Date: Wed, 2 Mar 2011 11:20:32 +0100 From: Emanuel Haupt <ehaupt@FreeBSD.org> To: Alexander Leidinger <Alexander@leidinger.net> Cc: emulation@FreeBSD.org Subject: Re: MASTER_SITE quality of emulation@, freebsd-emulation@ ports Message-ID: <20110302112032.7c2e65e7.ehaupt@FreeBSD.org> In-Reply-To: <20110301113555.59803j37iixnc7i8@webmail.leidinger.net> References: <20110226125814.3b0a46e5.ehaupt@FreeBSD.org> <20110227145125.12961olps6d8bapw@webmail.leidinger.net> <20110228101541.1926b275.ehaupt@FreeBSD.org> <20110228104240.32245ke5o34pco40@webmail.leidinger.net> <20110301100645.90f5eeb6.ehaupt@FreeBSD.org> <20110301113555.59803j37iixnc7i8@webmail.leidinger.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Alexander Leidinger <Alexander@leidinger.net> wrote: > Quoting Emanuel Haupt <ehaupt@FreeBSD.org> (from Tue, 1 Mar 2011 > 10:06:45 +0100): > > > Alexander Leidinger <Alexander@leidinger.net> wrote: > >> Quoting Emanuel Haupt <ehaupt@FreeBSD.org> (from Mon, 28 Feb 2011 > >> 10:15:41 +0100): > >> > >> > Alexander Leidinger <Alexander@leidinger.net> wrote: > >> >> Quoting Emanuel Haupt <ehaupt@FreeBSD.org> (from Sat, 26 Feb > >> >> 2011 12:58:14 +0100): > >> >> > >> >> > The distilator [1] shows that roughly 97% by a total of109969 > >> >> > all referenced distfiles fail to download. Resulting in > >> >> > countless timeouts. > >> >> > > >> >> > Excerpt from the most recent run: > >> >> > > >> >> > Maintainer ok bad % bad > >> >> > emulation@FreeBSD.org 1005 45696 97.85% > >> >> > freebsd-emulation@FreeBSD.org 1691 61577 97.33% > >> >> > >> >> > Ideas? > >> >> > >> >> Would it help to change the order of the master site list? I > >> >> assume that there are some which have everything. > >> > > >> > Unfortunately that wouldn't be much help to people who use > >> > MASTER_SORT. > >> > >> Yes. > >> > >> > I believe the best way would be to cleanup Mk/bsd.sites.mk. In > >> > particular MASTER_SITE_FEDORA_LINUX. > >> > >> I plan to have a look at a more recent linux distribution for our > >> linux_base. If I get the time, and if it will be fedora based > >> (probably it will, as it may make the update a little bit more > >> easy), this may mean the list will go back to what it was... > >> > >> > ports-mgmt/distilator can help identifying those sites. For > >> > instance: > >> > > >> > $ distilator /usr/ports/emulators/linux_base-f10/ > >> > > >> > Result: > >> > http://people.freebsd.org/~ehaupt/misc/linux_base-f10-distilator.log.gz > >> > > >> > A quick check would suggests to only keep: > >> > $ zcat linux_base-f10-distilator.log.gz | grep ^200 | awk > >> > '{print $3}' \ > >> > | perl -e 'use URI; while(<>) { chomp(); $u=URI->new($_); > >> > | print > >> > $u->host . "\n"; }' \ > >> > | sort | uniq > >> > > >> > archives.fedoraproject.org > >> > ftp.quicknet.nl > >> > ftp.rhd.ru > >> > ftp.udl.es > >> > mirror.steadfast.net > >> > > >> > ...but reducing MASTER_SITE_FEDORA_LINUX to those hosts shows > >> > that not even those > >> > mirrors are consistent. Hence my prior suggestion to get a > >> > complete > >> > >> :( > >> > >> > set of all distfiles > >> > needed by ports using MASTER_SITE_FEDORA_LINUX and mirror those > >> > on reliable mirrors. > >> > >> Should be easy to do (if a mirror is available). > >> > >> > I can assume that task if that's fine with emulation@ :-) > >> > >> I am fine with this. I try to deprecate the linux*f[6789]* ports > >> this week (I have to make up my mind if it will be one or two > >> months of grace time). If you just spend the time to have a look > >> at the fc4 and f10 ones, everything should arrive at a sane > >> situation after the expiration date. > > > > Ok, I think I'll look into it. I might wait until you're trough with > > expiring the said ports. > > It's up to you to wait or not, but with the following you can > download the relevant (official linux infrastructure for 7.x and 8.x > without the deprecated ones) distfiles directly: Done, thanks for your feedback. Emanuel
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110302112032.7c2e65e7.ehaupt>