Date: Thu, 22 Feb 2001 08:46:03 -0800 (PST) From: John Polstra <jdp@polstra.com> To: stable@freebsd.org Cc: all@biosys.net Subject: Re: cvsup confusion Message-ID: <200102221646.f1MGk3J27698@vashon.polstra.com> In-Reply-To: <4.3.2.7.2.20010221203901.00c8fd48@64.7.7.83> References: <4.3.2.7.2.20010221054413.00c443a0@64.7.7.83> <4.3.2.7.2.20010221054413.00c443a0@64.7.7.83> <4.3.2.7.2.20010221203901.00c8fd48@64.7.7.83>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <4.3.2.7.2.20010221203901.00c8fd48@64.7.7.83>, Allen Landsidel <all@biosys.net> wrote: > At 09:07 2/21/2001 -0800, you wrote: > > >There is a problem with this approach. Suppose everybody did it that > >way. Then at exactly the same time each week, everybody in a given > >time zone would be trying to CVSup from their nearby mirrors. The > >mirrors would quickly fill up, and you'd be retrying for hours, along > >with everybody else in your time zone. > > That starts off around a "probably true" and progresses on to "nearly > ludicrous." ;) [3 paragraphs of speculation omitted] That is one fine bunch of theorizing you've put together. Unfortunately, reality trumps theory every time. You are arguing with the guy who coordinates all 60 or so mirror sites, runs the master servers on freefall and cvsup-master, personally maintains 4 of the mirrors, and has logins on quite a few of the other ones. While you have been hypothesizing, I've been watching real mirror sites under real loads for almost 5 years. Oh yeah, and I wrote the software too. I know what to expect from it most of the time. The reality, based on observation, is that there are definite load spikes even now at the predictable times (like the top of every hour). If even a fraction of the users started doing their updates from /usr/local/etc/periodic, it would have a definite and observable effect. Nobody would be happy with the results. My condolences if that seems ludicrous to you. > >It's much better to pick a random time and put it in your crontab. > > I disagree. I think it would be "much better" if this was truly a > concern, to change the time that your periodics are executed, ... which is just a different way of achieving the same thing. > or to simply pick a different mirror that isn't saturated when your > default periodic runs. That isn't saturated _yet_. Mirrors tend to get more saturated as time goes on. I bet 99% of all users leave their crontab entries for the periodic scripts unchanged. So regardless of their time zone, they are running a 1 minute after some given hour (0301 in their local time zone). That's 24 possible starting times each day, instead of 1440. Many of the mirrors which are never saturated currently would become saturated at least several times a day under that scheme. > In any case, to the original poster, regardless of how you decide to > schedule your updates, you'll alleviate the problem by doing as was > suggested and reading the manpage, or (a better solution imho, but > reading the manpage anyway is always a great idea) condensing the > three seperate cvsupfiles you have into one and only invoking one > copy of cvsup to begin with. Yep, I agree with that! John -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Disappointment is a good sign of basic intelligence." -- Chögyam Trungpa 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?200102221646.f1MGk3J27698>