Closed morgabra closed 9 years ago
FYI @gregkh
I'll watch this, but networking bugs should be reported to the netdev@vger.kernel.org mailing list, not on bugzilla, if they want to get fixed....
@morgabra has someone filed this on netdev?
Someone forwarded the bugzilla message to netdev a long time ago but as far as I can tell that was all and it certainly didn't get a response: http://thread.gmane.org/gmane.linux.network/326672
A possibly related report: http://sourceforge.net/p/e1000/bugs/434/
The user in that case resolved it by a random shuffle of hardware but re-raising the issue with Intel folks is likely the best next step.
Please reopen if this is still an issue with the 4.2+ kernels.
Hi, hope this is the right place to mention this but I'm getting this error on the latest coreos alpha channel build within a virtualbox guest on a Windows host. sudo ethtool -K eth1 sg off
fixed it for me. Glad to add any more info, just let me know.
Mar 17 16:57:28 core01 kernel: ------------[ cut here ]------------
Mar 17 16:57:28 core01 kernel: WARNING: CPU: 0 PID: 328 at ../source/net/core/dev.c:2422 skb_warn_ba d_offload+0xd3/0x120()
Mar 17 16:57:28 core01 kernel: virtio_net: caps=(0x0000080400174a29, 0x0000000000000000) len=1654 da ta_len=1456 gso_size=1480 gso_type=2 ip_summed=0
Mar 17 16:57:28 core01 kernel: Modules linked in: xt_conntrack ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 xt_addrtype iptable_filter nf_nat nf_connt rack br_netfilter bridge stp llc overlay nfsv3 nfs_acl nfs lockd grace sunrpc fscache nls_ascii nls_ cp437 vfat fat ext4 crc16 mbcache jbd2 sd_mod crc32c_intel jitterentropy_rng hmac drbg virtio_net mo usedev ata_piix libata aesni_intel aes_x86_64 glue_helper lrw gf128mul ablk_helper cryptd scsi_mod p smouse evdev i2c_piix4 i2c_core virtio_pci virtio_ring virtio acpi_cpufreq tpm_tis tpm button sch_fq _codel ip_tables autofs4
Mar 17 16:57:28 core01 kernel: CPU: 0 PID: 328 Comm: kworker/0:1H Tainted: G W 4.4.4-co reos #2
Mar 17 16:57:28 core01 kernel: Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12 /01/2006
Mar 17 16:57:28 core01 kernel: Workqueue: rpciod rpc_wake_up_status [sunrpc]
Mar 17 16:57:28 core01 kernel: 0000000000000286 000000007e9d210b ffff880016cc37f0 ffffffff952d6973
Mar 17 16:57:28 core01 kernel: ffff880016cc3838 ffffffff95819051 ffff880016cc3828 ffffffff950746b6
Mar 17 16:57:28 core01 kernel: ffff88001706bb00 ffff880017355000 0000000000000002 ffff880017355000
Mar 17 16:57:28 core01 kernel: Call Trace:
Mar 17 16:57:28 core01 kernel: [<ffffffff952d6973>] dump_stack+0x63/0x90
Mar 17 16:57:28 core01 kernel: [<ffffffff950746b6>] warn_slowpath_common+0x86/0xc0
Mar 17 16:57:28 core01 kernel: [<ffffffff9507474c>] warn_slowpath_fmt+0x5c/0x80
Mar 17 16:57:28 core01 kernel: [<ffffffff952dc766>] ? ___ratelimit+0x86/0xe0
Mar 17 16:57:28 core01 kernel: [<ffffffff95455203>] skb_warn_bad_offload+0xd3/0x120
Mar 17 16:57:28 core01 kernel: [<ffffffff954596ce>] __skb_gso_segment+0x7e/0xd0
Mar 17 16:57:28 core01 kernel: [<ffffffff95459a6f>] validate_xmit_skb.isra.99.part.100+0x12f/0x2b0
Mar 17 16:57:28 core01 kernel: [<ffffffff9545a02b>] validate_xmit_skb_list+0x3b/0x60
Mar 17 16:57:28 core01 kernel: [<ffffffff9547d851>] sch_direct_xmit+0x171/0x210
Mar 17 16:57:28 core01 kernel: [<ffffffff9545a383>] __dev_queue_xmit+0x253/0x590
Mar 17 16:57:28 core01 kernel: [<ffffffff9545a6d0>] dev_queue_xmit+0x10/0x20
Mar 17 16:57:28 core01 kernel: [<ffffffff9549ae06>] ip_finish_output2+0x276/0x310
Mar 17 16:57:28 core01 kernel: [<ffffffff9549c609>] ip_finish_output+0x139/0x1f0
Mar 17 16:57:28 core01 kernel: [<ffffffff9548da43>] ? nf_hook_slow+0x73/0xd0
Mar 17 16:57:28 core01 kernel: [<ffffffff9549d03e>] ip_output+0x6e/0xe0
Mar 17 16:57:28 core01 kernel: [<ffffffff9549c702>] ? __ip_local_out+0x42/0x100
Mar 17 16:57:28 core01 kernel: [<ffffffff9549c4d0>] ? ip_fragment.constprop.50+0x80/0x80
Mar 17 16:57:28 core01 kernel: [<ffffffff9549c7f5>] ip_local_out+0x35/0x40
Mar 17 16:57:28 core01 kernel: [<ffffffff9549d9f9>] ip_send_skb+0x19/0x40
Mar 17 16:57:28 core01 kernel: [<ffffffff954c470d>] udp_send_skb+0x16d/0x280
Mar 17 16:57:28 core01 kernel: [<ffffffff954c485e>] udp_push_pending_frames+0x3e/0x60
Mar 17 16:57:28 core01 kernel: [<ffffffff954c5e41>] udp_sendpage+0x121/0x1a0
Mar 17 16:57:28 core01 kernel: [<ffffffff9543b648>] ? sock_sendmsg+0x38/0x50
Mar 17 16:57:28 core01 kernel: [<ffffffff9543b77b>] ? kernel_sendmsg+0x2b/0x30
Mar 17 16:57:28 core01 kernel: [<ffffffffc05de458>] ? csum_partial_copy_to_xdr+0x12d8/0x4630 [sunrp c]
