Date: Wed, 15 Dec 2004 07:37:51 -0800 From: sp0ng3b0b <sp0ng3b0b@sbcglobal.net> To: sp0ng3b0b <sp0ng3b0b@sbcglobal.net> Cc: questions@freebsd.org Subject: Re: FreeBSD 5.3: panic page fault with nge driver Message-ID: <41C05A4F.1080207@sbcglobal.net> In-Reply-To: <41C05789.3050300@sbcglobal.net> References: <41BF5A4C.2000203@sbcglobal.net> <41C05789.3050300@sbcglobal.net>
next in thread | previous in thread | raw e-mail | index | archive | help
I found some more info:
---snip---
To: <freebsd-hackers@freebsd.org>
Date: Tue, 16 Nov 2004 14:28:58 -0000
Subject: 5.3-RELEASE with nge crashes regularly under load
I tried enabling giant and setting the sysctl neither helped.
Unfortunately didn't have time to setup and get a dump of
the kernel panic as I had to get the machine up and running
ASAP. Replaced the netcard with an em and all was well.
Test to reproduce should be rsync on a dir with some large
files in.
Steve
---snip---
This could be the problem. Well, it was hard enough convincing mgmt to
let me use FreeBSD instead of RedHat. But I am not going to get budget
for new nics...especially when the worked before.
If ANYONE can help...I would appreciate it. Again, to restate, nge
driver is causing a "panic: page fault".
sp0ng3b0b wrote:
> I've been told that this might be fixed in CURRENT. That's nice, but I
> am trying to run a production quality server and CURRENT is not
> recommended for that purpose. Any recommendations?
>
> By the way, is Questions the wrong list for my problem?
>
> sp0ng3b0b wrote:
>
>> I am getting a page fault with a new install of FreeBSD 5.3 Release.
>> Something changed with the nge driver in FreeBSD 5.3. I did not have
>> this problem with 5.2.1. I just performed a clean install on a box
>> that had been running 5.2.1 RELEASE for a few months.
>>
>> The box is an Instrusion Detection System. I have fiber taps in place
>> on the network and use the rx ports of the nic to monitor the traffic.
>>
>> In 5.2.1, i would do this to bond ports and monitor them:
>>
>> # ifconfig nge0 media 1000baseSX mediaopt full-duplex promisc -arp up
>> # ifconfig nge1 media 1000baseSX mediaopt full-duplex promisc -arp up
>> # ngctl mkpeer fec dummy fec
>> # ngctl msg fec0: add_iface '"nge0"'
>> # ngctl msg fec0: add_iface '"nge1"'
>> # ngctl msg fec0: set_mode_inet
>> # ifconfig fec0 promisc -arp up
>>
>> However, when i set the media options for nge0, the machine panics and
>> dies. The panic message is hand typed below.
>>
>> FreeBSD 5.3 RELEASE.
>> Intel Pentium 4 3.06GHz CPU
>> 1GB RAM
>> 80GB Maxtor HD
>> (1x) Intel 82801DB (ICH4) Pro/100 Ethernet
>> (6x) NetGear GA621 fiber 1GB NIC
>>
>>
>> Fatal trap 12: page fault while in kernel mode
>> fault virtual address = 0x8
>> fault code = supervisor read, page not present
>> instruction pointer = 0x8:0xc055a002
>> stack pointer = 0x10:0xe32a2c9c
>> frame pointer = 0x10:0xe32a2cac
>> code segment = base 0x0, limit 0xfffff, type 0x1b
>> = DPL 0, pres 1, def 2 1, gran 1
>> processor eflags = interrupt enabled, resuep, IOPL = 0
>> current process = 27 (irq17: nge0)
>> trap number = 12
>> panic: page fault
>> _______________________________________________
>> freebsd-questions@freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
>> To unsubscribe, send any mail to
>> "freebsd-questions-unsubscribe@freebsd.org"
>>
> _______________________________________________
> freebsd-questions@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to
> "freebsd-questions-unsubscribe@freebsd.org"
>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41C05A4F.1080207>