This is all covered in the administration guide. See the section on
upgrading.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board - http://www.reviewboard.org
VMware, Inc. - http://www.vmware.com
2010/3/2 JohnHenry qhlonl...@163.com
Thanks Christian. There is a version on our
Off-hand, I'm not sure what will happen when you try this, but you can go
ahead and give it a try. If it does crash, feel free to file a bug and we'll
look into it.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board - http://www.reviewboard.org
VMware, Inc. -
Hi,all
We want the reviewboard to run faster. We found that on some
functions like view_diff and view All review reqeusts, the cpu
consume on apache (Reviewboard server) cost 90%, and less then 10% on
MySQL, So it seems that the bottleneck is not lie in Too Much SQL
Execution. We tried
Hi,Christian,
I had tried with an inexistent review-request-id 1000, Then the
post-review will reposrt error info as follows:
---
Looking for '192.168.3.155 /' cookie in
Thanks Christian,
I've looked into the Apache and the reviewboard logfiles (see below).
There isn't much in them. Nothing related to the creation of the
folders. There are some issues in the apache log when trying to
display pictures that are not present.
I had set Debug = True in the RB config.
Oh ok - thanks. :)
On Mar 2, 3:15 pm, Christian Hammond chip...@chipx86.com wrote:
The URLs should be fine. Both domains point to the same place.
There was a server issue. We've had a lot of traffic lately and Apache seems
to have died again last night. Fixing and restarting at the moment.
Still can't get this to work:
# load_rb_env.py
import os, sys
import reviewboard directory from site-packages
# Append the main RB python directory.
sys.path.append(reviewboard.__path__[0])
# Append the RB site directory.
sys.path.append(/srv/reviews/test/conf)
# Import the RB environment.
Go figure; this seems to work:
sys.path.append(/srv/reviews/test/conf)
os.environ['DJANGO_SETTINGS_MODULE'] = 'settings_local'
from django.contrib.auth.models import User
users = User.objects.all()
for user in users:
print user.username
On Mar 3, 1:17 pm, Stodge sto...@gmail.com wrote:
All,
I'm getting the following error on a recent ReviewBoard on Apache
+mod_python install on RHEL 5.1. I'm wondering if this is a django or
mod_python version issue. All relevant permissions are set such that
apache (the process owner) has full rwx to /var/rb/mysite directory.
Mod_python error:
We have ReviewBoard set up with ActiveDirectory but users are
uncomfortable with their passwords being snoopable on the wire. I
therefore want to make it fully supported over HTTPS - all web access
is through https, the email notifications have https links in them
etc.
Is there any existing
Hi,
I haven't been upgrading RB for quite some time now and on RB web page
I saw some information about GitHub feature. Unfortunately it wasn't
clear for me from what I read in the release summary if RB 1.5 B1
supports external/remote git repositories hosted for example on
Gitorious.
I know it
Considering all Visual Studio 2008 new source code files have initial
unicode BOM characters, any diff of source files from a default VS
install should do.
I am surprised *anyone* that uses VS can use post-review.
I set up a new user on the latest RBTools and had them modify a
checked in file and
A lot of what Review Board does is computationally expensive, and we've done
a lot to heavily optimize things. Best way to get speedup is to run it on a
faster server with more memory and more caching. We don't support anything
like psyco or other accelerators.
Christian
--
Christian Hammond -
One thing you might want to try, as I hit this once myself. Go into the
settings and the File Storage section. Make sure the values are what you'd
expect, and then, even if they are, hit Save. Just in case, restart Apache
after that.
We had a problem once (maybe it's still here) where, after an
If you're not using memcached, that will be a huge gain.
-David
On Wed, Mar 3, 2010 at 3:37 PM, Christian Hammond chip...@chipx86.com wrote:
A lot of what Review Board does is computationally expensive, and we've done
a lot to heavily optimize things. Best way to get speedup is to run it on
Sounds like it's not seeing Review Board in the Python path. From a command
line, try typing:
$ python
import reviewboard
See if that succeeds. If it does, type:
import reviewboard.settings
If that tells you it can't find settings_local.py, then that's fine. If it
can't find
The red rectangle doesn't come from us. This is a Pygments thing, so you'd
need to talk to them about changing that. We have no control over it.
It seems that change to post-review for encoding in UTF-8 breaks a lot of
things, so I'm going to remove it and look into an alternative fix.
Christian
Python 2.4.3 (#1, Jul 2 2009, 15:50:36)
[GCC 4.1.2 20080704 (Red Hat 4.1.2-44)] on linux2
Type help, copyright, credits or license for more information.
import reviewboard
import reviewboard.settings
Unable to read settings_local.py.
Please see http://www.reviewboard.org/docs/manual/dev/admin/
Okay, do that again and then type:
print reviewboard.__path__
It looks like what's happening is that, one way or another, Review Board
isn't in the system path that mod_python is searching.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board - http://www.reviewboard.org
Yes, I'm using 1.5 beta 1.
And yes, the api token is from my account (the owner of the
repository). The following is a screenshot for greater clarity to see
if I have anything wrong :
http://skitch.com/nbauman/n2gif/screen-shot-2010-03-03-at-4.04.52-pm
Logging is turned on and permissions look
Yes, earlier I tried explicitly including the path to ReviewBoard in
my PythonPath, and got past this error, but then apache/mod_python
started complaining that they couldn't find djblets, which had me
worried that i'd start having to include the absolute path to every
missing library - not
I don't know. It's possible there's something funky with RHEL and Python
paths when using mod_python, but I don't have an answer there. It'd be
helpful if someone else on RHEL 5.1 can chime in on this.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board -
Trying to install reviewboard.