Mar 17 16:57:28 core01 kernel: [<ffffffff954d2ad4>] inet_sendpage+0x74/0xd0
Mar 17 16:57:28 core01 kernel: [<ffffffff954d2a60>] ? inet_sendmsg+0xa0/0xa0
Mar 17 16:57:28 core01 kernel: [<ffffffffc05de5f5>] csum_partial_copy_to_xdr+0x1475/0x4630 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05de76f>] csum_partial_copy_to_xdr+0x15ef/0x4630 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05dc5a6>] xprt_transmit+0x66/0x350 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05d86d9>] rpc_peeraddr+0x1369/0x14d0 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05d8520>] ? rpc_peeraddr+0x11b0/0x14d0 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05d8520>] ? rpc_peeraddr+0x11b0/0x14d0 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05e33b5>] rpc_wake_up_status+0x2a5/0x6b0 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffffc05e37b5>] rpc_wake_up_status+0x6a5/0x6b0 [sunrpc]
Mar 17 16:57:28 core01 kernel: [<ffffffff9508caf5>] process_one_work+0x155/0x410
Mar 17 16:57:28 core01 kernel: [<ffffffff9508d59e>] worker_thread+0x4e/0x470
Mar 17 16:57:28 core01 kernel: [<ffffffff9508d550>] ? rescuer_thread+0x320/0x320
Mar 17 16:57:28 core01 kernel: [<ffffffff9508d550>] ? rescuer_thread+0x320/0x320
Mar 17 16:57:28 core01 kernel: [<ffffffff95092928>] kthread+0xd8/0xf0
Mar 17 16:57:28 core01 kernel: [<ffffffff95092850>] ? kthread_park+0x60/0x60
Mar 17 16:57:28 core01 kernel: [<ffffffff95558c0f>] ret_from_fork+0x3f/0x70
Mar 17 16:57:28 core01 kernel: [<ffffffff95092850>] ? kthread_park+0x60/0x60
Mar 17 16:57:28 core01 kernel: ---[ end trace 106ba202ee0e666d ]---
I am having same issue i have off LRO/GRO on NIC but still causing this error
[ 8162.302876] ------------[ cut here ]------------
[ 8162.302902] WARNING: CPU: 2 PID: 0 at net/core/dev.c:2573 skb_warn_bad_offload+0xcd/0xda
[ 8162.302906] : caps=(0x00003e2007db78e9, 0x0000000000000000) len=5844 data_len=4344 gso_size=1448 gso_type=1 ip_summed=3
[ 8162.302908] Modules linked in: btrfs raid6_pq xor loop iptable_raw veth bonding ip6table_mangle xt_CHECKSUM sb_edac intel_powerclamp coretemp intel_rapl iosf_mbi kvm_intel kvm irqbypass crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd vhost_net tun vhost ipmi_ssif macvtap macvlan iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_vs iTCO_wdt iptable_nat nf_conntrack_ipv4 iTCO_vendor_support nf_defrag_ipv4 nf_nat_ipv4 ipmi_si iptable_mangle iptable_filter ipt_REJECT nf_reject_ipv4 ipmi_devintf ipt_MASQUERADE pcspkr sg lpc_ich nf_nat_masquerade_ipv4 hpwdt hpilo ipmi_msghandler video nf_nat acpi_power_meter wmi ioatdma shpchp nf_conntrack pcc_cpufreq ip6table_filter ip6_tables ebtables dm_snapshot dm_bufio dm_multipath br_netfilter bridge 8021q garp mrp stp
[ 8162.302967] llc ip_tables xfs libcrc32c sd_mod crc_t10dif crct10dif_generic mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm ixgbe drm crct10dif_pclmul crct10dif_common mdio crc32c_intel hpsa serio_raw ptp pps_core i2c_core scsi_transport_sas dca dm_mirror dm_region_hash dm_log dm_mod
[ 8162.302994] CPU: 2 PID: 0 Comm: swapper/2 Kdump: loaded Tainted: G W ------------ 3.10.0-862.3.3.el7.x86_64 #1
[ 8162.302999] Hardware name: HP ProLiant BL460c Gen8, BIOS I31 06/01/2015
[ 8162.303001] Call Trace:
[ 8162.303005]
You are using a very old and obsolete kernel version (3.10), please go get support from the vendor that is forcing you to do that, you are paying them money for this :)
its CentOS 7.5 :(
Does this message means its impacting network performance or its just reporting stuff and nothing to do with performance?
If it is Centos, go ask on their developer forums, there is nothing that anyone here can do about this sorry.
You might want to use a more modern kernel, that one is very old and obsolete :(
I think you are right, i should just install kernel 4.x because there is no downside to do that i believe.
Finally i have upgraded my kernel to 4.17.12-1.el7.elrepo.x86_64 and i am still see same stack error which i posted earlier, I believe 4.17.x is the latest kernel so now where and what is the wrong here?
Locking this thread as off-topic for Container Linux. If you're seeing kernel issues with a current version of Container Linux, please report a new issue.
https://bugzilla.kernel.org/show_bug.cgi?id=82471
This affects alpha/beta kernel versions and significantly hurts network performance/reliability.
Disabling scatter/gather offloading appears to stop the pain:
sudo ethtool -K bond0.101 sg off
Is this a useful thing to report here?