On Sat, 13 Oct 2007, joseph blase wrote: > On 10/13/07, Garrick Staples <garrick at usc.edu> wrote: > > On Sat, Oct 13, 2007 at 03:45:14AM +0800, joseph blase alleged: > > It seem that bind to something takes too much time. > > with trace=bind output, please shed light what's happening in this? > > [root at linux root]# strace -e trace=bind -T id userid > bind(5, {sa_family=AF_INET, sin_port=htons(682), sin_addr=inet_addr("0.0.0.0")}, 16) = 0 <0.000000> > bind(5, {sa_family=AF_INET, sin_port=htons(683), sin_addr=inet_addr("0.0.0.0")}, 16) = 0 <0.000000> -snip- > bind(5, {sa_family=AF_INET, sin_port=htons(688), sin_addr=inet_addr("0.0.0.0")}, 16) = 0 <0.000000> > bind(5, {sa_family=AF_INET, sin_port=htons(689), sin_addr=inet_addr("0.0.0.0")}, 16) = 0 <0.000000> > uid=36923(userid) gid=36923(u_036923) groups=36923(u_036923) Maybe this ? man ypbind -broken-server lets ypbind accept answers from servers running on an ille- gal port number. This should usually be avoided, but is required by some ypserv(8) versions. -- -- dag wieers, dag at centos.org, http://dag.wieers.com/ -- [Any errors in spelling, tact or fact are transmission errors]