Server occasional disables it's public network card

[Animoski]

Hello everyone,

I've been running and updating the Proxmox server software I run to make sure it's as bug free as possible, yet occasionally I find the server inaccessible because it decides to disable it's public network card. I have another server in in the same datacenter which never this problem never happens to and it's running Windows Server 2008 R2 as the host OS.

This is the error I keep getting:

Oct 9 16:33:57 zeus kernel: ------------[ cut here ]------------
Oct 9 16:33:57 zeus kernel: WARNING: at net/sched/sch_generic.c:267 dev_watchdog+0x26d/0x280() (Not tainted)
Oct 9 16:33:57 zeus kernel: Hardware name: X9SCL/X9SCM
Oct 9 16:33:57 zeus kernel: NETDEV WATCHDOG: eth1 (e1000e): transmit queue 0 timed out
Oct 9 16:33:57 zeus kernel: Modules linked in: vzethdev vznetdev simfs vzrst nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 vzcpt nfs lockd fscache nfs_acl auth_rpcgss sunrpc nf_conntrack vzdquota vzmon vzdev ip6t_REJECT ip6table_mangle ip6table_filter ip6_tables xt_length xt_hl xt_tcpmss xt_TCPMSS iptable_mangle iptable_filter xt_multiport xt_limit xt_dscp ipt_REJECT ip_tables vhost_net macvtap macvlan tun kvm_intel kvm vzevent ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr ipv6 iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi fuse tpm_tis snd_pcsp tpm snd_pcm snd_timer tpm_bios snd video i2c_i801 soundcore serio_raw i2c_core output snd_page_alloc shpchp ext3 jbd mbcache ahci e1000e [last unloaded: scsi_wait_scan]
Oct 9 16:33:57 zeus kernel: Pid: 29, comm: ksoftirqd/6 veid: 0 Not tainted 2.6.32-14-pve #1
Oct 9 16:33:57 zeus kernel: Call Trace:
Oct 9 16:33:57 zeus kernel: <IRQ> [<ffffffff8106c608>] ? warn_slowpath_common+0x88/0xc0
Oct 9 16:33:57 zeus kernel: [<ffffffff8106c6f6>] ? warn_slowpath_fmt+0x46/0x50
Oct 9 16:33:57 zeus kernel: [<ffffffff8147b57d>] ? dev_watchdog+0x26d/0x280
Oct 9 16:33:57 zeus kernel: [<ffffffff8107fbfc>] ? run_timer_softirq+0x1bc/0x380
Oct 9 16:33:57 zeus kernel: [<ffffffff8147b310>] ? dev_watchdog+0x0/0x280
Oct 9 16:33:57 zeus kernel: [<ffffffff81075363>] ? __do_softirq+0x103/0x260
Oct 9 16:33:57 zeus kernel: [<ffffffff8100c30c>] ? call_softirq+0x1c/0x30
Oct 9 16:33:57 zeus kernel: <EOI> [<ffffffff8100df35>] ? do_softirq+0x65/0xa0
Oct 9 16:33:57 zeus kernel: [<ffffffff81074ec0>] ? ksoftirqd+0x80/0x110
Oct 9 16:33:57 zeus kernel: [<ffffffff81074e40>] ? ksoftirqd+0x0/0x110
Oct 9 16:33:57 zeus kernel: [<ffffffff81095606>] ? kthread+0x96/0xa0
Oct 9 16:33:57 zeus kernel: [<ffffffff8100c20a>] ? child_rip+0xa/0x20
Oct 9 16:33:57 zeus kernel: [<ffffffff81095570>] ? kthread+0x0/0xa0
Oct 9 16:33:57 zeus kernel: [<ffffffff8100c200>] ? child_rip+0x0/0x20
Oct 9 16:33:57 zeus kernel: ---[ end trace c358b7f106434a55 ]---
Oct 9 16:33:57 zeus kernel: e1000e 0000:02:00.0: eth1: Reset adapter

I need this to be fixed as soon as possible and I would appreciate any and all help.

 

[tom]

use newer e1000e drivers, included already in pvetest

(see http://forum.proxmox.com/threads/11...her-packages-on-pvetest-(on-the-road-to-V2-2))

 

[Animoski]

My server (as well as VPSes on the server) does not use the E1000E NIC anywhere though, i'm not sure why they're related in this instance.

 

[tom]

from your log:

e1000e 0000:02:00.0: eth1: Reset adapter

 

[Animoski]

Thanks Tom,

I've updated to PVE Test, I will see if it continues to cause this issue.

Any idea what would cause this error?

"Network error: could could connect to server (mainserversIP):5900" after trying to "console" into a KVM VPS?