GitLab

Fix XML. Don't try to canonicalize to print. That has too much unwanted baggage. Create a copy of the node instead which will automatically add namespace declarations to it.

Ensure that version 1.70 or later of the p5-XML-LibXML library is installed. There is a pretty nasty bug in earlier versions.

instead of the local hostname, so that the scripts work when located
on your local desktop machine.

Print canonicalized strings as output. If you don't the namespaces->prefix bindings won't always be printed if you print out a fragment.

Fixed Geni XML parse bugs. Added additional checks for absent tags. Fixed bug where the ticket was created using the original string rather than the updated rspec.

Added component_manager information to links in both versions 0.1 and 2 of the rspec. The request-tunnel.xml example uses component_manager tags on links.

"ifdef FOO" is always true if FOO is not the empty value, so use 'ifeq
(FOO, 1)" instead when checking if support for each filesystem is
enabled.

dhcpd isn't running at this time so invoking dhcpd_makeconf will blow up
the elabinelab setup.

Make sure power doesn't blow up if rmcp code is not installed.

When getting things working for boss/ops config, forgot to also make changes
here for boss/ops/fs config.

Since we don't currently load the Linux-based tools into an inner elab,
we have to munge DB state in the inner elab so that all nodes use the
standard BSD-based tools.

Reign in a rogue ifdef in the makefile.

The current version of graphviz "dot" appears to put out positions as
floating point numbers rather than integers.  Change to accomodate.

Add missing argument to RecordVlanInsertion().

Don't return 1 from RemoteDoList() on error.  The caller is expecting a list
and not any sort of error.  Instead, on error we tbdie() which is what other
backend modules do on errors.

I cannot seem to get the dependency into the boss and ops ports.

Some of these daemons could go nuts when accounts, etc. disappear while
they are running.

changes and subsequent bug fixes.

auth_mysql module (which is installed under different names depending on the
version).

This could have been done with configure or in boss-install.  I didn't do
the former because configure might get run before apache is installed.  I
didn't do the latter because you would have to remember to re-run configure
if you updated your mod_auth_mysql port.

variables in libtblog_simple and SWIG generated code.

The fix is to libtblog, to get rid of the warning.  The warning from swig
generated code is a swig problem that was supposedly fixed, but apparently
not.  So those warnings remain, and led to the workaround.

The workaround is in shommlists which parses output captured by SUEXEC.
That function captures all output, stdout and stderr, from the command it runs
and presents it in an array.  So if you use it to invoke a perl script that
evokes the "used once" messages, those messages wind up in the array.
showmmlists was not validating that output at all, it just assumed everything
returned was the name of a mail list.  Added some syntactic validation
(aka a regexp) to deal with that.

All we really do is apply one non-standard patch to avoid SSL certificate
mismatches due to "www.emulab.net" vs. "boss.emulab.net".  In older versions
of m2crypto we just patched the installed library files, but the newer versions
get installed as a .egg file, so it is easiest to just patch it while the
port is being built.

I really don't like creating a separate port for this, but the alternative
was to modify the standard port, adding a new option.  That seemed even worse.