Resetting network configuration is the only way

Apollo

Wizard
Joined
Jun 13, 2013
Messages
1,458
ya it came back up, and its snappy.
maybe i should make a task for it to do it every couple hours.
You shouldn't have to do that. I suspect the issue is with the network interface upon boot up.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
You shouldn't have to do that. I suspect the issue is with the network interface upon boot up.

Is there a config i can pull up and look at what it says?
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
is this any indication as to what is going on?
Code:
freenas.local kernel log messages:
> in6_purgeaddr: err=65, destination address delete failed
> em0: link state changed to DOWN
> em0: link state changed to UP

-- End of security output --

 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
I suspect the em(4) driver. Does lspci show you which Ethernet chipset is in use, or is that in dmesg somewhere? It’ll be Intel, of some description.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
I suspect the em(4) driver. Does lspci show you which Ethernet chipset is in use, or is that in dmesg somewhere? It’ll be Intel, of some description.

Code:
root@freenas:~ # lspci
00:00.0 Host bridge: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor Host Bridge/DRAM Registers (rev 07)
00:01.0 PCI bridge: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor PCIe Controller (x16) (rev 07)
00:01.0 PCI bridge: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor PCIe Controller (x8) (rev 07)
00:02.0 VGA compatible controller: Intel Corporation HD Graphics 530 (rev 06)
00:14.0 USB controller: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller (rev 31)
00:14.0 USB controller: Intel Corporation Sunrise Point-H Thermal subsystem (rev 31)
00:16.0 Communication controller: Intel Corporation Sunrise Point-H KT Redirection (rev 31)
00:16.0 Communication controller: Intel Corporation Sunrise Point-H CSME HECI #1 (rev 31)
00:17.0 SATA controller: Intel Corporation Sunrise Point-H SATA controller [AHCI mode] (rev 31)
00:1f.0 ISA bridge: Intel Corporation Sunrise Point-H LPC Controller (rev 31)
00:1f.0 ISA bridge: Intel Corporation Sunrise Point-H PMC (rev 31)
00:1f.0 ISA bridge: Intel Corporation Ethernet Connection (2) I219-LM (rev 31)
00:1f.0 ISA bridge: Intel Corporation Sunrise Point-H Northpeak (rev 31)
00:1f.0 ISA bridge: Intel Corporation Sunrise Point-H SMBus (rev 31)
02:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03)
root@freenas:~ #


If thats the case, would it work fine for 4+years then all of the sudden have problems?
also what is a em(4) driver

i called Att and had them send me a new gateway i need just need to connect everything.
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
If thats the case, would it work fine for 4+years then all of the sudden have problems?

If you change OS and thus driver base, sure. You said the issues started when you upgraded to 11.2 from 11.1, or did I misunderstand that?

You have an i219-LM, which I also see referenced as 82185. A quick man 4 em does not list it as officially supported by it, and it's not part of igb(4) either. The hardware recommendations guide says i219-LM is officially supported by FreeBSD, so I don't know what to think.
We had someone else this week who used em(4) with a Desktop Intel adapter, and had better luck with igb(4) once he moved to 11.3.

what is a em(4)

That's the FreeBSD driver handling your Ethernet controller chip. (4) references the man page section: devices. See https://www.kernel.org/doc/man-pages/ .

Could be your gateway as well, of course. Hopefully swapping that out resolves your woes.

If not, and you are willing to experiment, you could go for an Intel i210 off eBay (here's an example: https://www.ebay.com/itm/PCIe-x1-intel-I210-GbE-Network-card/233388854989), which gets you onto igb(4). That's the same controller used in a lot of SuperMicro boards. I think. A little hazy on whether there's a difference between i210 and i210-t1.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
You said the issues started when you upgraded to 11.2 from 11.1, or did I misunderstand that?
it started having problems when i went from 11.1U7 to 11.2U8

i cant remember when/why i bought it but i do have a intel Ethernet i210-t1 GbE NIC it actually has a hp part number on it as well.

would i put myself in a bad situation if i update to 11.3 and not promote my data sets just yet?

