Date: Thu, 05 Mar 2015 15:53:19 -0700 From: Ian Lepore <ian@freebsd.org> To: Brett Wynkoop <freebsd-arm@wynn.com> Cc: freebsd-arm@freebsd.org Subject: Re: crash on writing usbstick Message-ID: <1425595999.28798.11.camel@freebsd.org> In-Reply-To: <20150305153421.2e2bca98@ivory.wynn.com> References: <20150301041855.5352663e@ivory.wynn.com> <20150301144653.63b38cdf@ivory.wynn.com> <20150301184456.7b5e6487@ivory.wynn.com> <1DC8221F-64EA-418C-8CE5-5FFA4F3DBC64@bsdimp.com> <20150301203244.55578413@ivory.wynn.com> <F79FBD40-6002-4C38-A191-A74A88DA929B@bsdimp.com> <20150305064318.2f35f2c0@ivory.wynn.com> <1425567301.3471.6.camel@freebsd.org> <20150305153421.2e2bca98@ivory.wynn.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 2015-03-05 at 15:34 -0500, Brett Wynkoop wrote: > On Thu, 05 Mar 2015 07:55:01 -0700 > Ian Lepore <ian@freebsd.org> wrote: > > > I spent some time yesterday trying to reproduce this, and couldn't. > > The only usb thumb drive I have handy right now is 4gb, which isn't > > quite big enough to hold the entire ports tree, but using your tar|tar > > incantation it will run to the point where the destination filesystem > > is full without any errors; I did that twice. > > > > You didn't say what you were using as a source for the copy (sdcard, > > nfs, etc). I was using an nfs mount as source, which on BBB means > > that usb is involved as both the source and destination. > > > > Hmm, I just realized the ports tree I'm copying includes .svn, no > > wonder it's so big. I'll bet if I exclude that from the copy it'll > > run to completion. > > Greeting- > > Would you like access to my BB via ssh and also access to it's console > via ssh? > > I have had the same results with several usb sticks, but all of them > have been at least 16GB. I am also swapping to the current USB stick > on a swap partation, but I have had the same results on this stick and > others without swapping to it. > > /usr/ports is on the root fs on the sd card. > I got my ports tree onto sdcard then copied it to the usb stick using the tar|tar command, still can't reproduce the problem at all. Nobody else has reported any beaglebone IO-related panics or errors for months either. I wonder what's different about your setup? (Nice side effect... doing the nfs->usb copy created hundreds of spurious interrupt messages, I've never been able to reproduce them reliably before. I think I've got them fixed now; tests still running.) -- Ian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1425595999.28798.11.camel>