Date: Fri, 9 Aug 2002 13:25:05 +0200 (CEST) From: Attila Nagy <bra@fsn.hu> To: Mike Barcroft <mike@FreeBSD.org> Cc: Yar Tikhiy <yar@FreeBSD.org>, <cvs-committers@FreeBSD.org>, <cvs-all@FreeBSD.org> Subject: Re: cvs commit: src/libexec/ftpd ftpd.8 ftpd.c Message-ID: <Pine.LNX.4.44.0208091313140.4115-100000@scribble.fsn.hu> In-Reply-To: <20020809013552.B94821@espresso.q9media.com> References: <200208081753.g78Hrq0X078614@freefall.freebsd.org> <20020809013552.B94821@espresso.q9media.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello, > Are these changes making it back to lukem to avoid lost functionality > when we make the switch to lukemftpd? I think before the switch could happen it is necessary to bring lukemftpd at least for the same level of performance as the current ftpd. Two examples: sendfile() support and resource usage. And I think they are really important... sendfile(): I did "performance measurements" with four machines. One NFS server (held the data), one FTP server (data over NFS) and two machines with some wget -O /dev/null ftp://server/bigfile running. With ftpd I got about 8 MB/s with 10-20% idle processor time, with lukemftpd I got about 4-5 MB/s with 0% idle... resource usage: without any hacks made, I could do about 750 concurrent ftp connection to ftpd (started from inetd) without any hicks and only about 200-270 to lukemftpd (I can't remember the exact number) with some weird kernel messages. So it seems that currently, lukemftpd needs some work to be able to do massive FTP services... Just my opinion. --------[ Free Software ISOs - ftp://ftp.fsn.hu/pub/CDROM-Images/ ]------- Attila Nagy e-mail: Attila.Nagy@fsn.hu Free Software Network (FSN.HU) phone @work: +361 210 1415 (194) cell.: +3630 306 6758 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.44.0208091313140.4115-100000>