On my main server, I just upgraded from 6.3.5 to 6.4.0 stable a few days ago.  The system seems to be functioning properly as far as I can tell, but, in examining the syslog, I found several call traces which I have not seen before.

 

Call traces appear to be related to macvlan and ip addressing; perhaps the cause is that I recently assigned an IP address to the UniFi docker.

 

Any idea as to the cause and how to address this?

 

Diagnostics attached.

medianas-diagnostics-20180121-1921.zip

Edited January 23, 20188 yr by Hoopster

Needed to reboot server to test some additions to go file unrelated to this issue.  There are no call traces on boot, but, the first one appeared about two hours later and a second (almost identical) one about 5-6 hours after the first. 

 

As mentioned previously, I have not yet observed any functional issues on the server, but, something could be happening that I have not yet observed.

 

Here's the latest call trace; same as before.

 

Jan 22 11:24:15 MediaNAS kernel: Call Trace:
Jan 22 11:24:15 MediaNAS kernel: <IRQ>
Jan 22 11:24:15 MediaNAS kernel: ipv4_confirm+0xac/0xb4 [nf_conntrack_ipv4]
Jan 22 11:24:15 MediaNAS kernel: nf_hook_slow+0x31/0x90
Jan 22 11:24:15 MediaNAS kernel: ip_local_deliver+0xab/0xd3
Jan 22 11:24:15 MediaNAS kernel: ? inet_del_offload+0x3e/0x3e
Jan 22 11:24:15 MediaNAS kernel: ip_rcv+0x30e/0x343
Jan 22 11:24:15 MediaNAS kernel: ? ip_local_deliver_finish+0x1b2/0x1b2
Jan 22 11:24:15 MediaNAS kernel: __netif_receive_skb_core+0x67b/0x6f4
Jan 22 11:24:15 MediaNAS kernel: process_backlog+0x8c/0x12d
Jan 22 11:24:15 MediaNAS kernel: net_rx_action+0xf6/0x24a
Jan 22 11:24:15 MediaNAS kernel: __do_softirq+0xc7/0x1bc
Jan 22 11:24:15 MediaNAS kernel: do_softirq_own_stack+0x2a/0x40
Jan 22 11:24:15 MediaNAS kernel: </IRQ>
Jan 22 11:24:15 MediaNAS kernel: do_softirq+0x46/0x52
Jan 22 11:24:15 MediaNAS kernel: netif_rx_ni+0x21/0x35
Jan 22 11:24:15 MediaNAS kernel: macvlan_broadcast+0x117/0x14f [macvlan]
Jan 22 11:24:15 MediaNAS kernel: macvlan_process_broadcast+0xe4/0x114 [macvlan]
Jan 22 11:24:15 MediaNAS kernel: process_one_work+0x146/0x239
Jan 22 11:24:15 MediaNAS kernel: ? rescuer_thread+0x258/0x258
Jan 22 11:24:15 MediaNAS kernel: worker_thread+0x1c3/0x292
Jan 22 11:24:15 MediaNAS kernel: kthread+0x10f/0x117
Jan 22 11:24:15 MediaNAS kernel: ? kthread_create_on_node+0x3a/0x3a
Jan 22 11:24:15 MediaNAS kernel: ? kthread_create_on_node+0x3a/0x3a
Jan 22 11:24:15 MediaNAS kernel: ret_from_fork+0x1f/0x30
Jan 22 11:24:15 MediaNAS kernel: Code: 48 c1 eb 20 89 1c 24 e8 24 f9 ff ff 8b 54 24 04 89 df 89 c6 41 89 c5 e8 a9 fa ff ff 84 c0 75 b9 49 8b 86 80 00 00 00 a8 08 74 02 <0f> ff 4c 89 f7 e8 03 ff ff ff 49 8b 86 80 00 00 00 0f ba e0 09 
Jan 22 11:24:15 MediaNAS kernel: ---[ end trace a3f8df7d7f81044d ]---

Edited January 23, 20188 yr by Hoopster

Continuing this conversation with myself in case it is helpful to others.

 

The IP address assignment to the UniFI docker does appear to be source of the call traces.  I removed the IP address assignment on the docker last night and let it go back to the same IP address as the unRAID server.  Since then (~12 hours) there have been no call traces generated in the syslog.  Previously, they would occur every 4-6 hours.  Of course, I will keep monitoring the situation to see if they return; however, indications so far are that this was the cause.