Date: Mon, 17 Sep 2001 17:31:21 +0300 From: Peter Pentchev <roam@ringlet.net> To: .@babolo.ru Cc: Satoshi - Ports Wraith - Asami <asami@FreeBSD.ORG>, sobomax@FreeBSD.ORG, john_m_cooper@yahoo.com, gnome@FreeBSD.ORG, ports@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: bugbuddy build failure Message-ID: <20010917173121.A698@ringworld.oblivion.bg> In-Reply-To: <200109171417.SAA28643@aaz.links.ru>; from .@babolo.ru on Mon, Sep 17, 2001 at 06:17:24PM %2B0400 References: <yfkiteirhdf.fsf@vader.clickarray.com> <200109171417.SAA28643@aaz.links.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Sep 17, 2001 at 06:17:24PM +0400, .@babolo.ru wrote:
> Satoshi - Ports Wraith - Asami writes:
> > * From: Maxim Sobolev <sobomax@mail.ru>
> >
> > * Umm, no, bugbuddy already receives oaf dependency through USE_GNOME knob. From
> > * what I saw in the log I can tell that to me it looks like there is some more
> > * fundamental problem with bento itself, because it tries to build bugbuddy
> > * even though one of the packages it depends upon is missed (see piece of log
> > * below). I have no idea why it may be happening, though.
> >
> > As I reported before, it is sshd that is misbehaving. The package
> > building scripts are fine -- they have been handling dependencies the
> > exact same way for several years now.
> >
> > It's this piece of code that's failing.
> >
> > ===
> > while [ $# -gt 0 ]; do
> > if ssh -a ${master} [ -f ${packages}/All/$1 ]; then
> > if [ ! -f ${chroot}/tmp/depends/$1 ]; then
> > echo "copying package $1 for ${pkgname}"
> > scp -p $master:${packages}/All/$1 ${chroot}/tmp/depends
> > fi
> > else
> > echo "skipping package $1 for ${pkgname} since it is missing"
> > fi
> > shift
> > done
> > ===
> >
> > If the "ssh" command fails, it will not copy the necessary package.
> > The package *is* on the master when this happens. For instance, I
> > just got an error saying libiconv is missing, but it was built two
> > days ago according to the timestamp so there is no race condition.
> >
> > I have seen similar errors before, when Paul Saab told me that it was
> > ConnectionsPerPeriod in /etc/ssh/sshd_config that needed to be tweaked
> > to allow a very high access rate like the package cluster master.
> > However, that knob has gone away, (apparently) replaced by
> > MaxStartups. I changed it to 200:5:300 as shown below but the problem
> > still persists.
> >
> > ===
> > # ConnectionsPerPeriod has been deprecated completely
> >
> > # After 10 unauthenticated connections, refuse 30% of the new ones, and
> > # refuse any more than 60 total.
> > MaxStartups 200:5:300
> > ===
> >
> > I just noticed that there ase some of these messages:
> >
> > ===
> > rresvport: af=2 Resource temporarily unavailable
> > ===
> ssh -x resolves Resource temporarily unavailable problem
> This is nothing with MaxStartups
Of course, coupled with 'UsePrivilegedPort no'.
G'luck,
Peter
--
No language can express every thought unambiguously, least of all this one.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010917173121.A698>