GitLab

(originally called for just the main address) so that all the routes
get installed properly.

New 'blog-like news system.
Should make it easier to keep up with news updates!
Since:
  1. Doesn't require editing news.html in the CVS tree and
     recopying it/reinstalling just to add a news item.
  2. Doesn't involve editing menu.php3,
     since the modified date and "new" icon is automatic now.
  3. Looks nicer.

Also, it is based off the db, so other sites can have other news,
which they can maintain independently.

There are of course, neat features that have been suggested that
could go with this, such as message updates and automatic
mailing list mailings of news, but those don't exist yet...

Added new feature 'Experiment Modify'.
Now available (to admins only for now) from the showexp page.

Warning! doing a modify which alters the topology will probably
require a "reboot all nodes" afterwards.
(There will be a checkbox soon in the modify experiment page.)

Adding/removing delay nodes seems to work fine without reboots, though.

Warning! If the new version of the experiment cannot be mapped
 (not enough nodes available, for instance) the experiment will be
 swapped out! This will get fixed later.

Prerun backs up the experiment topology, so using a bad NS
file doesn't result in experiment termination.

As part of this, added library functions to libdb to
delete, backup, and restore both virtual and physical experiment state.

forthcoming, along with addition of idletimes, idlemail, and webidlemail.

options in the future, they'll get this one too.

Obviously, this won't scale to too many sites, so we'll have to
figure something else out.

clients so they can create a proper .forward file. 2) Add some
test code for seeing how feasible it is to do accounts on ops/tips
using tmcd instead of direct ssh. Initial results are not good; too
many accounts; does not scale well. Going to need a smarter protocol
for doing hundreds of accounts, but I will leave this code in place
for now. Note, unlike testbed nodes, control nodes must connect with
ssl.

Site-specific header graphic for gatech.

the hosts files that we put on free nodes. For reserved nodes, we
start a new hosts file from scrath, and qualify localhost with
the domain found from the hostname command.

and havoc upon all mankind and testbed users.

Before:
The main defs file (ie for configure) had an entry for WWWDEFS that
pointed to a <@WWWDEFS@>-defs.php3 file in the www/ directory. The www
defs file loaded some values about web pages, URLs, and some web
configuration parameters.

Problem:
Anything that was only in the www defs file was not accessible in the rest
of the universe (ie perl, C, and any other non-web-page scripts). For
instance, you couldn't have a perl script send an email to a user with a
link to the web site.

Solution:
Nuke all the www-defs files, move any important values into the main
configure, and change the web defs infrastructure to respect that. This
also meant adding about 3 lines each to all of the configure defs files.
(There really are about 10 new values you can change in your defs file,
but in almost all cases, the default values are the right thing.)

Upgrading:
External sites will need to move a few variables from their www-defs file
into their configure defs file. The example file should make it pretty
obvious. They may also want to customize some of the other vars that are
mentioned in configure.in and www/defs.php3.in .

work for virtual links when the node was a router, so pull them out
until I have the ipfw changes installed (the changes to use nexthop).

and when the last one was sent.

Subject: EMULAB.NET: 5 PCs idle 3 hours: gtecho/bar

think it is idle because it has been sent a swap request, but it isn't
really inactive, reset the swap request counter.

NOT have any options set.

clients so they can create a proper .forward file. 2) Add some
test code for seeing how feasible it is to do accounts on ops/tips
using tmcd instead of direct ssh. Initial results are not good; too
many accounts; does not scale well. Going to need a smarter protocol
for doing hundreds of accounts, but I will leave this code in place
for now. Note, unlike testbed nodes, control nodes must connect with
ssl.

shaping with them. Also a discussion of jails and how they relate, and a
proposed solution using MAC addresses (potentially made up ones) to
perform the demultiplexing at the receiving node.

bogus email to tbops, but rather spits the error directly to the
user.

swapfile.sys does not exist don't abort with an error message anymore.

we can send links that cause those fields to be filled in for people.
Add those links to email generated by newproject and newgroup pages
so that they can be saved by the leaders.

Add a bit of referrer magic to login page. When clicked from the join
or new project pages, tell login page to pass along the referrer page
so that when login is complete, user is zapped back to the original
page. This is especially nice when combined with the above change to
joinproject, whereby we send along the target pid/gid, but the user
has not yet logged in and remembers to do so via the link at the top
of the page.

make one in /usr/testbed/expwork/$pid. Too much cruft getting left
behind and it was causing even more log copy errors! Besides,
typically its just tbops people who need to look at that stuff.

I've fixed all existing group directories.