GitLab

run through configure.

nodes itself.

but the boss one is special. It does not listen on the standard port,
since we are going to transparently forward that port to ops. Instead,
it listens on localhost:@BOSSEVENTPORT@ which is where all the
internal event based programs on boss are going to connect. Note, this
file should not be installed yet ...

just a bare skeleton of the client side is installed; just enough to
send state events when ops boots. Since ops has to boot before boss
(NFS mounts) the main script goes into the background and retries
until success.

Note that the ssl version of tmcc is installed, and so the proper pem
files must be installed on ops. Thats ctrlnode.pem from the ssl obj
directory as well as emulab.pem (public part of CA) into /etc/emulab.

SQL to do it.

Add an intentionally undocumented -b option to skip the interactive
confirmation, for use with scripts, etc. that have already asked for
it.

script.

"arbitrary" script as defined in the stated_triggers table. Currently
using this to invoke the new opsreboot script whenever ISUP comes in
from ops.

The opsreboot script is currently a skeleton. All it does is send
email.  I'll add the rest later (which really won't be much at first;
just getting the event schedulers started).

Add OPSNODEBSD stated machine for aforementioned OSID.
Add a new trigger for aforementioned OSID that causes sbin/opsreboot
to be invoked whenever ops reboots.

have have gotten.

shells on boss) to use the rpc server without an agent running.
Using the no-passphrase key, these changes allow us to use the server
from ops in a very restricted manner. This change is temporary, until
I have something better in place. In the meantime, admin people change
their auth keys files on *boss* as such:

command="/usr/testbed/sbin/sshxmlrpc_server.py -ro",from="ops.emulab.net" ... rest of emulab generated key ...

Note the -ro argument; very important!

shell doesn't try to interpret any of it.

In detail:

1. Added TBDB_NODESTATE_ALWAYSUP to libdb.pm for representing the ALWAYSUP
eventstate.

2. Modified free node calculation in ptopgen to include ALWAYSUP nodes.

3. Added code to ptopgen to correctly handle the case of a NULL iface
column, which happens when switches have interface (as they do in
Wisconsin), but assign_wrapper expects (null) for their iface rather than
"".

commit the code that checks for unanachored entries in the routines
that use the regexes.

of all references to a node. Mainly intended for when you have a
mishap with the newnode stuff and need to clean it up.

Added a big list of which tables contain information about physical
nodes to libdb, so that this and other scripts can find it all.

borken the code that helps assign figure out which desire was not
matched.

Also remove some old dead code while I'm in here.

Removed ipassign routing module. All routing must new be done through distributed-dijkstra or through the old routing module. RIP.

resolve it using a library function that will go through multiple
levels of indirction, so that we can point FBSD-JAIL at FBSD-STD.

easier to nfree them if they are all on one line. Listed in order by
date they entered hwdown.