Date: Mon, 22 Aug 2016 16:30:17 -0700 From: Kevin Oberman <rkoberman@gmail.com> To: Jan Mikkelsen <janm@transactionware.com> Cc: "freebsd-emulation@freebsd.org" <emulation@freebsd.org> Subject: Re: Odd behavior with virtualbox-ose-4.3.38 Message-ID: <CAN6yY1s=EPL-nYxDrgvgxG4xQnhVaOiODwHLy3MXX-dqpddouA@mail.gmail.com> In-Reply-To: <59E9F6CE-0F2B-40E1-B78D-AF9DFBCED280@transactionware.com> References: <CAN6yY1vn_o%2B=bX7MRNwnurp_G%2BDRhD3sfSS63-1hx4JbYsVaSw@mail.gmail.com> <CAN6yY1sG92DUR0-7Y_wE-r94%2B22JZszgcsOy=XrY=J77njUC_Q@mail.gmail.com> <2418C811-3C6E-4837-B6A5-51DB4C425EB4@transactionware.com> <CAN6yY1v%2BfxZKYexP5m=xUtqaoPjm1zM_4LHXtY5YD_2qAecOoQ@mail.gmail.com> <59E9F6CE-0F2B-40E1-B78D-AF9DFBCED280@transactionware.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Aug 22, 2016 at 12:54 PM, Jan Mikkelsen <janm@transactionware.com> wrote: > Hi, > > On 23 Aug 2016, at 02:25, Kevin Oberman <rkoberman@gmail.com> wrote: > > On Mon, Aug 22, 2016 at 1:50 AM, Jan Mikkelsen <janm@transactionware.com> > wrote: > >> Do you have aio loaded or compiled into your kernel? If so, does it keep >> happening without aio? >> [ =E2=80=A6 ] >> > Thanks for the suggestion, but, no. This is on my trusty Lenovo T520 > laptop and no aio to be found. My configuration is just: > include GENERIC > ident GENERIC_BSD4 > > nooptions SCHED_ULE # ULE scheduler > options SCHED_4BSD # 4BSD scheduler > > I still can't be sure if the problem was triggered by the move to VB 5 or > t FreeBSD-11 (or a combination of both). > > > Yes, the problem could be in many places. However, if you=E2=80=99ve gone= to > FreeBSD 11, you have more aio than you think. From UPDATING: > > 20160301: > The AIO subsystem is now a standard part of the kernel. The > VFS_AIO kernel option and aio.ko kernel module have been removed. > Due to stability concerns, asynchronous I/O requests are only > permitted on sockets and raw disks by default. To enable > asynchronous I/O requests on all file types, set the > vfs.aio.enable_unsafe sysctl to a non-zero value. > > Are you running the VM against a raw disk device? You can check the > VBox.log output =E2=80=94 If you are not using aio, you should see an ent= ry like > this: > > AIO: Async I/O manager not supported (rc=3DVERR_NOT_SUPPORTED). Falling b= ack > to simple manager > > If you are using AIO, you will should see entries like: > > AIOMgr: Default manager type is "Async" > AIOMgr: Default file backend is =E2=80=9CNonBuffered" > > > AIO issues were not a big issue for a long time but some recent updates t= o > 4.3.something (or other environmental issues, possibly load) led me to tu= rn > aio off on a bunch of 10.x machines running a significant number of VMs > (~25). I=E2=80=99m have been curious about whether the aio change would c= ause > problems with an upgrade to 11. > > Regards, > > Jan. > > BINGO! Thank you so much, Jan! This does explain a great deal of what I was seeing and why. Now the question... how can I disable it? I'm guessing that "kern.features.aio=3D0" in /boot/loader.conf might do the trick, but the ma= n page does not mention this at all. Also, I guess it is time to open a bug report on this. FWIW, the problem does seem to be linked to disk activity. I had the system locked up and had to kill the process. When I rebooted, the system told me that a disk check was needed, but I had it lock up several time (3, I think) before it made it through the check without locking up again. Once that was complete, the system started and ran normally again. This also now fits rather well with an aio issue. Not too sure how to go about collecting more information for the PR,but I'll open it with what I have. Thanks again, Jan! -- Kevin Oberman, Retired Network Engineer
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAN6yY1s=EPL-nYxDrgvgxG4xQnhVaOiODwHLy3MXX-dqpddouA>