Greetings!
[EMAIL PROTECTED] (Mark Carmichael) writes:
> A single datapoint for "head" hecklers:
>
> [554] mxc@suse61: time ypcat hosts | wc -l
> 18259
>
> real 0m11.435s
> user 0m0.340s
> sys 0m0.050s
> [555] mxc@suse61: time ypcat hosts | head -1 | wc -l
> 1
>
> real 0m0.176s
> user 0m0.050s
> sys 0m0.010s
> [556] mxc@suse61: ypwhich
> yphost
> [557] mxc@suse61: ping -c 4 yphost
> PING yphost (153.32.2.254): 56 data bytes
> 64 bytes from 153.32.2.254: icmp_seq=0 ttl=254 time=1.332 ms
> 64 bytes from 153.32.2.254: icmp_seq=1 ttl=254 time=1.473 ms
> 64 bytes from 153.32.2.254: icmp_seq=2 ttl=254 time=1.487 ms
> 64 bytes from 153.32.2.254: icmp_seq=3 ttl=254 time=2.837 ms
> --- yphost ping statistics ---
> 4 packets transmitted, 4 packets received, 0% packet loss
> round-trip min/avg/max = 1.332/1.782/2.837 ms
>
> Seems worthwhile give the "wc -l" approach; I'd probably try
>
Thanks for this!
> if ypwhich -m $map 2>&1 >/dev/null; then
>
> myself, but I don't know enough to be confident that there aren't lots of
> empty maps being served in the world. I also don't know whether the exit
> status of 'ypwhich' can be trusted across likely implementations/versions.
I rather like this too, but I thought it best to stray as little as
possible from Peter's existing script, which uses ypcat on auto.master
piped into wc -l already. Relying on fewer working parts.
> So, keeping with the check-ypcat-stdout approach:
>
> if ypcat $map 2>/dev/null | read dummy; then
>
> I'm a little more confident that that will work in likely Bourne shell
> variants.
>
Does this read force an early return like the head? Fine with me, if
everyone else agrees.
>
> --
> Mark Carmichael
Take care,
--
Camm Maguire [EMAIL PROTECTED]
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah
