GitLab

viewing the log.

experiment code. No longer uses another table. Rather, the experiment
record contains a couple of extra fields for the batch system. Also
combined some of the backend code (no longer a killbatch script).
Also added scriptable experiments; the batchexp program in the bin
directory can start an experiment from the command line, and in fact
is used from the web page for both batch experiments and immediate
experiments (-i option). All of the DB code that was in the web
interfaces was moved to batchexp.

testbed boot process and the freebsd pxeboot/MFS kernel. I'm hoping
this will be useful enough for users to load their own disk images,
but we shall see. Usage is simple:

	/usr/testbed/bin/node_admin on|off pcXXX

The "on" state boots the MFS, and the "off" states sets it back to
normal.

using the libdb library. They have also been cleaned up somewhat, with more
comments added, etc.

I've merged nodeip and mac2if, which had very similar functions. The merged
version goes under the name nodeip.

Lastly, I've added some access checks to db2ns and node_status, to make sure
they're only called by the right people.

/proj/<proj>/tiplogs/<exp>/ . If set, virutal node name are used for the
destination filename.

Run by batch_daemon on experiment termination, and also available to users
(through paperbag/plasticwrap) any time during an experiment.

the list of nodes for a pid/eid, and fires off an update on each one.
This is for updating accounts and mounts on nodes after boot.
Currently invoked from the experiment sharing web page, but could be
more general purpose.

Manages the launching of new frisbee servers, and recording the
addresses the use in the database. If called for an image that is
already associated with a running server, exits quitely. Otherwise,
registers the new server's address and goes into the background, waiting
for the frisbee server to die so that it can unregister the address.

Not exactly what Mike wanted, but this was quick and easy. I'll make
it pretty later (sure you say!).

things, although this solution is not hugely better ...

	BOSSNODE=boss.emulab.net
	USERNODE=users.emulab.net
	OURDOMAIN=emulab.net

This uses the pxe booted freebsd kernel and MFS. In addition, I use
the standard testbed mechanism of specifying a startup command to
run, which will do the imagezip to NFS mounted /proj/<pid>/.... The
controlling script on paper sets up the database, reboots the node,
and then waits for the startstatus to change. Then it resets the DB
and reboots the node so that it returns back to its normal OS. The
format of operation is:

	create_image <node> <imageid> <filename>

Node must be under the user's control of course. The filename must
reside in the node's project (/proj/<pid>/whatever) since thats the
directory that is mounted by the testbed config software when the
machine boots. The imageid already exists in the DB, and is used to
determine what part of the disk to zip up (say, using the slice option
to the zipper). Since this operation is rather time consuming, it does
the usual trick of going to background and sending email status later.

and produces a ns file that will produce an identical topology to the one
stored in the DB.  It does not keep virtual ports the same but does map IP
addresses to connections correctly.  I.e. the topology is the same even
though the virt_nodes table differ slightly.

plastic. Useful for when plastic reboots, although it needs to be
run from paper of course. We should probably set up some kind of
reboot utility for when plastic reboots, since when capture fires up,
the "acl" files are all in the wrong group/mode. On the other hand,
this shared file based permission scheme is going to end up changing
anyway soon, since it won't work when the serial lines are on other
machines. Will need a tip permserver at that point.

customization for the log paths. Also added an increase of max
connections to 200.

Usage: dhcpd_makeconf [-h] [-v] <templatefile>
templatefile should be a skeleton dhcp.conf file, without entires for each node.
The string '%%nodetype=<type>' will be expanded to entries for all nodes of type
<type> in the following form:
host $ip {
	hardware ethernet $mac;
	option host-name '$node_id';
	fixed_address $ip;
}

The generated dhcp conf file is printed on stdout.

The -v option causes dhcpd_makeconf to use virtual names, rather than
physical ones, for hostnames.

Rob and I noticed that when we need to do maintainance to a node, its hard
to grab it when it gets freed before it gets taken again. So this will allow
admins to set up nodes so that when the current owner frees them, they
immediately are moved into another experiment. For instance, if I need to do
an upgrade on all the machines in the testbed, I can do:

sched_reserve testbed down pc1 pc2 pc3 [...] pc40

and it will immediately reserve any free nodes, skip any that are already
in testbed/down, and add an entry to the next_reserve table for any nodes
that are already reserved. Then nfree checks for entries in the next_reserve
table before it lets any nodes go (much like it does for the reloads table),
and moves them into the expt before it can get reserved again. So we can
guarantee that the next time a node is freed, it will be saved for us.

Note that this can also be useful to accumulate nodes for a large experiment,
since you can make sure you get them as soon as the current owner is done.

the new list files the users node, instead of using NFS.

updated.  This was breaking the testing stuff as some scripts were still
using the One True DB.

they are not yet installed from the top-level makefile, pending a decision on
the best way to install files outside the @prefix@ directory.

Added configure option '--with-rcdir', which defaults to /usr/local/etc/rc.d,
that specifies where to install these.

(respectively).  I.e.:

$TESTMODE = @TESTMODE@;
...
if ($TESTMODE) {
	...
}

sshtb is a _very_ simple shell script that runs ssh with a few commandline
parameters, which make it play nicer in an script environment. These
parameters can be changed with the '--with-ssh-args' argument, but default to:
'-q -o "BatchMode yes" -o "StrictHostKeyChecking no"'
All ssh calls now use this script.

Fixed delay config, and had to add it to configure since it needs to know the path to its libdb.pm and needs the database.

(the new one is logs email address), move all those silly -with
areguments into a definitions file. See defs-defaults and
defs-stoller-emulab for examples. When you configure a development
tree, you want to create your own defs file, and then configure like
this:
	/src/testbed/configure  \
		--prefix=/usr/testbed/devel/stoller \
		--with-TBDEFS=/src/testbed/defs-stoller-emulab \
		--with-LEDA=/usr/testbed/LEDA/LEDA-4.0

perl script in the bin dir, so its available as a command line program
too. The link is available from the Begin Experiment page.

machines in the experiment. exports_setup runs on paper and queries
the DB, generating a set of export lines. That is sent over to
plastic, and piped into exports_setup.proxy, which generates a new
/etc/exports file and restarts mountd. All of the client setup scripts
have been changed to mount the project and user directories.

Change to startexp/endexp, which are almost scriptable now (can be
called directly in addition from the web page). Add front ends to
these for the web page (webstartexp and webendexp). These changes are
mostly support for batch mode.

where script email goes. Defaults to testbed-ops@fast.cs.utah.edu. Change
it for your development tree with: --with-TBOPSEMAIL=stoller@fast.cs.utah.edu