Date: Tue, 29 Jun 2021 20:22:31 +0000 From: Rick Macklem <rmacklem@uoguelph.ca> To: Peter Eriksson <pen@lysator.liu.se> Cc: freebsd-net <freebsd-net@freebsd.org> Subject: Re: RFC: NFS trunking (multiple TCP connections for a mount Message-ID: <YQXPR0101MB09686A55B3373E452E09B8B7DD029@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM> In-Reply-To: <362300CE-30DA-4552-A3E4-0F3DFE385B2A@lysator.liu.se> References: <YQXPR0101MB0968DC173855A82AAF45F08FDD039@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM>,<362300CE-30DA-4552-A3E4-0F3DFE385B2A@lysator.liu.se>
next in thread | previous in thread | raw e-mail | index | archive | help
Thanks everyone, for your comments. I now see (never used it nor looked at it before) that lagg/lacp pins traffic for a TCP connection to an interface (not infallibly for Peter). I also now learned a little bit w.r.t. multiple queues for fast NICs. Sounds like the "nconnect" option is worth implementing. Thanks for your input, rick ________________________________________ From: Peter Eriksson <pen@lysator.liu.se> Sent: Tuesday, June 29, 2021 5:11 AM To: Rick Macklem Cc: freebsd-net Subject: Re: RFC: NFS trunking (multiple TCP connections for a mount CAUTION: This email originated from outside of the University of Guelph. Do= not click links or open attachments unless you recognize the sender and kn= ow the content is safe. If in doubt, forward suspicious emails to IThelp@uo= guelph.ca > I don't understand how multiple TCP connections to the same > server IP address will distribute the load across multiple network > interfaces? > I thought that lagg would have handled this? A lagg typically keeps all data in a TCP stream on a specific lagg member (= depending on how the lagg is set up, unless you select the =93roundrobin=94= option in freebsd - don=92t do that unless you like out-of-order packets= =85) Network equipment with laggs typically hash the IP streams over the lagg me= mbers based on MAC addresses (source&target), IP addresses (source&target) = and port numbers. (We have been diagnosing a fun problem locally where we see packet losses/p= erformance drops over our internal backbone network for certain combination= s of odd/even IP addresses/port numbers when things pass certain SPB =93rou= ters=94 (which typically hash the streams over many =93channels=94 between = routers)=85 Fun fun. :-) I think the multiple NFS TCP streams could make for some nice performance i= mprovements in certain cases. And it would be a more generalisation of havi= ng multiple streams between two hosts - one-or-many over IPv4 and one-or-ma= ny over IPv6 at the same time. Windows SMB has a similar feature. Just avoid the Linux NFS mounting deadlock issue with =93down=94 servers pl= ease :-) - Peter
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?YQXPR0101MB09686A55B3373E452E09B8B7DD029>