存启动VM：
ioctl(KVM_CREATE_VM) failed: 12 Cannot allocate memory
failed to initialize KVM: Cannot allocate memory
据我所知，这可能是因为没有足够的内存或提交限制太低，但显然不是...倾倒caches可用5.9 GB，没有提交限制：
$ free -m
total used free shared buff/cache available
Mem: 7696 1287 135 139 6274 5973
Swap: 2892 525 2367
$ cat /proc/sys/vm/overcommit_memory
1
因此，我写了一个C 程序来依次分配更大的块。我发现它无法分配超过2.1 GB。（N.B.它被编译为64位。）这与Qemu不启动一致，但是为什么？
因此，我将其修改为写入内存。这导致一些缓存被倾倒，免费报告约2 GB：
：$ free -m
total used free shared buff/cache available
Mem: 7696 2988 288 143 4420 4268
Swap: 2892 525 2367
...当程序终止时：
$ free -m
total used free shared buff/cache available
Mem: 7696 1258 2253 147 4185 5994
Swap: 2892 525 2367
现在，我试图启动Qemu，并神奇地起作用！免费报告：
$ free -m
total used free shared buff/cache available
Mem: 7696 2438 4451 147 806 4834
Swap: 2892 530 2362
因此，当被问到一个大块时，似乎内核不愿意释放一些缓存，但愿意让他们走几步。发生了什么事？
我正在进行Debian测试：Linux-4.11.0-1-AMD64＃1 SMP Debian 4.11.6-1（2017-06-19）X86_64 GNU/Linux
我找到了一个解决方法，这表明问题是内存碎片。在最近的问题再次发生时，我运行了以下命令，该命令解决了问题：
sudo su -
echo 1 > /proc/sys/vm/compact_memory
相关帖子：https：//unix.stackexchange.com/questions/44481/defragging-ram-oom-failure/147860#147860
这是从压实之前的DMESG转储，它可能会显示有关问题的更多信息，但我无法理解：
[618172.910238] qemu-system-x86: page allocation failure: order:4, mode:0x16040c0(GFP_KERNEL|__GFP_COMP|__GFP_NOTRACK), nodemask=(null) [618172.910244] qemu-system-x86 cpuset=/ mems_allowed=0 [618172.910248] CPU: 1 PID: 19454 Comm: qemu-system-x86 Not tainted 4.13.0-1-amd64 #1 Debian 4.13.13-1 [618172.910249] Hardware name: System manufacturer System Product Name/P8Z68-V LX, BIOS 4105 07/01/2013 [618172.910250] Call Trace: [618172.910256] ? dump_stack+0x5c/0x85 [618172.910257] ? warn_alloc+0x114/0x1b0 [618172.910259] ? __alloc_pages_direct_compact+0x4a/0xf0 [618172.910261] ? __alloc_pages_slowpath+0xd57/0xd60 [618172.910261] ? __alloc_pages_slowpath+0xd57/0xd60 [618172.910263] ? __alloc_pages_nodemask+0x228/0x250 [618172.910266] ? cache_grow_begin+0x80/0x530 [618172.910267] ? cache_grow_begin+0x80/0x530 [618172.910269] ? fallback_alloc+0x161/0x200 [618172.910271] ? kmem_cache_alloc_trace+0x1c3/0x5a0 [618172.910292] ? kvm_dev_ioctl+0xb6/0x6b0 [kvm] [618172.910305] ? do_vfs_ioctl+0x9f/0x600 [618172.910306] ? SyS_ioctl+0x74/0x80 [618172.910308] ? system_call_fast_compare_end+0xc/0x97 [618172.910309] Mem-Info: [618172.910313] active_anon:337090 inactive_anon:152155 isolated_anon:0
active_file:140849 inactive_file:198001 isolated_file:0
unevictable:28 dirty:8 writeback:0 unstable:0
slab_reclaimable:773968 slab_unreclaimable:14206
mapped:112839 shmem:69965 pagetables:12329 bounce:0
free:320850 free_pcp:0 free_cma:0 [618172.910315] Node 0 active_anon:1348360kB inactive_anon:608620kB active_file:563396kB inactive_file:792004kB unevictable:112kB isolated(anon):0kB isolated(file):0kB mapped:451356kB dirty:32kB writeback:0kB shmem:279860kB shmem_thp: 0kB shmem_pmdmapped: 0kB anon_thp: 215040kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no [618172.910315] Node 0 DMA free:15900kB min:136kB low:168kB high:200kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15984kB managed:15900kB mlocked:0kB kernel_stack:0kB pagetables:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB [618172.910317] lowmem_reserve[]: 0 3200 7657 7657 7657 [618172.910319] Node 0 DMA32 free:1205068kB min:28184kB low:35228kB high:42272kB active_anon:462188kB inactive_anon:167656kB active_file:159796kB inactive_file:164356kB unevictable:0kB writepending:0kB present:3362060kB managed:3296492kB mlocked:0kB kernel_stack:976kB pagetables:4164kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB [618172.910321] lowmem_reserve[]: 0 0 4457 4457 4457 [618172.910323] Node 0 Normal free:62432kB min:39260kB low:49072kB high:58884kB active_anon:886172kB inactive_anon:440964kB active_file:403600kB inactive_file:627648kB unevictable:112kB writepending:32kB present:4708352kB managed:4568752kB mlocked:112kB kernel_stack:12672kB pagetables:45152kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB [618172.910324] lowmem_reserve[]: 0 0 0 0 0 [618172.910326] Node 0 DMA: 1*4kB (U) 1*8kB (U) 1*16kB (U) 0*32kB 2*64kB (U) 1*128kB (U) 1*256kB (U) 0*512kB 1*1024kB (U) 1*2048kB (M) 3*4096kB (M) = 15900kB [618172.910332] Node 0 DMA32: 59795*4kB (UME) 67824*8kB (UME) 21933*16kB (UME) 2275*32kB (UE) 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1205500kB [618172.910337] Node 0 Normal: 1460*4kB (UME) 1303*8kB (UME) 2698*16kB (UME) 104*32kB (UME) 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 62760kB [618172.910343] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB [618172.910343] 384298 total pagecache pages [618172.910344] 9802 pages in swap cache [618172.910345] Swap cache stats: add 1402693, delete 1392949, find 861777/1070392 [618172.910345] Free swap = 1537612kB [618172.910345] Total swap = 2962428kB [618172.910346] 2021599 pages RAM [618172.910346] 0 pages HighMem/MovableOnly [618172.910346] 51313 pages reserved [618172.910347] 0 pages hwpoisoned
您可以释放 buffer/cache 。
安全的方法是用root：
启动此命令#sync; echo 3 > /proc/sys/vm/drop_caches
运行echo 1 > /proc/sys/vm/compact_memory对我不起作用，即使我不是从云服务器运行VM，而是在本地服务器上运行。它需要我有一些权限（我在内核中读了类似配置的内容）。错误消息是Permission denied。
所以，我尝试停止在计算机上运行的软件，而当时我不需要。我尝试关闭Torrent应用Transmission和VOILà，我获得了足够的免费连续内存以运行我的VM。
另一个选择是重新启动机器，但是我正在等待一项非常长的任务来完成并且不想取消它，因此，重新启动并不是从上方获取错误的选项（Cannot allocate memory）。