jonathanross
Tux's lil' helper
Joined: 19 Mar 2005
Posts: 77
Location: Planet Earth
|
 Posted: Mon Sep 22, 2008 8:07 pm Post subject: Call Trace errors in dmesg: page allocation failures |
 |
|
Hi All,
Last week I suddenly started getting these in dmesg on a machine build no more than three weeks old. I've confirmed the swapper, rsync and kswapd0 were all during an emerge --sync.
| Code: |
[47833.518905] qmail-smtpd: page allocation failure. order:3, mode:0x4020
[88192.886463] swapper: page allocation failure. order:3, mode:0x4020
[88188.386556] rsync: page allocation failure. order:3, mode:0x4020
[47833.525012] kswapd0: page allocation failure. order:3, mode:0x4020
|
Amongst the repetitive debug data this seems to be the common theme:
| Code: |
[47833.518940] Call Trace:
[47833.518954] [00000000004b7574] __slab_alloc+0x1b4/0x5f4
[47833.518997] [00000000004b95d0] __kmalloc_track_caller+0x98/0xf0
[47833.519021] [0000000000625a38] __alloc_skb+0x5c/0x108
[47833.519051] [00000000100003f0] tulip_interrupt+0x2a8/0xd94 [tulip]
[47833.519111] [000000000048fd58] handle_IRQ_event+0x34/0x74
[47833.519138] [000000000049145c] handle_fasteoi_irq+0xe0/0x13c
[47833.519163] [000000000042db94] handler_irq+0x8c/0xb4
[47833.519197] [00000000004208b4] tl0_irq5+0x1c/0x20
[47833.519223] [00000000004c84fc] do_select+0x3a8/0x3bc
[47833.519247] [00000000004bbb60] vfs_read+0x90/0x10c
[47833.519272] [00000000004bbe9c] sys_read+0x34/0x60
[47833.519292] [00000000004062d4] linux_sparc_syscall32+0x3c/0x40
[47833.519324] [0000000000012368] 0x12370
|
It's a SPARC Netra X1 box and across the few of those boxes that I admin they all have varying NIC errors with the tulip drivers depending on their speed/duplex settings but nothing causing loss of service.
Am I right in thinking that these errors are simply complaints about the tulip drivers and the NICs ? There were apparently only 8 errors and 8 packets dropped during the emerge --sync according to ifconfig.
The annoying thing is that this is a fresh build with the latest kernel and I've based disk images around this build:
Linux Loopy 2.6.25-gentoo-r7 #4 SMP Sun Sep 7 19:22:55 BST 2008 sparc64 sun4u TI UltraSparc IIe (Hummingbird) GNU/Linux
I can post all the dmesg output if someone wants a closer look but I thought I'd save you the eyestrain first off 
I suppose my main question is ... is it harmless or something to worry about for machine stability ? These are all production Servers.
Any help appreciated,
JR  |
|
Kernel & Hardware
