On Aug 4, 9:27 pm, Christian Hammond <chip...@chipx86.com> wrote:
> On Wed, Jul 29, 2009 at 3:58 AM, gbjbaanb <andy.bolstri...@gmail.com> wrote:
> > Hi everyone.
> > I'm a very new user of ReviewBoard, I've just set it up, and as you
> > asked for information on how I intend to use it, and suggestion for
> > improvement... I'd thought I'd post now while my initial thoughts are
> > still fresh.
> Feedback is always welcome!
> So far... I've got it set up and it looks good, there were a few pains
> > getting going though:
> > 1. in my setup, I only have a single DNS entry for the webserver. I'm
> > using it in a corporate environment, and I don't want to set up a new
> > box just for RB, so I thought I'd install it on the existing web
> > server (which runs a few other development-related sites such as
> > mantis). Unfortunately the installer assumes that it will be running
> > as a vhost, and takes over the entire web on port 80. If this could be
> > changed to a directory instead of a vhost, I think that would be a
> > good improvement to make. Currently I have it running on a different
> > port, as I cannot affect the other sites (and I'm not too hot on
> > mod_python requirements)
> Most of the time, a vhost suffices and is the easiest to work with. It's a
> sample configuration file, so we expect people who are using more custom
> setups to customize the config file. We can probably look into generating
> other forms of config files in the future, but it's the kind of thing we'd
> prefer to take a patch for.
> > 2. When installing, the dependancy on PyLucene made me decide not to
> > bother with search :) Would it be a good idea to try using Clucene
> > instead as it might be easier to get installed, and possibly faster/
> > efficient to run?
> PyLucene is definitely a pain to install. I wish there was something we
> could do about that.. Clucene isn't an option, because it's, well, not
> Python. We need to be able to hook into it, and need Python bindings for
> this, which is what PyLucene is.
> We may look into alternative, pure-Python search engines that we can use in
> the future, though they will almost surely be slower than PyLucene. Maybe
> that's okay though
> 3. When I installed it, I immediately had a 'taking a nap' error.
> > Looking at the logs showed a problem with the python egg cache.
> > Setting the environment variable inside the vhost directive fixed
> > this, but it was an annoyance that it didn't work out of the box.
> > Also, the "error message" could be worded better, 'taking a nap'
> > suggests a delay, not a fatal error!
> Which config file did you use? The python egg cache should have been set
> correctly in generated config files.
> The errors are something we're working to improve. That particular error is
> from back in the day when Review Board was used in only 2 places:
> review-board.org and VMware.
> > 4. LDAP support. I access active directory from a linux box, no
> > problem with that except that the corporate AD doesn't allow anonymous
> > reads. Most other sites I use from my linux server (eg wikipedia,
> > mantis) allow you to set a ldap proxy user to connect with. That user
> > then attempts to search for the user specified and logs you in. I
> > couldn't see any option for this in the web UI, is it supported?
> I don't know a lot about LDAP. Our support comes from third parties who need
> LDAP support. So, I can't really answer this, except to say that I don't
> believe we have support for a proxy user, and we'd need a patch to support
1. vhost: I'll have a play and see if I can fix it, if so I'll send
you a patch.
2. Lucene: I thought it had python bindings already, but I can't find
much info on the web. I may have a look at converting the perl or php
bindings to python when I have some time.
3.egg cache, I used the conf/apache_modpython.conf file. (I had to
copy it as I use CentOS - you copy the file into /etc/httpd/conf.d to
get apache to pick it up). I can't find what version of RB I am using
in the installed code (I thought it'd be the latest though).
4. LDAP... I don't know python, so I can't help much. The idea behind
the proxy user is that you log on as that user and perform a search
for the user you want to log on as. If that succeeds, you use the same
connection to bind as the user with the user's password. Lots of
people on this thread have wanted ldap, might I point you in the
direction of http://www.python-ldap.org/docs.shtml
You received this message because you are subscribed to the Google Groups
To post to this group, send email to firstname.lastname@example.org
To unsubscribe from this group, send email to
For more options, visit this group at