Date: Sat, 14 Mar 2015 17:39:41 +0200 From: Konstantin Belousov <kostikbel@gmail.com> To: Michael Fuckner <michael@fuckner.net> Cc: Steven Hartland <killing@multiplay.co.uk>, Ryan Stone <rysto32@gmail.com>, "freebsd-hackers@freebsd.org" <freebsd-hackers@freebsd.org> Subject: Re: Server with 3TB Crashing at boot Message-ID: <20150314153941.GL2379@kib.kiev.ua> In-Reply-To: <5504530F.5040508@fuckner.net> References: <CAFMmRNzmr0Z1vXYK5y2QLk3eyUqXyK8TELgh4KHZ%2BDQVDXWZAg@mail.gmail.com> <275339388.395931.1426279324879.JavaMail.open-xchange@ptangptang.store> <CAFMmRNya8ZV=D3z0dE1%2BDsprAVBvcHZFGN1g1Ezd8k84nE3kVA@mail.gmail.com> <5503DC66.40409@fuckner.net> <55041EF5.9080200@multiplay.co.uk> <5504362C.9000904@fuckner.net> <20150314134733.GI2379@kib.kiev.ua> <550448B3.7050100@fuckner.net> <20150314145607.GJ2379@kib.kiev.ua> <5504530F.5040508@fuckner.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 14, 2015 at 04:26:07PM +0100, Michael Fuckner wrote: > On 3/14/2015 3:56 PM, Konstantin Belousov wrote: > > On Sat, Mar 14, 2015 at 03:41:55PM +0100, Michael Fuckner wrote: > >>>> > >>>> In the Processor Configuration tab there also was > >>>> extended apic support (default is disabled) > >>> On 10.1 this option should not result in any behaviour change. > >>> > >>>> > >>>> Should I give this option a try? > >>> > >>> Up to you. I am somewhat curious whether it boots with DMAR enabled and > >>> what happens if it does not. > >>> ____ > >> > >> looks good, doesn't it? > >> > >> http://dedi3.fuckner.net/~molli123/temp/dmar.log > > No, it does not. At least I suspect that your ix* interfaces are not > > functional. The issue is with DMAR code, not with the ixgbe driver, and > > it was probably fixed in stable/10. > > why do you think it is not working? > > root@s4l:~ # ping dedi3.fuckner.net > PING dedi3.fuckner.net (81.169.184.133): 56 data bytes > 64 bytes from 81.169.184.133: icmp_seq=0 ttl=56 time=15.522 ms > 64 bytes from 81.169.184.133: icmp_seq=1 ttl=56 time=11.266 ms Ok. I looked at the lines DMAR1: <unknown dev>:pci3:0:0 fault acc 0 adt 0x0 reason 0x5 addr 0 Due to the nature of the bug, the pci rid above could be unrelated. Anyway, good that ix* work, but it is not clear which device transfer faulted. > > > > [root@mf5 ~]# grep ix0 dmar.log > ix0: Link is Down > ix0: link state changed to DOWN > ix0: <Intel(R) PRO/10GbE PCI-Express Network Driver, Version - 2.5.15> > port 0x4020-0x403f mem 0xaaa00000-0xaabfffff,0xaac04000-0xaac07fff irq > 32 at device 0.0 on pci1 > ix0: attempting to allocate 9 MSI-X vectors (64 supported) > ix0: using IRQs 272-280 for MSI-X > ix0: Using MSIX interrupts with 9 vectors > ix0: dmar3 pci0:1:0:0 domain 2 mgaw 48 agaw 48 re-mapped > ix0: bpf attached > ix0: Ethernet address: c4:54:44:92:73:d2 > ix0: PCI Express Bus: Speed 5.0GT/s Width x8 > ix0: Link is up 1 Gbps Full Duplex > ix0: ix0: link state changed to UP > ix0: link state changed to DOWN > Starting Network: lo0 ix0 ix1 ix2 ix3. > ix0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500 > nd6 options=29<PERFORMNUD,IFix0: LiDISABLED,AUTO_LInk is up 1 > Gbps Full Duplex > mediaix0: link state changed : Ethernet autosto UP > DHCPREQUEST on ix0 to 255.255.255.255 port 67 > > > I'll checkout STABLE to check now... Thank you.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150314153941.GL2379>