Stefano Mazzocchi wrote:
Sure: the CVS/list services on cocoondev.org would only be for non-Apache Cocoon-based projects.Sylvain Wallez wrote:I like this idea, as we need, as shown by cocoondev.org, to host some
live Cocoon apps. But we *must* also use the common infrastructure
provided by daedalus and icarus when it makes sense : static website,
distros, mailing lists and cvs (IMHO including the ones of
cocoondev.org).
Absolutely. It would be plain silly to redo something that already works great (CVS/mail/static web are just great and managed very well, IMO).
What we lack is the ability to host our own web applications and this is what we want to fix.No worries here too: while Cocoon is a great project & community, I would never started working with it if it wasn't an Apache project. And I'd opt for the proxypass solution rather than switching DNS.
This ensures we won't be tempted by a "Cocoon software
foundation" (even if I'm sure Stefano will take great care to avoid
this), save bandwidth, CPU and administration costs for people/entities offering resource for live applications.
Oh, you can bet your ass here :)
From daedalus:Proxypassing will also allow to have _several_ machines behind cocoon.apache.org. This can provide simple load partitioning and allow different members of the community to offer CPU and bandwidth (nothing sure for now, but I'd like my company to offer some).
Yep. Also, as Steven and I discussed on the phone, most of the bandwidth is used by distributions and Pier told me that infrastructure@ is building a pretty nice rsync-based mirroring system.
So it makes perfect sense to use the ASF machinery when it works and just *patch* what we need.
However, I wonder if proxypassing from San Diego (IIRC this is where icarus and daedalus are) to Ghent or another european location is technically ok?
Don't use geographical topology as a metric. Steven, what is the network topology between the ASF machines and the one we would be using?
bash-2.05a$ traceroute cocoondev.org
traceroute to cocoondev.org (209.15.201.32), 64 hops max, 44 byte packets
1 fa3-13.br1.sfo.collab.net (63.251.56.130) 0.293 ms 0.239 ms 0.161 ms
2 200.ge2-0.er2b.sjc3.us.mfnx.net (64.124.78.35) 0.657 ms 0.280 ms 0.438 ms
3 so-1-2-0.cr1.sjc3.us.mfnx.net (208.185.175.193) 0.554 ms 0.406 ms 10.012 ms
4 pos9-1.mpr1.pao1.us.mfnx.net (208.184.232.178) 0.940 ms 0.891 ms 0.877 ms
5 64.124.11.198.cogent-above.above.net (64.124.11.198) 1.464 ms 1.525 ms 1.466 ms
6 p6-0.core02.sfo01.atlas.cogentco.com (66.28.4.149) 1.512 ms 1.948 ms 1.577 ms
7 p15-0.core01.den01.atlas.cogentco.com (66.28.4.129) 43.539 ms 43.037 ms 43.010 ms
8 p5-0.core01.mci01.atlas.cogentco.com (66.28.4.30) 43.073 ms 42.990 ms 43.088 ms
9 g50.ba01.b006121-1.mci01.atlas.cogentco.com (66.28.65.130) 43.142 ms 43.001 ms 43.050 ms
10 hosting4u.demarc.cogentco.com (66.250.5.98) 56.774 ms 56.990 ms 56.631 ms
11 core2-POS3-0.KCY.hosting4u.net (209.15.6.97) 58.486 ms 56.611 ms 56.589 ms
12 net2-vlan1-sup2.KCY.hosting4u.net (209.15.116.246) 56.965 ms 57.326 ms 56.902 ms
13 fw.kc.aoindustries.com (209.15.126.2) 56.802 ms 57.651 ms 57.188 ms
14 cocoondev.org (209.15.201.32) 56.805 ms 57.370 ms 57.527 ms
The machine is located in Kansas City, US, not in Ghent. And maybe we can tweak routing tables from daedalus to cocoondev.org.
Ping times are averaging 60ms. Cannot test from nagoya, don't have an account there. While it doesn't reach the level of interconnectedness as daedalus & nagoya, I think the interconnection is more than decent enough for a proxypass solution. Let's ask the network gurus of [EMAIL PROTECTED] (I feel like in a 'coming-out' mode anyhow :-)
</Steven>
--
Steven Noels http://outerthought.org/
Outerthought - Open Source, Java & XML Competence Support Center
[EMAIL PROTECTED] [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
