Date: Fri, 25 Apr 2014 15:55:29 +0100 From: "Steven Hartland" <killing@multiplay.co.uk> To: =?iso-8859-1?Q?Gerrit_K=FChn?= <gerrit.kuehn@aei.mpg.de> Cc: freebsd-net@freebsd.org Subject: Re: NFS over LAGG / lacp poor performance Message-ID: <006183CE786F4EAC9435B797E5F271CF@multiplay.co.uk> References: <535A1354.2040309@wp.pl><20140425113711.e7c7d1c2.gerrit.kuehn@aei.mpg.de><535A482E.1030106@wp.pl><20140425140123.a76c18f9.gerrit.kuehn@aei.mpg.de><535A5268.100@wp.pl><8247FE6336414E1F97ADA561D0680097@multiplay.co.uk><535A5DD9.9060206@wp.pl><AC1233D368D94519BB59B484D6F385C0@multiplay.co.uk><535A69CE.9010800@wp.pl><4E7A280CACC84910A3DE068085976E30@multiplay.co.uk><20140425161312.3ad95e3a.gerrit.kuehn@aei.mpg.de><1BD2C00D4F0D405C939C975246D41C60@multiplay.co.uk> <20140425163433.e35e5de3.gerrit.kuehn@aei.mpg.de>
index | next in thread | previous in thread | raw e-mail
----- Original Message -----
From: "Gerrit Kühn" <gerrit.kuehn@aei.mpg.de>
> On Fri, 25 Apr 2014 15:25:04 +0100 "Steven Hartland"
> <killing@multiplay.co.uk> wrote about Re: NFS over LAGG / lacp poor
> performance:
>
> SH> We saw the issue with 10-RELEASE last weekend on a machine with 6 x
> SH> igb's where by the system was burning CPU in the interrupt handlers,
> SH> setting num_queues to 2 fixed the issue we where seeing.
>
> Sounds like reducing the queues is the first thing to try then, thanks!
> How did you verify that the irqs were the culprit? Apart from the high irq
> rate in vmstat I do not see any other load values skyrocketing.
It was more CPU usage in top.
Run top then enable system processes (shift+S) and threads (shift+H) then
disable idle process display (z) if your seeing significant CPU usage to
the igb interrupt handlers try reducing the number.
Regards
Steve
home |
help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?006183CE786F4EAC9435B797E5F271CF>