On 10/23/05, Dimitar Vasilev <[EMAIL PROTECTED]> wrote: > > I don't like the style of djb, sorry. > > I know, but his tools work and are great for such things. > There have been occasions on which the bgp "blinks"/ntpd decides to go play > and after a while the box goes disfunct and I have to start/stop manually. >
I guess that's why we let cisco handle bgp (and ntp, by the way). > > The box has 1 ide and 4 scsi disks. But I had to > > remove scsi cables and can't find a replacement > > at the moment, so it's on ide for now. > > > > Squid is also on the server. So I guess, when > > scsi disks are back online, I'll dedicate 2 to > > squid, and 2 in a vinum mirroring array with > > softupdates disabled for more important > > things, like cvs repos, distfiles, etc. > > > > Try setting up a md through which to buffer the read/write operations for > cvsupd > For CVS there is a hint in the hubs article > /dev/da0s1b /anoncvstmp mfs > rw,-s=786432,-b=4096,-f=512,-i=560,-c=3,-m=0,nosuid,nodev 0 0 > > /etc/inetd.conf > cvspserver stream tcp nowait root /usr/bin/cvs cvs -f -l -R -T > /anoncvstmp --allow-root=/home/ncvs pserver > > I suppose if you play a bit, you can come with a similar solution for CVSUP. > > Btw, have you extended an existing stripe with growfs? > > I have to extend my stripe on which ftp+rsync repo lies with 0,5TB and > would like to preserve data if possible. > All the best > -- > Димитър Василев > Dimitar Vassilev > > GnuPG key ID: 0x4B8DB525 > Keyserver: pgp.mit.edu > Key fingerprint: D88A 3B92 DED5 917E 341E D62F 8C51 5FC4 4B8D B525 > It doesn't really seem that disk performance is a bottle-neck. CPU is saturated much more quickly. It's a dual P3 box, 5-6 cvsup sessions with compression turned on is as much as it is gonna take without major slowdowns. I guess there's no way to make a new IDE disk lag behind over a 100Mbit link. But I'm thinking about gigabit - and cvs instead of cvsup. That's where your recommendations will come into play. No, I haven't been growing any stripes in my garden lately :-) Thanks, Andrew P.
_______________________________________________ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
