On Jul 3, 2008, at 20:46, Robert Dionne wrote:
I've been using the 0.7.2 version for some time now and it's rock
solid. There were issues getting it all built on OSX, etc.. but I
run it for days at a time without even thinking about it.
I'm using the Java client on one end to run some legacy code. I'm
essentially using couchdb as a poor mans RDF store. On the other end
I have an Erlang based app that is loading large RDF graphs where
each node is running in it's own process. It uses erlcouch to talk
to couchdb. I'm not able to discuss the app, it's a research
prototype.
I'm a fan of couchdb. I caught the Erlang bug three years ago and
have tried to use it whenever I can. I think couchdb is a very
timely project and have always thought the best solution to O-R
mapping is get rid fo the R :). I've poked at the code some and am
hopeful to spend more time with it, and even help out if I can.
I just bit the bullet and built 0.8 this morning, knowing erlcouch
and the Java clients would probably break as there's been a lot of
good work since 0.7.2, surprisingly all my stuff more or less works.
Great story! Thanks for sharing!
I did run the unit tests. Under FIrefox, with the exception of
replication (see below) test they all ran. However when running with
Webkit it gets to the "attachments" test and Webkit goes off the
deep end, cpu at 99%, etc... I'm running MAC, dual core OSX 10.4.1
Both are known issues. We haven't got around to fix the WebKit weird
behaviour yet, since FF works, this is not a priority. The replication
problem surfaces when IPv6 is enable which it is on OS X by default
IIRC so localhost will resolve to :::1 where CouchDB is not listening.
Using http://127.0.0.1:5984/ to access the test suite or disabling
IPv6 dose the trick here. So all is well with your installation :)
Again, thanks for sharing!
Cheers
Jan
--
All the best,
Bob Dionne
[info] [<0.2798.0>] HTTP Error (code 500): {'EXIT',{{badmatch,
{error,econnrefused}},
[{couch_rep,do_http_request,3},
{couch_rep,open_doc,3},
{couch_rep,replicate,3},
{couch_httpd,handle_replicate_request,2},
{couch_httpd,handle_request,2},
{mochiweb_http,headers,4},
{proc_lib,init_p,5}]}}
[info] [<0.2798.0>] 127.0.0.1 - - "POST /_replicate" 500
On Jul 2, 2008, at 5:55 AM, Jan Lehnardt wrote:
Heya,
in a recent discussion with somebody who follows these
lists, he pointed out that CouchDB appears to have a lot
of problems and has some sort of negative aura. Mostly
because the issues discussed on these lists are about
problems people have with CouchDB.
Obviously, when looking at other information sources,
it is clearly visible that CouchDB is loved by many and
used successfully.
Yet these stories don't make it to the mailing lists. So
please, if you have a CouchDB success-story to share,
let us know. If that means shamelessly advertising
a project or product, just go ahead and do so :)
Anything from your personal blog to the "next big
thing" is interesting to us.
To let us know, just reply to this post!
If you can't talk publicly about what you do in detail, a
message like "We are using CouchDB successfully in
our sub-radar startup" (I know you are there ;) will do.
If you don't want to post to the list directly, send me a
private mail and I can relay the totally anonymous
information then.
--
If you have a problem with CouchDB, don't let this
post discourage you from posting it here, we'd love
to hear about any issues you might encounter to
make your life easier in the future.
Cheers and thanks in advance!
Jan
--
PS: If you find this a useless exercise, just ignore me :-)
