GitLab

containers.

This is a hack.  The "blockstore" procedure is just returning a node
object, which causes skew in the resulting topology vs. what is generated
by the implicitly created node objects in the real testbed parser.  Harmless,
but not a good validity test.  Need to revisit.

for mere users.

This makes it convenient to request public addresses for virtual nodes
through the NS interface, like this:

    tb-set-node-routable-ip $node 1

Previously, if the drift file didn't exist at boot time, drift was not tracked.
Now we check at each report time instead, since ntpd can create the file and
that might not happen for a hour or more after boot.

Replace an exec of the perl os_select script with a call to the OSSelect()
node method. This cut in half the time spent in the DB setup for each node.
Note that this change had already been made to libosload_new.

Reworked the code that setup the partitions table entries. We were potentially
updating each DB row for each image loaded. Now we just work out all the changes
in a perl data struct and make one set of DB changes at the end. The code is
more comprehensible now as well (I hope!)

Finally, disable the "swapinfo" stuff which was the first step in doing stateful
swapout of disk state. That code never got finished.

Previously if you reverted an image to an older version, the check would
not detect it and we would not update cached copies.

Also, improve a couple of info messages.

Add node attribute setting memory size to 768MB.

That sort order is: reservations that need nodes, priority, creation date.
It used to be: priority, pid.

The new order makes the most sense for reservations that still need nodes,
so you can see easily where they are going to land first.

Also fix typo in BlockstoreType.pm

For every tarball and rpm, tmcd will now pass a SERVER= string as well
telling the client where the file should be downloaded from (if using
the web rather than NFS). Right now this value is the same for all
tarballs and rpms, and is hardwired in tmcd as either "www" (if
SPEWFROMOPS=0) or "users" (if 1). Note: BUMPED THE TMCC VERSION NUMBER
for this.

Made a common routine for doing an error-check-and-retry copy of a file
across "racy" NFS. This is used by install-{tarfile,rpm} and rc.topomap.

Clean up the web page a bit, and add a checkbox to retain
the image file in case the user wants to keep it around,
but delete the descriptor (perhaps to recreate it).

To turn this option on, define SPEWFROMOPS=1 in your defs file. This
will result in a redirect message from boss which will send the wget
client over to ops. 

A perl setuid root cgi script is run from the webserver on ops when a
/spewrpmtar request is made. This script sends the key,nodeid,file
over to boss via XMLRPC (as elabman). The return is simple yes or no,
the caller is allowed (not allowed) to have that file. Since the
ops script runs as root, it can spew the file back to the caller.

Note that the elabinelab checks for the elabinelab source code are
gone; we are now open source. Also, we spew that file from /share now,
to be consistent.