Conflicts:
	software/erp5/instance-erp5-input-schema.json
	stack/erp5/buildout.cfg
	stack/erp5/haproxy.cfg.in
	stack/erp5/instance-erp5.cfg.in
	stack/erp5/instance-mariadb.cfg.in
	stack/erp5/instance-zope.cfg.in
	stack/erp5/instance.cfg.in
	stack/monitor/buildout.cfg
	stack/monitor/monitor.py.in
	stack/slapos.cfg

Zope integrated Products.signalstack back in 2010:

    https://github.com/zopefoundation/Zope/commit/16796274
    ( Integrated the Products.signalstack / z3c.deadlockdebugger
      packages. You can now send a SIGUSR1 signal to a Zope process and get a
      stack trace of all threads printed out on the console. This works even
      if all threads are stuck. )

In 404150b7 (wendelin: Teach -dev build to dump all threads traceback on
SIGUSR1) it was my mistake to add Products.signalstack.

/reviewed-by TrustMe
/cc @Tyagov

This fixes the following random failures:

While:
  Installing slapos-deps-eggs.
  Getting distribution for 'ZODB3==3.10.5+SlapOSPatched001'.

An internal error occurred due to a bug in either zc.buildout or in a
recipe being used:
Traceback (most recent call last):
  ...
  File "zc/buildout/easy_install.py", line 968, in _get_dist
    shutil.copytree(dist.location, newloc)
  File "shutil.py", line 177, in copytree
    os.makedirs(dst)
  File "os.py", line 157, in makedirs
    mkdir(name, mode)
OSError: [Errno 17] File exists: '.../eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg'

By default, this is only true when a partition has a dedicated unix user.
Inside a webrunner, we can have:
  ${buildout:directory} = /srv/slapgrid/slappart0/srv/runner/instance/slappart1
  but $HOME = /srv/slapgrid/slappart0

Zope partitions override HOME as follows:
  home-path = ${buildout:directory}

The link was pointing to the RSS url itself instead of
pointing to the monitoring interface.

also add --disable-host-tool option to prevent possible installation
error that happens when pkg-config is rebuild without removing
parts/pkg-config directory.

Fixing almost random error: cannot get `parentNode` of null