GitLab

root=/dev/XXX param in /boot/grub/grub.conf, since grub.conf gets read at
boot by the bootloader.

Previously it was not being filled in and would wind up '0'.  Since it was
part of the primary key along with the service, this basically meant that
only one experiment at a time could successfully use the sync-server; i.e.,
only one could have (0, 'emulab-syncd') as their primary key.

describing the number of outlets possible on a power controller.

Also, attempt reconnect once if DB access fails:
>               /*
>                * Try once to reconnect.  In theory, the caller (client)
>                * will retry the tmcc call and we will reconnect and
>                * everything will be fine.  The problem is that the
>                * client may get a different tmcd process each time,
>                * and every one of those will fail once before
>                * reconnecting.  Hence, the client could wind up failing
>                * even if it retried.
>                */

recently.  Reduce plab_nodehist logging to get rid of implicit
redundancies.  Remove some debug code.

table... thus it no longer starts over every time it gets restarted.

last three days remain.  Saves cleaned rows to a logfile for future use.

times to give the vserver a chance to settle after creation via xmlrpc.
On v4 nodes, we can't seem to login immediately after sliver creation;
waiting a few seconds solves that problem.

Add needed method Template::gid_idx.

the exp directory, which happened when we started archiving the entire
experiment directory.

fix isn't that great, but it works for now.

nodes.

as 'plabstatus-STATUS'.  Also add support for multiple ticket versions. I
had avoided this because I thought rollout would happen fast and I didn't
want to whack the db just to do this... but rollout is taking a long time
and hitting PLC for legacy tickets is too slow to ignore anymore.

(e.g., /share/freebsd/6.2/src on /usr/src)

(which supports that option).  Otherwise some nodes will still timeout in
the elabinelab environment.

experimental LAN.

tries to connect later. We started seeing this with pubsub when the
server dies, so I expect our failure mode is different!

like thing.

Add fine-grained Datapository Access based based on "dp_projects" node_attributes.

"I implemented and tested extensions to snmpit & friends so that an
elabinelab could additional request that an experimental interface be
placed in trunked mode, to discover the vlan tags associated with
vlans, and to request modifications to existing vlans belonging to an
elabinelab without tearing it down and reconstructing it."

renaming the startup file in /usr/local/etc/rc.d (which of course
needs to be there, and will be, cause someone will turn pubsub into a
package that we can pkg_add).

illegal.

I also added elvin_gateway since I think we want to run that in our
elabinelab for testing and using older images that still have elvin
clients on them.