freebsd-update painfully slow - slower than source code build of world and kernel

[Daniel Bond]

Hi Stefan.

Yes, I am also noticing this. Luckily interrupting it and starting it  
again resumes. Judging from the speed of http://www.daemonology.net/  
(hosted on same site), the freebsd-update server must be absolutely  
hammered.



On Jan 6, 2009, at 9:50 AM, Stefan Miklosovic wrote:

> Hi,
>
> My opinion is same. I tried to upgrade from 7.0-RELEASE to 7.1- 
> RELEASE but even after
> copying all the stuff from 7.0-RELEASE CD (src etc) and having  
> GENERIC kernel in /boot/,
> "freebsd-update upgrade -r 7.1-RELEASE" started to work properly but  
> hase not done its work.
> All tries stopped at some failure during a downloading. I have been  
> trying this about half a day,
> three times, but no change :((
>
> On Tue, Jan 6, 2009 at 9:03 AM, Daniel Bond <db at danielbond.org> wrote:
> Hi,
>
> I'm not sure where to post this, I had trouble finding a suitable  
> mailing-list. Please point me in the right direction, if this is the  
> wrong place to post this message.
>
> First off, I love the binary update tool for FreeBSD. It is an  
> excellent tool, and saves a lot of time and trouble compared to the  
> old method (or so I thought, until recently).
> I also like seeing the freebsd-update method is in the release notes  
> for 7.1-RELEASE, as a official way to upgrade a system.
>
> Yesterday I was struck by happiness, as I noticed 7.1-RELEASE was  
> out on ftp.freebsd.org - and decided to start off by upgrading one  
> of my companies development servers.
> Usually an update with FreeBSD-update is quite quick, but today and  
> yesterday it has just been to slow to use, after two days of trying  
> - I've still not completed a single upgrade. The
> server in question is connected to gigabit internet.
>
> I think it is embarrassing that the binary update tool, is actually  
> slower to use than compiling the whole operating system and kernel -  
> even on a slow machine! The reason for this,
> is not the tool it self, the tool is excellent - but there are no  
> mirrors.. We need some mirrors, or such a great tool is not really  
> usable at all (except for the really patient).
>
> This also goes for portsnap. Portsnap is also an excellent tool, but  
> the experience from using it could be much better. The european  
> portsnap mirror is actually slower, than the one in the US.
> I've been in contact with Colin, twice, about hosting another  
> portsnap mirror. Using a proxy server, does not cut it - not for my  
> use, sorry. I tried it, it didn't help. The last time I didn't  
> receive an
> answer.
>
> As I was saying to Colin, both myself and a friend who works for the  
> Norwegian government, should be able to run a mirror for portsnap on  
> good bandwidth. Many other people have offered
> to host mirrors, why is having mirrors a bad thing?
>
> I know the 6.4 and 7.1 releases have very many patches, due to  
> conversion from CVS to SVN. I have previously upgraded servers in  
> Norway and UK to 6.4-RELEASE with freebsd-update,
> and speed has been acceptable, not great, but enough to keep me  
> using and loving the tool. Still, I think more people will use  
> freebsd-update, since it is more practical to use, especially for
> non homogenous environments.
>
> Hopefully this will improve in the future, I don't mean to come  
> across as a whining grunge, but it is quite frustrating to me, as a  
> loving freebsd user.
>
> Congrats on a new release, I will be using it in a another day or so  
> (or whenever freebsd-update is done - maybe I will eat my own words,  
> and just do a regular build)!
>
>
>
> Best regards,
>
>
> Daniel Bond.
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org 
> "
>