New seup: Reviewboard seems to fail on thinking SVN repo should be a checked out version

2014-07-21 Thread Michael Kovacs
I am plunging into setting up a reviewboard per request. At this time the general environment is that an SVN repository exists and is hosted on a Win2k8R2 server. The repo is using a self signed cert so I have been battling with the SSL issue for a bit and believe I have most of that figured ou

Re: New seup: Reviewboard seems to fail on thinking SVN repo should be a checked out version

2014-07-22 Thread Michael Kovacs
> > The SVNRoot\NetworkConfigs (my test SVN) is the data itself.. Forgive my > novice questions but from what you have said it seems like this would be a > task for the end user to have to do each time? Is there a way that when the > project is checked in (programmers use VS2010 & the Win tools

Re: New seup: Reviewboard seems to fail on thinking SVN repo should be a checked out version

2014-07-23 Thread Michael Kovacs
David, I seriously appreciate the help you are giving me on this issue OK so I have the syntax and implemented the --repository-url in the post-commit-run.cmd file, which is called by the actual post-commit.cmd on VisualSVN. As can be seen above, I was passing a set of authentication creden

Re: New seup: Reviewboard seems to fail on thinking SVN repo should be a checked out version

2014-07-23 Thread Michael Kovacs
I found a reference (of yours, actually!) which shows the syntax of the repo url to be "...=file://...". Tried the line as --repository-url=file:///data/SVNRoot/NetworkConfigs but back to the original issue then: >>> Current directory = C:\Data\SVNRoot\NetworkConfigs >>> Checking for a Subv

Upgrade from working 1.7.25 on Centos 65 fails

2014-07-30 Thread Michael Kovacs
Trying to get up to 2.x to work on the SSL certificate errors. Environment is: Centos 65 updated. Python 2.6 (yum installed) ReviewBoard 1.7.25 (yum installed) Setuptools are installed, so easy_install is available. Attempting to do ANY upgrade (easy_install -U reviewboard or reviewboard=1.7.2

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-07-30 Thread Michael Kovacs
> > OK so given the yum install is topped out at RB 1.7.25, Python at 2.6 and > others accordingly, would the best way to be rpm -e on everything and > reinstalling from scratch? Again, I have a snapshot of the running base > system so reverting is not an issue when I get down a dead end.. > O

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-07-30 Thread Michael Kovacs
> > Christian: > Again, just to be clear the machine is snapshot'd at the running 1.7.25 yum install, which is working. The steps above all related to getting 1.7.25 upgraded to 1.7.27; the snapshot in question had never been used to attempt a 2.0.x install. I will revert, then uninsta

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-07-31 Thread Michael Kovacs
OK wow... Went back to square one and reloaded everything to keep it clean. I believe the bulk of my problems were actually a combination of selinux & iptables; once I temporarily disabled them and tracked out a bunch of deps I have the 2.0.5 system up on a clean install (no yum reviewboard re

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-07-31 Thread Michael Kovacs
On Thursday, July 31, 2014 2:53:44 PM UTC-5, Christian Hammond wrote: > > During that run, is selinux still disabled? > > Can you run this in a shell: > > $ python > >>> import whoosh > >>> whoosh.__file__ > >>> import reviewboard > >>> reviewboard.__file__ > > $ python >

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-07-31 Thread Michael Kovacs
On Thursday, July 31, 2014 3:26:08 PM UTC-5, Christian Hammond wrote: > > It’s definitely valid syntax. > > Make sure the __ is two underscores. > > Can you copy/paste that to a text file and attach it, so that the > formatting is preserved? > > Christian > > > > > > Doing it with notepad as a m

Re: Upgrade from working 1.7.25 on Centos 65 fails

2014-08-01 Thread Michael Kovacs
> > Looks like a lot of the battle is with SELINUX... I now have 2.0.5 running > and it seems to be happy in its little world. Now to start tying it into > the VisualSVN server again. At this time I have a snapshot of the unit in > good condition however do you have an SELINUX exception that al

ReviewBoard 2.0 & VisualSVN - Tying it together

2014-08-07 Thread Michael Kovacs
OK so first off thanks for all the help getting things this far; we at least are up and running with RB itself! Now to tie it all together. Versions: RBTools 0.6.2 VisualSVN server v2.5.7 Standard Edition on Windows 2008R2 ReviewBoard 2.0.5 on Centos 6.5 Submitters will be running AnkhSVN integr

Re: ReviewBoard 2.0 & VisualSVN - Tying it together

2014-08-07 Thread Michael Kovacs
Thanks David. None of the options worked... Here is three different configs and the output of a snip of the post-commit.log file (names changed to protect the innocent): With parameters --repository-url=file:///c:/data/svnroot/NetworkConfigs >>> Making HTTP POST request to http://itt-d-review.

Re: ReviewBoard 2.0 & VisualSVN - Tying it together

2014-08-07 Thread Michael Kovacs
No joy there either. Tried with and without the repo name in the file:// call >>> Checking for a Subversion repository... >>> Running: svn info file://C:/data/SVNRoot/svn/ --non-interactive >>> Command exited with rc 1: ['svn', 'info', 'file://C:/data/SVNRoot/svn/', '--non-interactive'] svn: E18

Re: ReviewBoard 2.0 & VisualSVN - Tying it together

2014-08-07 Thread Michael Kovacs
>From a win8 desktop, I do a very simple edit of one text file: svn co https://itt-d-cvs1.abc.local/svn/NetworkConfigs cd NetworkConfigs notepad test.txt (make small change in file) svn ci -m "0807TestChgXX" The rest is done in the post-commit hook itemized above. Just for clarity, this is the

Re: ReviewBoard 2.0 & VisualSVN - Tying it together

2014-08-11 Thread Michael Kovacs
This seems to be getting overly complex... With the RBTools installed on the VisualSVN server running Win2008R2, and the certificate now a valid cert instead of self gen'd, should the rbt call be running as a post-commit? It seems like what is happening is the RBT call is trying to access the