1. 07 Jan, 2009 1 commit
    • Leigh B. Stoller's avatar
      Various changes to support storing control network vlans in · f55e466c
      Leigh B. Stoller authored
      the DB, and eventually elabinelab firewalls.
             # Allow arbitrary the names for the ids. Forgot who wanted this.
             alter table elabinelab_vlans change `inner_id`
                      `inner_id` varchar(32) NOT NULL default '';
             alter table elabinelab_vlans change `outer_id`
                      `outer_id` varchar(32) NOT NULL default '';
             # Tag elabinelab created vlans with the stack module.
             alter table elabinelab_vlans add `stack`
                 enum('Control','Experimental') NOT NULL default 'Experimental';
             # Tag vlans softstate table with the stack module.
             alter table vlans add `stack`
                 enum('Control','Experimental') NOT NULL default 'Experimental';
             # A new table, to store softstate on interfaces. snmpit will
             # modify this table whenever a port is enabled/trunked(tagged)
             # on the command line. Eventually speed/duplex should go here.
             CREATE TABLE `interface_state` (
      	  `node_id` varchar(32) NOT NULL default '',
      	  `card` tinyint(3) unsigned NOT NULL default '0',
      	  `port` tinyint(3) unsigned NOT NULL default '0',
      	  `iface` varchar(32) NOT NULL,
      	  `enabled` tinyint(1) default '1',
      	  `tagged` tinyint(1) default '0',
      	  PRIMARY KEY  (`node_id`,`card`,`port`),
      	  KEY `nodeiface` (`node_id`,`iface`)
      	# Initialize the above table.
      	insert into interface_state (node_id,card,port,iface)
      	   select node_id,card,port,iface from interfaces;
      	# Cleanup the vlans table which was filling with cruft
      	# cause of a bug in the swapout code. Finally figured
      	# out how to combine delete and join!
      	delete v FROM vlans v LEFT JOIN lans l on l.lanid=v.id
      	       WHERE l.lanid IS NULL;
              # Fix an unrelated bug; slot overflow.
              alter table nodelog change `log_id`
                      `log_id` int(10) unsigned NOT NULL auto_increment;
  2. 11 Dec, 2008 1 commit
  3. 08 Dec, 2008 1 commit
  4. 04 Dec, 2008 1 commit
  5. 17 Nov, 2008 1 commit
    • Leigh B. Stoller's avatar
      Minor changes to user certs to support certificate revocation lists · 4809cd65
      Leigh B. Stoller authored
      in the protogeni code. We no longer save the unencrypted certs after
      they are revoked, since protogeni will ignore them. I redid the the
      DB table as well, adding a revoked stamp, and the DN so that we can
      generate the CRL list from the DB directly, without having to run them
      all through openssl.
      This commit requires all certs to be regenerated, and the ssl xmlrpc
      server to be restarted.
  6. 20 Oct, 2008 1 commit
  7. 17 Oct, 2008 1 commit
  8. 28 Aug, 2008 1 commit
  9. 19 Aug, 2008 1 commit
  10. 16 Jul, 2008 1 commit
  11. 10 Jul, 2008 2 commits
  12. 11 Jun, 2008 2 commits
  13. 03 Jun, 2008 1 commit
  14. 30 May, 2008 1 commit
  15. 20 May, 2008 1 commit
  16. 19 May, 2008 1 commit
  17. 13 May, 2008 1 commit
  18. 12 May, 2008 1 commit
  19. 17 Apr, 2008 3 commits
  20. 24 Mar, 2008 1 commit
  21. 21 Feb, 2008 1 commit
  22. 11 Feb, 2008 2 commits
  23. 09 Feb, 2008 1 commit
  24. 14 Jan, 2008 1 commit
  25. 10 Jan, 2008 2 commits
  26. 20 Dec, 2007 1 commit
  27. 26 Nov, 2007 2 commits
  28. 06 Nov, 2007 1 commit
    • Leigh B. Stoller's avatar
      This started out as a simple change to turn the datastore into a CVS · c1cff09b
      Leigh B. Stoller authored
      sandbox, and that I did. It falls back to the older archive when
      the template is older then CVS repos.
      But along the way I got annoyed with the fact that template instantiation
      does not provide a logfile to the web interface. The reason is that
      the current logfile stuff is very experiment centric; there has to be an
      experiment and an attached logfile. An instance does not have an experiment
      until really late in the game so the code was just not bothering.
      Anyway, I've started to generalize the logfile stuff with a new table
      and the approach that a logfile is named by a random key, and if you
      know the key you can look at the logfile in the web (since without an
      experiment it is hard to do permission checks unless we make logfiles
      uid/gid owned, and I did not want to do that.
  29. 17 Oct, 2007 1 commit
  30. 21 Aug, 2007 1 commit
    • Leigh B. Stoller's avatar
      Another round of widearea node hacking for CMU. These changes add · 99346dc0
      Leigh B. Stoller authored
      widearea reloading support.
      * New slot in the images table to store an access key which remote
        sites must provide in order to download an image (via https).
      * tmcd returns a different kind of ADDRESS field from doloadinfo.
        Instead of the multicast stuff, return a URL that points to boss'
        web server. The URL is of the form:
        which as you can see is fully specified; the client does not need
        to know anything else.
      * New webpage and backend scripts appropriately called "spewimage"
        which also includes support for the http HEAD request (from wget) to
        avoid downloading images that are already on the node. I just
        learned about this HEAD request stuff today ... but otherwise these
        operate as expected, spewing the image if the access key is provided.
      * Changes to rc.frisbee to deal with remote loading. In addition to
        URL support, I also added support for simple paths, the intent being
        that we will probably distribute images offline (say, at night) so
        that when a node reboots it doesn't actually have to wait 60 minutes
        for an image to download. I have not added any server side support
        for this yet though. Maybe later this week.
      * Other bits and pieces and fixes to make this work.
  31. 16 Aug, 2007 1 commit
    • Leigh B. Stoller's avatar
      Add IP and privkey slots (key INDEX slots) to widearea_nodeinfo table. · 0fe9fe1a
      Leigh B. Stoller authored
      This will superceed the widearea_privkeys table that was used for
      RON/Netbed but was never done well.
      The privkey is just something we generate and store on the dongle (as
      a simplification, we just create a dongle per node instead of
      generating them on the fly) since there are not very many of these
      widearea (CMU) nodes. We store the IP here for convenience (there is
      an interfaces table entry of course).
      The IP is updated when the node tells us it has changed, using the
      privkey to uniquely identify it (for checkin).
  32. 31 Jul, 2007 1 commit
  33. 27 Jul, 2007 1 commit