Network dropping and random crashing.

[Vlash]

Recently my server has been loosing connection. I have noticed it when visiting the tower/main, or any other page for that matter. I will often get an unable to connect page when navigating around the gui. To make matters worse, the server will be down when I get home. I mirrored the logs and have the last log when the system powered down. I am not good at understanding the logs so maybe someone can point me to the issues. I am seeing lots of eth0: Link Down but just cannot figure out what is causing it. Looks like it is happening on disk spindown, but like I said I am horrible at these logs. Really the only thing I have tried thus far is changing out the controller card, and changing the ethernet cable. Besides that, I really don't have a clue. Thanks in advance.

 

May 23 17:27:07 Tower kernel: alx 0000:01:00.0 eth0: Link Down
May 23 17:27:07 Tower kernel: bond0: link status definitely down for interface eth0, disabling it
May 23 17:27:07 Tower kernel: device eth0 left promiscuous mode
May 23 17:27:07 Tower kernel: bond0: now running without any active interface!
May 23 17:27:07 Tower kernel: br0: port 1(bond0) entered disabled state
May 23 17:27:09 Tower ntpd[1581]: Deleting interface #5 br0, 192.168.1.102#123, interface stats: received=86, sent=190, dropped=0, active_time=52874 secs
May 23 17:27:09 Tower ntpd[1581]: 45.33.103.94 local addr 192.168.1.102 -> <null>
May 23 17:27:09 Tower ntpd[1581]: Deleting interface #6 br0, fe80::4cf2:daff:fe1b:a409%10#123, interface stats: received=0, sent=0, dropped=0, active_time=52874 secs
May 23 17:27:13 Tower kernel: alx 0000:01:00.0 eth0: NIC Up: 1 Gbps Full
May 23 17:27:13 Tower kernel: bond0: link status definitely up for interface eth0, 1000 Mbps full duplex
May 23 17:27:13 Tower kernel: bond0: making interface eth0 the new active one
May 23 17:27:13 Tower kernel: device eth0 entered promiscuous mode
May 23 17:27:13 Tower kernel: bond0: first active interface up!
May 23 17:27:13 Tower kernel: br0: port 1(bond0) entered blocking state
May 23 17:27:13 Tower kernel: br0: port 1(bond0) entered forwarding state
May 23 17:27:14 Tower ntpd[1581]: Listen normally on 7 br0 192.168.1.102:123
May 23 17:27:14 Tower ntpd[1581]: Listen normally on 8 br0 [fe80::4cf2:daff:fe1b:a409%10]:123
May 23 17:27:14 Tower ntpd[1581]: new interface(s) found: waking up resolver

 

syslog

[Vlash]

**Bump**

Anyone have any suggestions?

[Squid]

Ethernet:  Have you checked / replaced the cable or switched ports on the switch yet?

 

Crashing: Post your diagnostics

 

Other:  Quick glance it also appears that you've got a loose / poor connection to one or more drives from the SAS controller

[Vlash]

I replaced the cable yes.

I have reseated all my cables to my drives. Actually unplugged everything and plugged them back in even though the sata cables have the locking clips to make sure. Plus the power cables as well. Same on the controller, unplugged cables and plugged back in.

 

What is it that I need to do to grab a diagnostics?

 

Thanks

 

Edit:

Forgot to say switched ports on router too. Even switched from my router to my hub with same results.

Edited May 30, 2020 by Vlash
Forgot information

[Vlash]

Ok I figured out how to grab a diagnostic report. Thought that download was just to grab the system log, but wiki said otherwise. Anyway here it is

tower-diagnostics-20191009-0616.zip

[Vlash]

Guess this one has everyone stumped lol.

[tripleflip]

I'm having the same issue, have you tried booting into SAFE MODE NO PLUGINS. I've noticed that when booted into there it is stable. So does that mean its the plugins? Isn't it odd a lot of us are having the same issue. 

[tripleflip]

On 5/30/2020 at 4:59 AM, Squid said:

Ethernet:  Have you checked / replaced the cable or switched ports on the switch yet?

 

Crashing: Post your diagnostics

 

Other:  Quick glance it also appears that you've got a loose / poor connection to one or more drives from the SAS controller

Even if the network cable was disconnected that wouldn't cause a system reboot would it? 

[Vlash]

I didn't try booting into safe mode, so I can't say much about your plugins and if it's causing your issue.

 

I don't think the network cable being disconnected would play any role in the system rebooting, but I have not tried that either.

 

I do think I found my issue. I'm testing this weekend to see, but so far the results are good. I think that one of the cables on my multi-lane forward cable is bad. With @Squid pointing out the poor/loose cable connection, I kept thinking about it because I did unplug everything and plug it in again to make sure everything was connecting properly. So I started with disk 6 since it was the one showing in the logs as failing disk and replaced that cable with one not in use. Two of the four sata cables were in use, so I had two I could test with.

 

Once I swapped out the cable and booted back up, I have not had any issues, even network issues. Lesson learned on this one, and it's crazy that the data cable caused my network to drop like it was. So @tripleflip you might try changing cables out to see if it helps.