CentOS 7: Want a functional loggerhead

[Kevin R. Bulgrien]

> -----Original Message-----
> From: Richard Wilbur [mailto:richard.wilbur at gmail.com]
> Sent: Friday, September 11, 2015 1:24 PM
>
> On Wed, Sep 9, 2015 at 5:00 PM, Kevin R. Bulgrien
> <kevin.bulgrien at freedomcte.com> wrote:
> [...]
> >   $ cat /etc/redhat-release && /home/app/bin/serve-branches ~/bzr/blah
>
> The only other thing that comes to mind right now is to try:
> $ cd /home/app/bin
> $ cat /etc/redhat-release && serve-branches ~/bzr/blah
>
> to make the CentOS 7 invocation look exactly like your CentOS 6
> invocation.  It will only effect the current path when running
> serve-branches, but that may change the availability of something it
> cares about.

I may be missing the point, but that literal command sequence won't work.
serve-branches wants to write its log to the CWD.  Obviously I have to
have some differences because CentOS 6 / EPEL supplies all loggerhead
dependencies in-distro.  CentOS 7 doesn't.  /home/app is just a place
to put python site-packages I built.  PYTHONPATH points into it.

I already ran the serve-branches tool from the same relative folder
(i.e. ~/bzr, ~, and ~/bzr/blah ) on both CentOS 6 and CentOS 7.
I've tested a lot of variations of invocation (not all of which
are documented in the question).  I'm pretty confident I ruled
out simple permissions issues.

Nevertheless...

$ cd /home/app/bin
$ cat /etc/redhat-release && serve-branches --log-folder=${HOME}/bzr ~/bzr/blah
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629983553280: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629975160576: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629966767872: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629958375168: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629949982464: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629941589760: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629933197056: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629453899520: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629445506816: Initial 
worker pool
DEBUG:paste.httpserver.ThreadPool:Started new worker 139629437114112: Initial 
worker pool
serving on 0.0.0.0:8080 view at http://127.0.0.1:8080
DEBUG:paste.httpserver.ThreadPool:Added task (0 tasks queued)
INFO:loggerhead:127.0.0.1 - - [11/Sep/2015:14:24:50 -0500] "GET / HTTP/1.0" 
404 - "-" "Lynx/2.8.8dev.15 libwww-FM/2.14 SSL-MM/1.4.1 OpenSSL/1.0.1e-fips"

This produces the same symptom (404 error).

I wish loggerhead/python/site-packages were able to be put into a much
more verbose logging mode so it was more easy to see where the problem
is being encountered.  I suppose they can be and I just don't know
how to do that at this time.  I tried `python -m trace ...` but I
had trouble identifying useful information in the list of
executed commands.

Kevin R. Bulgrien