# easy_install -v P4PythonInstaller
Searching for P4PythonInstaller
Reading http://pypi.python.org/simple/P4PythonInstaller/
Reading http://www.review-board.org/
Found link:
http://reviewboard.googlecode.com/svn/trunk/P4PythonInstaller#egg=P4PythonInstaller-0.2
Sorry for the confusion - it's actually RHEL 4.1.2-44. It's funny
because I have a box with the exact same OS running just fine.
On Mar 3, 8:16 pm, Christian Hammond chip...@chipx86.com wrote:
I don't know. It's possible there's something funky with RHEL and Python
paths when using mod_python,
Looks like a temporary network issue. I can connect to that SVN server just
fine. Maybe try it again, or check your network setup?
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board - http://www.reviewboard.org
VMware, Inc. - http://www.vmware.com
On Wed, Mar 3, 2010 at 5:21
Could it be that my generate Raw file URL mask url is too long? When i
switch to custom it has the value filled out for me, but the
validation doesn't work -
Ensure this value has at most 128 characters (it has 131).
On Mar 3, 8:02 pm, Nik Bauman n...@foodzie.com wrote:
Yes, I'm using 1.5 beta
Good thought! I investigated the git setup - everything looks fine
though.
git ls-remote g...@github.com:user/repo.git HEAD works fine.
On Mar 3, 9:14 pm, Christian Hammond chip...@chipx86.com wrote:
Oh, is Git installed on the server? Maybe it's failing its `git ls-remote
repository path
So if you call that on the server, it works?
That's the check we're actually performing when it comes down to it. If it
fails, we report that the repository is invalid. Otherwise, it should allow
it to continue.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board -
Correct,
CLI output:
:/home/nbauman$ git ls-remote g...@github.com:foodzie/foodzie.git HEAD
f5546a2dea7746f7670f04ddc5d94c814f4cHEAD
:/home/nbauman$
App response:
http://skitch.com/nbauman/n28se/screen-shot-2010-03-04-at-1.17.48-am
:(
-Nik
On Mar 4, 1:13 am, Christian Hammond
Is logging turned on? If not, try turning it on in Settings and restarting
Apache. When the ls-remote fails, it should log the reason why.
I committed a change to bump up the max length for the raw URL field, so in
the next nightly rebuild (in about 2.5 hours) you may want to upgrade.
Christian
Aha!
I figured it out. It was a public key issue. My apache user (www-data)
needed to authenticate with github using my public key and my key was
only in my user's account. So when the web app tried authenticating,
it was puking.
On Mar 4, 1:40 am, Christian Hammond chip...@chipx86.com wrote:
Well, not in the web interface. The log had the details that I needed
(turns our I was embarrassingly looking in the wrong location for the
log):
2010-03-03 22:51:05,394 - ERROR - Git: Failed to find valid repository
ssh://g...@github.com/foodzie/foodzie.git: Host key verification
failed.
fatal:
Looks like our SSH verification code doesn't kick in for the Private GitHub
URLs. I'll have to fix that. It should have handled all that from the web
UI.
Christian
--
Christian Hammond - chip...@chipx86.com
Review Board - http://www.reviewboard.org
VMware, Inc. - http://www.vmware.com
On Wed,
hi,
sorry for the delay in replying.
i checked the path you specified and i see a known hosts file created
there. it includes the cvs host too. (i am sure this was created by
reviewboard)
When i try to add the repo, it asks me whether i trust that host and i
clicked yes. The hang happens
Status: New
Owner:
Labels: Type-Defect Priority-Medium
New issue 1525 by jamesmcdougall: Function to expand hidden lines in a diff
does not work for 1.5 Beta 1
http://code.google.com/p/reviewboard/issues/detail?id=1525
*NOTE: Do not post confidential information in this bug report.*
Comment #6 on issue 769 by rponton: post-review should warn if there's
no diff in the user's path
http://code.google.com/p/reviewboard/issues/detail?id=769
This needs to be made a configuration option. The default diff util in my
default
path is a crappy old version that is not
Comment #6 on issue 1500 by trowbrds: xx lines hidden [Expand] link
stopped working
http://code.google.com/p/reviewboard/issues/detail?id=1500
Issue 1525 has been merged into this issue.
--
You received this message because you are listed in the owner
or CC fields of this issue, or because
Updates:
Status: Duplicate
Mergedinto: 1500
Comment #1 on issue 1525 by trowbrds: Function to expand hidden lines in a
diff does not work for 1.5 Beta 1
http://code.google.com/p/reviewboard/issues/detail?id=1525
This is fixed in the nightlies.
--
You received this message
Comment #20 on issue 1060 by trowbrds: easy_install P4PythonInstaller
*Error*
http://code.google.com/p/reviewboard/issues/detail?id=1060
The error you just posted looks like gcc isn't fully installed. Maybe
you're missing
C++ compilers?
--
You received this message because you are listed
Comment #21 on issue 1060 by hbetts3: easy_install P4PythonInstaller *Error*
http://code.google.com/p/reviewboard/issues/detail?id=1060
Trowbrds,
the gcc was installed but not the g++. I installed that and now everythign
works.
--
You received this message because you are listed in the
Comment #7 on issue 769 by chipx86: post-review should warn if there's
no diff in the user's path
http://code.google.com/p/reviewboard/issues/detail?id=769
The plan after the RBTools 0.2 release is to bundle a custom diff tool that
we can
use instead of GNU diff for this.
--
You received
41 matches
Mail list logo