Re: bin/49593: lib/libpthread/t_resolv:resolv

[christos%zoulas.com@localhost (Christos Zoulas)]

On Jan 21, 12:00am, Garrett%NetBSD.org@localhost (Garrett%NetBSD.org@localhost,Cooper%NetBSD.org@localhost) wrote:
-- Subject: bin/49593: lib/libpthread/t_resolv:resolv

| >Number:         49593
| >Category:       bin
| >Synopsis:       lib/libpthread/t_resolv:resolv
| >Confidential:   no
| >Severity:       non-critical
| >Priority:       low
| >Responsible:    bin-bug-people
| >State:          open
| >Class:          sw-bug
| >Submitter-Id:   net
| >Arrival-Date:   Wed Jan 21 00:00:00 +0000 2015
| >Originator:     yaneurabeya%gmail.com@localhost
| >Release:        n/a
| >Organization:
| EMC / Isilon Storage Division
| >Environment:
| FreeBSD fuji-stable-10.local 10.1-STABLE FreeBSD 10.1-STABLE #11 r276486: Wed Dec 31 17:40:39 PST 2014     ngie@fuji-stable-10.local:/usr/obj/usr/src.svn/sys/FUJI  amd64
| >Description:
| I've seen resolve_test fail once before on FreeBSD CURRENT and 10-STABLE with the following error (from kyua report-html):
| 
| <h2>Standard error</h2>
| 
| <pre>h_resolv: Error: host centaurus.4web.cz ask 2 got 1
| 
| </pre>
| 
| I hit it once before in the past in 11/07/2014, and hit it today with the same host and output.

It means that getaddrinfo() failed once to resolve that host.
The test picks hosts to resolve from a host list randomly, so it
might try the same host more than once. We can make print more information
(keep the error return from getaddrinfo()), but it is not worth it. Perhaps
you have a flakey internet connection?

christos