PfSense 2.3 LAN interface stops routing traffic - stops working after 2 or 3 day

j.koopmann

Sure..


#!/usr/local/bin/perl

use Net::Ping;
use Sys::Syslog;

$server_to_ping="192.168.1.1";
$server_to_ping2="192.168.1.2";

sub check_ping_server
{
$host_alive=1;
$ping=Net::Ping->new('icmp');
if( $ping->ping($_[0]) ) { $host_alive=1;}
 else  {$host_alive=0;}
return $host_alive;
}

if(!check_ping_server($server_to_ping) && !check_ping_server($server_to_ping2))
    {
    system("ifconfig igb2 down");
    system("sleep 2");
    system("ifconfig igb2 up");
    system("sleep 5");
    openlog("checkigb2", "ndelay", LOG_USER);
    syslog('notice', 'IP check failed, igb2 restarted');
    closelog();
    }
    else
    {
      openlog("checkigb2", "ndelay", LOG_USER);
      syslog('notice', 'IP check ok');
      closelog();
    }

exit;

Not the best piece of coding but it seems to do the trick.

Regards,
JP

OLBaID

Hi been tracking the bug (thanks CMB!):

https://redmine.pfsense.org/issues/6296

A comment on the bug tracker had this link:

https://forum.pfsense.org/index.php?topic=107471.msg602590#msg602590

This user disabled the following in System/ Advanced / Networking:

I had all enabled:

Disable hardware checksum offload
Disable hardware TCP segmentation offload
Disable hardware large receive offload

Can anyone else here confirm if this helps if they are experiencing the issue (I am working to get this hardware back online soon to test myself)

Not trying to muddy any waters, but hoping for a fast resolution.

Thanks

erdmensch

I had these options disabled even before the upgrade to 2.3.
Only disabling the CPU cores solves the issue for me.

jeffvfren

I'm having the same issue. Disable CPU cores, still under monitoring.
Hopefully the fix release asap.

Update:
It does not work for me, the issue just happen again.

h311m4n

Add us to the list as well.

We have a virtual pfsense cluster on our DRC site with and ipsec tunnel to our prod site. Since the update, we've had constant crashes from the master on the DRC site.

It has been stable for the past 2-3 days now, but with a VCENTER session open, I see a constant CPU usage warning on the master on the other side. We have done the CPU workaround too but as soon as we launch our VEEAM replications, we can be pretty much sure that the ipsec tunnel will fall. Def looks like all the UDP traffic is basically doing a DOS…

cmb

We're still working on tracking this down. Have it narrowed down to something in our IPsec changes (which are just back-ports from FreeBSD -CURRENT), but that's still 80 change sets potentially related.

h311m4n

@cmb:

We're still working on tracking this down. Have it narrowed down to something in our IPsec changes (which are just back-ports from FreeBSD -CURRENT), but that's still 80 change sets potentially related.

Good to know you're working on it! :)

marcvb

We also are having this problem with 2.3 , it happens each week.
We use multiple ipsec connections.
The pfsense is inside a vmware.
2Gb ram
4 cores

I think the state table size is also bigger: 82% (165500/201000)
When i want to open the state table it also crashes Allowed memory size of 268435456 bytes exhausted.
I will put more ram in the system when i have a downtime window.

–--
Yes i got my down window upgraded the ram to 4Gb, can now view the states without any problem.

j.koopmann

Happened again today twice but with different details.

LAN went down. For whatever reason my script doing ifconfig down and up did not help. I tried manually with no luck. Retried and introduced longer sleep between down and up and then it worked. This was the first crash in several days.
appr. 60 minutes later LAN was down again. However this time I did not even manage to get any result on the serial console. I had to power down/up. No core dump I was able to find however after login I saw a crash and uploaded it. not sure if it is related. I would say it happened around the time of (1).


Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address	= 0x0
fault code		= supervisor read data, page not present
instruction pointer	= 0x20:0xffffffff80d22566
stack pointer	        = 0x28:0xfffffe001a38c590
frame pointer	        = 0x28:0xfffffe001a38c770
code segment		= base 0x0, limit 0xfffff, type 0x1b
			= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags	= interrupt enabled, resume, IOPL = 0
current process		= 12 (irq260: igb2:que 0)
version.txt06000025412713111367  7616 ustarrootwheelFreeBSD 10.3-RELEASE #31 01118b4(RELENG_2_3): Thu Apr 28 03:57:55 CDT 2016
    root@ce23-amd64-builder:/builder/pfsense/tmp/obj/builder/pfsense/tmp/FreeBSD-src/sys/pfSense

Regards,
JP

jswope

I have 5 locations and we use the Dell R210 servers all sites except 2 have this issue. Only difference between them are Internet Service Providers.

