Date: Wed, 18 Jan 2012 20:25:25 +0100 From: peter h <peter@hk.ipsec.se> To: freebsd-stable@freebsd.org Subject: Re: about thumper aka sun fire x4500 Message-ID: <201201182025.26736.peter@hk.ipsec.se> In-Reply-To: <4F16FE42.3090300@egr.msu.edu> References: <201201171859.10812.peter@hk.ipsec.se> <20120117220912.GA32330@icarus.home.lan> <4F16FE42.3090300@egr.msu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday 18 January 2012 18.15, Adam McDougall wrote: > On 01/17/12 17:09, Jeremy Chadwick wrote: > > On Tue, Jan 17, 2012 at 06:59:08PM +0100, peter h wrote: > >> I have been beating on of these a few days, i have udes freebsd 9.0 an= d 8.2 > >> Both fails when i engage> 10 disks, the system craches and messages : > >> "Hyper transport sync flood" will get into the BIOS errorlog ( but not= hing will > >> come to syslog since reboot is immediate) > >> > >> Using a zfs radz of 25 disks and typing "zpool scrub" will bring the s= ystem down in seconds. > >> > >> Anyone using a x4500 that can comfirm that it works ? Or is this box b= roken ? > > >=20 > I've seen what is probably the same base issue but on multiple x4100m2=20 > systems running FreeBSD 7 or 8 a few years ago. For me the instant=20 > reboot and HT sync flood error happened when I fetched a ~200mb file via= =20 > HTTP using an onboard intel nic and wrote it out to a simple zfs mirror=20 > on 2 disks. I may have tried the nvidia ethernet ports as an=20 > alternative but that driver had its own issues at the time. This was=20 > never a problem with FFS instead of ZFS. I could repeat it fairly=20 > easily by running fetch in a loop (can't remember if writing the output=20 > to disk was necessary to trigger it). The workaround I found that=20 > worked for me was to buy a cheap intel PCIE nic and use that instead of=20 > the onboard ports. If a zpool scrub triggers it for you, I doubt my=20 > workaround will help but I wanted to relate my experience. The problem i had was most likley the disc-io itself. It was always there=20 whenever a larger number of discs was in motion.It was never there as=20 violent networking ( i even used myri2000 to increase traffic, never a prob= lem) A scrub on the 20-or-so zpool was all that was needed, andn when rebooting= =20 the scrub continued and whoops - a new reboot. Sometimes the bios reported not even 16G mem but 10.5 ( which also freebsd = noticed) Right now i am torturing the box with same load ( minus myri2000) and sunk-= os, i'll report if it does show simular problems. >=20 > > Given this above diagram, I'm sure you can figure out how "flooding" > > might occur. :-) I'm not sure what "sync flood" means (vs. I/O > > flooding). >=20 > As I understand it, a sync flood is a purposeful reaction to an error=20 > condition as somewhat of a last ditch effort to regain control over the=20 > system (which ends up rebooting). I'm pulling this out of my memory=20 > from a few years ago. > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" >=20 =2D-=20 Peter H=E5kanson =20 There's never money to do it right, but always money to do it again ... and again ... and again ... and again. ( Det =E4r billigare att g=F6ra r=E4tt. Det =E4r dyrt att laga fel.= )
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201201182025.26736.peter>