Firebox Marvel ports locking up (CORE-E SERIES)

[power_matz]

I have a X750e with deactivated MSK intefaces.
I will give it a try to use them instead of the sk interface I currently use….

[corvey]

2.2.5 is out, maybe they fixed this bug.

[deanot]

I don't think 2.2.5 is a nano build is it?

[deanot]

No you're right, it is a nano build. I might try it out later, just to see if it is better or not.

[deanot]

2.2.5 Nano installed, I had to reset the ports about 18 times before I could get it on.  But it is on and running, and I am testing and abusing the ports to see if I can get it to lock up again.  At this time, I do believe it will do the same…..

[deanot]

That was short lived, locked up within minutes of posting this.  Oh well, time to ditch this brick and find something more reliable or go back to 2.1x

[corvey]

That sucks, thanks for trying 2.2.5 for us, deanot.  I will be retiring my Firebox on 2.1.5 and use it as a solid backup router.

I'm still waiting on my 1265Lv2 Xeon CPU to arrive from Fleabay.  I will be going ESXI this time around to install Pfsense on.  Hopefully it works.

[deanot]

I fired the box up again this morning, this time I uninstalled Snort, so far it has been up and running solid for the past few hours.  I am still expecting it to lock up on me, just have to wait for when it happens.

The only extension I am running right now, is LCDproc-Dev.  Nothing more, nothing less.

[deanot]

I did receive my HP 4 port intel nic card, I should be building up a new PFSense system sometime today, I am going to load a full install of the latest build on it and see what happens.  I am unsure if I want to restore my settings or start out completely fresh and hope for the best.

Being unsure of what is causing this, I feel the best approach would be to do everything from fresh.  For all I know, it could be my config that is doing it, it could be from a previous install doing it.  All I know, when I was on 2.1.x, it worked just fine.

Without much information from devs, I have no idea if drivers got updated/changed or what could be causing this.  I do know it is a known issue for some network cards, but nothing was mentioned about the Marvel nics, and this does seem a little more wide spread than just my system doing this as of right now.  It is apparent, this is not just firebox related anymore.

[deanot]

Just a quick update, 25 hours and still going.  Not one reboot in this time, not to say it is stable, but it could go at any moment.

I will continue to update…

Only thing I did was remove snort, I can't believe that would be the fix, we shall see..

[deanot]

3 days, 12 hours and still chugging along.

[rudelerius]

Maybe the hardware is just finally giving up the ghost?
https://forum.pfsense.org/index.php?topic=101920.0

[deanot]

@rudelerius:

Maybe the hardware is just finally giving up the ghost?
https://forum.pfsense.org/index.php?topic=101920.0

I don't believe this is the case, the logs show nothing like that and this has only started since going onto 2.2.3.  It is also across multiple systems, mine being the box listed below, the other on a PC system with an intel chipset card.  Mine is still up and running and appears to be stable, but saying that, it has gone as long as 15 days if I remember correctly.

Thanks for the link anyway.

[deanot]

It crapped itself this afternoon, twice in 5 minutes.  Kinda weird it does this, running great for 6 days 10 hours or so.

[deanot]

If anyone is following this thread, I would like to say I have scrapped the FireBox project, too much down time from the ports dropping out.  I have replaced it with another box with a lot better specs on the CPU and Memory…. Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz 4 CPUs: 1 package(s) x 4 core(s).

This thing is pretty sweet compared to the FireBox Celeron CPU.. lol

Take care all, see you in the next thread.

[chpalmer]

Im running a Watchguard x-550e box with donated 4 port switch from a 1250 at a site and have absolutely no issues what so ever with it.  (Now that Ive opened my mouth though…)

13 days uptime right now and only since the power outage that hit the area. Otherwise it had been up since 2.2.5 release day.  This box see's more than 40GB a day of transfer due to all the Netflix en such that goes on in the house there.

This is on the SK interfaces.

1.82Gb per hour average.

43.65Gb a day average.

23 active devices on the network right now.  :o    Big house.  Lots of people.

[BThunderW]

Can anyone confirm that this is an issue only with the SK interfaces or are both SK and MSK interfaces affected?

[dkrizic]

I had massive problems with my sk interfaces (I don't have the msk interfaces, because it is an x550e). A workaround is, that the interfaces are now running on 100baseTX and are stable now.

[chpalmer]

Im running a 550-e at gigabit with no issues.  Makes me wonder what the difference is between what Ive done and what I maybe haven't done.

Are you guys running the interface LED "fix" on your models?

[power_matz]

Yes, I have this fix installed. Do you think that this might cause the behaviour?