tyvm for all your help/
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
Fine to go to 11.3, just do not update feature flags. As long as you leave the pool the way it is, you can always go back. Once you upgrade feature flags, the pool becomes read-only in 11.2.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
Fine to go to 11.3, just do not update feature flags. As long as you leave the pool the way it is, you can always go back. Once you upgrade feature flags, the pool becomes read-only in 11.2.
my wife works from home so i have to wait an hour or so but im going to swap out the att gateway first.
Then ill update to 11.3 if it doesnt make things batter then ill pop in the i210.
i just finished testing my 6 wd easystores for a 6 by 8TB for a space upgrade.
 

Patrick M. Hausen

Hall of Famer
Joined
Nov 25, 2013
Messages
7,776
Did you try the obvious - check "disable hardware acceleration" for the interface in question? Just a quick shot from the hip, sorry if this is already covered, I did not reread the whole thread thoroughly.
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
i just finished testing my 6 wd easystores for a 6 by 8TB for a space upgrade.

Nice. I'm curious, what was in those? Are these still HGST He8 rebadged, or something else?
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
Nice. I'm curious, what was in those? Are these still HGST He8 rebadged, or something else?

Their HGST He8, but what is weird is the label on one of them.
Wd8.png
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
Did you try the obvious - check "disable hardware acceleration" for the interface in question? Just a quick shot from the hip, sorry if this is already covered, I did not reread the whole thread thoroughly.
Remind me where the check box is again.
thank you

i do have Enable autotune checked under system advanced
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
i do have Enable autotune checked under system advanced

Aha! He cried.

Turn that off, like, post haste. It does more harm than good. Check tunables and see whether it created any in there. I think one can just uncheck that and reboot, and, others may have more input.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
Aha! He cried.

Turn that off, like, post haste. It does more harm than good. Check tunables and see whether it created any in there. I think one can just uncheck that and reboot, and, others may have more input.
There was 20 or so things in the tuneables "set by autotune" let me reboot and see if that removes them.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
This is what i have in there after checking the box.



Code:
net.inet.tcp.sendspace
131072
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.arc_max
34650000000
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.l2arc_headroom
2
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.l2arc_noprefetch
0
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.l2arc_norw
0
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.l2arc_write_boost
40000000
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.l2arc_write_max
10000000
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.metaslab.lba_weighting_enabled
1
sysctl
Generated by autotune
yes
more_vert
vfs.zfs.zfetch.max_distance
33554432
sysctl
Generated by autotune
yes
more_vert
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
Code:
net.inet.tcp.sendspace


Not sure how that would interact with things. My recommendation is to remove all of those tunables, manually create a vfs.zfs.arc_max if you need additional room for jails and VMs, reboot, and see whether it's stable that way.

vfs.zfs.arc_max can be human-readable, so if for example you have 32GiB of RAM, and need 4GiB for the system and jails, you can set it to 28G and that will take effect on boot.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
Code:
net.inet.tcp.sendspace


Not sure how that would interact with things. My recommendation is to remove all of those tunables, manually create a vfs.zfs.arc_max if you need additional room for jails and VMs, reboot, and see whether it's stable that way.

vfs.zfs.arc_max can be human-readable, so if for example you have 32GiB of RAM, and need 4GiB for the system and jails, you can set it to 28G and that will take effect on boot.

so if it was created by autotune then i should remove it?
i dont remember if i have ever looked in tunables before.
 

Yorick

Wizard
Joined
Nov 4, 2018
Messages
1,912
so if it was created by autotune then i should remove it?

Yes. Disable autotune, and remove all the tunables it created.

This may not be where your issue lies, mind you. And, it's a way to simplify the setup and troubleshooting. You wouldn't be the first user to get in trouble with autotune.
 

tfran1990

Patron
Joined
Oct 18, 2017
Messages
294
After replacing router and removing autotune tunables, i think that solved my problem. once i tried to upgrade to 11.3 i started having other issues. on reboot the middlewared would try to load but take so long the process would stop and i would be at a command line.
 
Top