Date: Sat, 25 Apr 2015 12:43:36 +0200 From: Baptiste Daroussin <bapt@FreeBSD.org> To: Dmitry Morozovsky <marck@rinet.ru> Cc: freebsd-stable@FreeBSD.org Subject: Re: protecting some processes from out-of-swap killer Message-ID: <20150425104336.GD13141@ivaldir.etoilebsd.net> In-Reply-To: <alpine.BSF.2.00.1504251316020.43520@woozle.rinet.ru> References: <alpine.BSF.2.00.1504251316020.43520@woozle.rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
--rqzD5py0kzyFAOWN Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Apr 25, 2015 at 01:31:14PM +0300, Dmitry Morozovsky wrote: > Hi there colleagues, >=20 > I have stable/10 on a rather big machine (2*8*2 e5 Xeon, 64G RAM, SAS+SSD= ZFS=20 > raid10+ZIL+L2ARC) acting as a PostgreSQL server.=20 >=20 > To use such a big resource pool that is mostly idle, I'd deployed poudrie= re=20 > there (using tmpfs) too. >=20 > Most times this combination works like a charm: LA could be 60+ and no vi= sual=20 > latency increase on SQL queries. >=20 > However, sometimes postgres processes got killed by 'out of swap space'. = =20 > I suppose the source of problem could be that VSZ size of postgres proces= ses=20 > (8-9 G) is bigger than swap congigured (4G). >=20 > Is there any way to prevent this, besides reallocating space for swap? >=20 > Quick googling does not help, at least I could not find answers relevant= =20 > enough. >=20 > Thanks! protect(1) ? Best regards, Bapt --rqzD5py0kzyFAOWN Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iEYEARECAAYFAlU7b9gACgkQ8kTtMUmk6EyPWgCdG6jHIEs8UyIPhQDJZ0AAMknP 3BQAnivyrsSl79NCoyA4gKNoCOj7NovQ =BDgz -----END PGP SIGNATURE----- --rqzD5py0kzyFAOWN--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150425104336.GD13141>