Date: Thu, 28 Jul 2016 14:09:34 +0300 From: Konstantin Belousov <kostikbel@gmail.com> To: Johan Hendriks <joh.hendriks@gmail.com> Cc: freebsd-stable@freebsd.org Subject: Re: swapoff: Cannot allocate memory FreeBSD 11 Beta2 Message-ID: <20160728110934.GD83214@kib.kiev.ua> In-Reply-To: <2601b5e7-8611-d4a3-9d24-4e727f60ad7a@gmail.com> References: <2601b5e7-8611-d4a3-9d24-4e727f60ad7a@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Jul 28, 2016 at 12:38:51PM +0200, Johan Hendriks wrote: > Hello all. > > Lately I noticed above (error) when I reboot machine. > > It looks like this. > > Stopping cron. > Waiting for PIDS: 1183 > swapoff: /dev/gpt/swap-9M286954: Cannot allocate memory > stopping zfsd. > Waiti...... > > My swap space concists of two GPT swap partitions on the zfs mirror OS > disks, both 4GB each. > The machine has 4 GB mem. > > This does not happen after normal use, but if i wait a day, do a > buildworld then the swapoff error shows. Swapoff ENOMEM means that there is too much anonymous memory allocated for it to be fit into RAM. E.g. if you have tmpfs or md(4) swap backed disks which consume too much space, you will get this error. Another possibility is some process which is still alive at the swapoff point, consuming enough anon memory.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20160728110934.GD83214>