Hello everybody, I have some puzzling issue. I see this in dmesg:
ksoftirqd/1: page allocation failure. order:1, mode:0x20 Pid: 9, comm: ksoftirqd/1 Not tainted 2.6.32-220.4.1.el6.x86_64 #1 Call Trace: <IRQ> [<ffffffff81123daf>] ? __alloc_pages_nodemask+0x77f/0x940 [<ffffffff81061f00>] ? enqueue_entity+0x390/0x420 [<ffffffff8115dc62>] ? kmem_getpages+0x62/0x170 [<ffffffff8115e87a>] ? fallback_alloc+0x1ba/0x270 [<ffffffff8115e5f9>] ? ____cache_alloc_node+0x99/0x160 [<ffffffff8115f3db>] ? kmem_cache_alloc+0x11b/0x190 [<ffffffff8141f4b8>] ? sk_prot_alloc+0x48/0x1c0 [<ffffffff8141f742>] ? sk_clone+0x22/0x2e0 [<ffffffff8146c9b6>] ? inet_csk_clone+0x16/0xd0 [<ffffffff81485883>] ? tcp_create_openreq_child+0x23/0x450 [<ffffffff8148326d>] ? tcp_v4_syn_recv_sock+0x4d/0x2a0 [<ffffffff81485641>] ? tcp_check_req+0x201/0x420 [<ffffffffa0442762>] ? nf_ct_deliver_cached_events+0xa2/0xe2 [nf_conntrack] [<ffffffff81482c8b>] ? tcp_v4_do_rcv+0x35b/0x430 [<ffffffffa04aa557>] ? ipv4_confirm+0x87/0x1d0 [nf_conntrack_ipv4] [<ffffffff81484401>] ? tcp_v4_rcv+0x4e1/0x860 [<ffffffff814620b0>] ? ip_local_deliver_finish+0x0/0x2d0 [<ffffffff8146218d>] ? ip_local_deliver_finish+0xdd/0x2d0 [<ffffffff81462418>] ? ip_local_deliver+0x98/0xa0 [<ffffffff814618dd>] ? ip_rcv_finish+0x12d/0x440 [<ffffffff81461e65>] ? ip_rcv+0x275/0x350 [<ffffffffa057c15b>] ? ingress_enqueue+0x2b/0x94 [sch_ingress] [<ffffffff8142be6b>] ? __netif_receive_skb+0x49b/0x6e0 [<ffffffff8108b789>] ? delayed_work_timer_fn+0x39/0x50 [<ffffffff8142c14a>] ? process_backlog+0x9a/0x100 [<ffffffff814307c3>] ? net_rx_action+0x103/0x2f0 [<ffffffff81072001>] ? __do_softirq+0xc1/0x1d0 [<ffffffff8100c24c>] ? call_softirq+0x1c/0x30 [<ffffffff8100c24c>] ? call_softirq+0x1c/0x30 <EOI> [<ffffffff8100de85>] ? do_softirq+0x65/0xa0 [<ffffffff81071b60>] ? ksoftirqd+0x80/0x110 [<ffffffff81071ae0>] ? ksoftirqd+0x0/0x110 [<ffffffff81090726>] ? kthread+0x96/0xa0 [<ffffffff8100c14a>] ? child_rip+0xa/0x20 [<ffffffff81090690>] ? kthread+0x0/0xa0 [<ffffffff8100c140>] ? child_rip+0x0/0x20 I googled a bit for similar bugs, and, well, I just find it strange to get these on my system. These can be typical on systems pressured for kernel memory, but I find it hard to believe that the system in question has it. It's SL6.2 running 2.6.32-220.4.1.el6.x86_64 kernel, with 8GB RAM and vm.min_free_kbytes = 131072 sysctl parameter, typically 150-250 MB are always free (and like 4 GB cache+buffers). slabtop in the very same session (nothing bad seems to happened after this message) shows Active / Total Objects (% used) : 1533015 / 2020657 (75.9%) Active / Total Slabs (% used) : 68451 / 68477 (100.0%) Active / Total Caches (% used) : 120 / 202 (59.4%) Active / Total Size (% used) : 212940.66K / 274324.79K (77.6%) Minimum / Average / Maximum Object : 0.02K / 0.14K / 4096.00K I wonder if anyone knows what might be going on and can suggest any tweak for solving this. While this isn't a problem per se, I fear that it might be a symptom of something worse. This is what /proc/meminfo looks like (having swapped stuff isn't a problem, it's just something that wasn't used in ages that was thrown there): MemTotal: 7854796 kB MemFree: 231400 kB Buffers: 182564 kB Cached: 3433284 kB SwapCached: 197872 kB Active: 4365836 kB Inactive: 2800612 kB Active(anon): 2686648 kB Inactive(anon): 877976 kB Active(file): 1679188 kB Inactive(file): 1922636 kB Unevictable: 0 kB Mlocked: 0 kB SwapTotal: 10485744 kB SwapFree: 9942660 kB Dirty: 328 kB Writeback: 0 kB AnonPages: 3467456 kB Mapped: 33580 kB Shmem: 13880 kB Slab: 298616 kB SReclaimable: 129076 kB SUnreclaim: 169540 kB KernelStack: 5976 kB PageTables: 30128 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 14413140 kB Committed_AS: 3952728 kB VmallocTotal: 34359738367 kB VmallocUsed: 50208 kB VmallocChunk: 34359677904 kB HardwareCorrupted: 0 kB AnonHugePages: 2140160 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 6912 kB DirectMap2M: 8380416 kB -- Vladimir