Site 1 R210 Charter Fiber no issues

Site 2 AT&T LTE charter fiber bo issues

Site 3 Charter Fiber issues every 2-3 days watchdog error no lan routing

Site 4 charter coax internet issues every 3-days

marcvb

A other firewall of us got stuck just now, when i looked at the console it was showing a different wan adres.
It seems to have been reverted to a private, will take a screenshot next time.

cmb

We seem to have a fix for this. It's not yet merged into the source tree, but will be soon.

@jswope:

I have 5 locations and we use the Dell R210 servers all sites except 2 have this issue. Only difference between them are Internet Service Providers.

It's not universal or even close to it or we wouldn't have released with the issue. Not ISP-specific. Just happens the traffic profile of some systems will encounter it. Certain UDP streams is what triggers it.

afreaken

@cmb:

We seem to have a fix for this. It's not yet merged into the source tree, but will be soon.

@jswope:

I have 5 locations and we use the Dell R210 servers all sites except 2 have this issue. Only difference between them are Internet Service Providers.

It's not universal or even close to it or we wouldn't have released with the issue. Not ISP-specific. Just happens the traffic profile of some systems will encounter it. Certain UDP streams is what triggers it.

Do you have an ETA?

cmb

Sometime this week.

eeit

Hello,

any news on this ? Is there a special setting or a patch to fix the issue available ?

Thx.

cmb

Reducing the system to one core is the only workaround. The fix was merged into devel branch today and hopefully into RELENG_2_3 tomorrow, in which case it'll hit snapshots tomorrow.

jonnyaalgaard.org

Then I have upgraded to the new snapshot that came this morning (2.3.1-DEVELOPMENT (amd64) - built on Thu May 12 00:18:20 CDT 2016 - FreeBSD 10.3-RELEASE-p2), and will check it out.
I have also removed the temp workaround, from /boot/loader.conf.local
The temp workaround for me were to add these lines into loader.conf.local
hint.lapic.1.disabled=1
hint.lapic.2.disabled=1
hint.lapic.3.disabled=1

When I entered these lines pfsense were stabile for me, but now I have enabled all cpu cores again, and hopefully the new snapshot will work just fine.
Before the workaround I did have 10-12 “crashes” a day.
I will post back if the new snapshot works Ok or not.

UPDATE! Before i managed to post this my pfsense crashed again with the new snapshot. I will try to disable all but 1 cpu core again, still using todays snapshot.

cmb

There is no fix yet, as I noted. Watch the bug ticket for updates.

cmb

The fix for this was merged this morning and is in the most recent available snapshot, Thu May 12 14:01:47 CDT and newer.

My test setups are around 4.5 hours run time in a scenario that never lasted more than 3-4 hours without the fix. Another user who had a circumstance that was much faster to replicate than anything I could duplicate in a lab (a matter of a handful of minutes, rather than hours) has also confirmed it's no longer happening.

Needs more runtime and more feedback from others, but initial results are good.

Those of you impacted, please upgrade to latest 2.3.1 (instructions here), remove the disabling cores workaround if you did that, and let us know how it goes.

byusinger84

@cmb:

The fix for this was merged this morning and is in the most recent available snapshot, Thu May 12 14:01:47 CDT and newer.

My test setups are around 4.5 hours run time in a scenario that never lasted more than 3-4 hours without the fix. Another user who had a circumstance that was much faster to replicate than anything I could duplicate in a lab (a matter of a handful of minutes, rather than hours) has also confirmed it's no longer happening.

Needs more runtime and more feedback from others, but initial results are good.

Those of you impacted, please upgrade to latest 2.3.1 (instructions here), remove the disabling cores workaround if you did that, and let us know how it goes.

OP here. Testing this for you now. It usually crashed on me within 6-12 hours but up to 24 hours was the latest. I should know no later than this weekend if it is working or not. I'll update you when I know more. Thank you!