PAX: size overflow detected in function tbf_enqueue net/sche
Posted: Thu Mar 24, 2016 3:59 am
Hello PAX/grsec,
when trying to rate limit bridged veth interface with tc qdisc and when generating some traffic from that interface, I'm always getting this size overflow:
Kernel is 4.4.5, with only pax-linux-4.4.5-test10.patch applied and with ported pax_size_overflow_report functionality from mainstream grsec patch.
It's a VPS machine running under KVM. If you need more info, please let me know.
Many thanks for helping me solve this!
Regards,
A.
when trying to rate limit bridged veth interface with tc qdisc and when generating some traffic from that interface, I'm always getting this size overflow:
- Code: Select all
[ 1963.536160] PAX: size overflow detected in function tbf_enqueue net/sched/sch_tbf.c:186 cicus.188_148 min, count: 36, decl: qdisc_tree_decrease_qlen; num: 2; context: fndecl;
[ 1963.537414] CPU: 0 PID: 3 Comm: ksoftirqd/0 Tainted: P O 4.4.5 #7
[ 1963.538013] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 1963.538348] 0000000000000000 3c9dca40a898b545 0000000000000286 0000000000000000
[ 1963.538994] ffffffff9b4956a0 ffffffff9bdbd8e3 3c9dca40a898b545 ffffffff9bdbd8e3
[ 1963.540058] 00000000000000ba ffffffff9b1eb295 ffff880032ebd040 ffff8800296af200
[ 1963.540058] Call Trace:
[ 1963.540058] [<ffffffff9b4956a0>] ? dump_stack+0x70/0xc0
[ 1963.540058] [<ffffffff9b1eb295>] ? report_size_overflow+0x65/0x80
[ 1963.540058] [<ffffffff9b795dfd>] ? tbf_enqueue+0x2cd/0x300
[ 1963.540058] [<ffffffff9b7489a0>] ? __dev_queue_xmit+0x300/0x620
[ 1963.540058] [<ffffffff9b9574a3>] ? br_dev_queue_push_xmit+0x93/0x2f0
[ 1963.540058] [<ffffffff9b957978>] ? br_forward_finish+0x38/0xb0
[ 1963.540058] [<ffffffff9b957410>] ? br_fdb_external_learn_del+0xf0/0xf0
[ 1963.540058] [<ffffffff9b957779>] ? __br_deliver+0x79/0x160
[ 1963.540058] [<ffffffff9b957940>] ? br_forward_finish.part.8+0x30/0x30
[ 1963.540058] [<ffffffff9b9549e7>] ? br_dev_xmit+0x227/0x310
[ 1963.540058] [<ffffffff9b748355>] ? dev_hard_start_xmit+0x315/0x520
[ 1963.540058] [<ffffffff9b747bbc>] ? validate_xmit_skb.isra.117.part.118+0x1c/0x4a0
[ 1963.540058] [<ffffffff9b748be9>] ? __dev_queue_xmit+0x549/0x620
[ 1963.540058] [<ffffffff9b7593bc>] ? neigh_resolve_output+0xfc/0x1d0
[ 1963.540058] [<ffffffff9b856200>] ? ip_forward_options+0x1d0/0x1d0
[ 1963.540058] [<ffffffff9b8564a6>] ? ip_finish_output2+0x1d6/0x340
[ 1963.540058] [<ffffffff9b8596ca>] ? ip_output+0x7a/0x100
[ 1963.540058] [<ffffffff9b858920>] ? ip_fragment.constprop.40+0xa0/0xa0
[ 1963.540058] [<ffffffff9b854963>] ? ip_forward+0x3e3/0x4e0
[ 1963.540058] [<ffffffff9b854510>] ? ip_frag_mem+0x50/0x50
[ 1963.540058] [<ffffffff9b852614>] ? ip_rcv+0x284/0x5c0
[ 1963.540058] [<ffffffff9b851e40>] ? ip_local_deliver_finish+0x260/0x260
[ 1963.540058] [<ffffffff9b741e00>] ? __netif_receive_skb_core+0x360/0xc40
[ 1963.540058] [<ffffffff9b745ce4>] ? napi_gro_receive+0x94/0xb0
[ 1963.540058] [<ffffffff9b611101>] ? virtnet_receive+0x701/0x9e0
[ 1963.540058] [<ffffffff9b74504a>] ? netif_receive_skb_internal+0x2a/0xa0
[ 1963.540058] [<ffffffff9b7451d0>] ? napi_gro_flush+0x50/0x70
[ 1963.540058] [<ffffffff9b745253>] ? napi_complete_done+0x63/0xb0
[ 1963.540058] [<ffffffff9b6114f0>] ? virtnet_poll+0x40/0x80
[ 1963.540058] [<ffffffff9b746962>] ? net_rx_action+0x332/0x500
[ 1963.540058] [<ffffffff9b0e3d8d>] ? __do_softirq+0xfd/0x200
[ 1963.540058] [<ffffffff9b0e3eb5>] ? run_ksoftirqd+0x25/0x40
[ 1963.540058] [<ffffffff9b104a01>] ? smpboot_thread_fn+0x161/0x1e0
[ 1963.540058] [<ffffffff9b1048a0>] ? sort_range+0x30/0x30
[ 1963.540058] [<ffffffff9b1018a7>] ? kthread+0xe7/0x100
[ 1963.540058] [<ffffffff9b1017c0>] ? kthread_create_on_node+0x1a0/0x1a0
[ 1963.540058] [<ffffffff9b98817e>] ? ret_from_fork+0x3e/0x70
[ 1963.540058] [<ffffffff9b1017c0>] ? kthread_create_on_node+0x1a0/0x1a0
Kernel is 4.4.5, with only pax-linux-4.4.5-test10.patch applied and with ported pax_size_overflow_report functionality from mainstream grsec patch.
It's a VPS machine running under KVM. If you need more info, please let me know.
Many thanks for helping me solve this!
Regards,
A.