Fix autoinstalls to work around a modern d-i bug

[invirt/packages/invirt-xen-config.git] / debian / changelog

invirt-xen-config (0.0.60) unstable; urgency=low

  * Fix autoinstalls to work around a modern d-i bug

    Modern d-i (at least Debian Buster, and probably others) reads
    /proc/consoles and adds entries to /etc/inittab to spawn a d-i
    on each of them.  With debconf/priority>=high (i.e. without
    the menu showing, and with d-i autodriving through the items),
    if there are multiple consoles, this results in the d-i's racing
    to use /var/lib/dpkg/status to record what happens as they
    run the postinst of each udeb, and as a result, menu items
    fail and the autoinstall fails.  See Debian Bug #944125.

    We only set console=hvc0, but for reasons unknown, tty0 also
    appears in /proc/consoles despite not being in /proc/cmdline,
    and we don't seem to have control over that.  The script that
    reads /proc/consoles is /sbin/reopen-console, which also
    happens to be the main rc script on sysinit in /etc/inittab.
    That script (through a chain of other scripts) is also what
    runs preseed/early_command if it's set.

    Normally, preseed/early_command cannot run until way too late
    to address this, because you don't have the preseed until several
    tasks into the install by which point it will have failed.
    However, if you get the quoting right, you can set it as a
    boot argument and it will run before any of the menu items,
    and in fact, before d-i has started.

    Because /sbin/reopen-console ultimately is what runs
    preseed/early_command, when the command runs, it is too late
    to use sed to edit reopen-console, and you can't kill it
    because it runs the rest of the process.  At the point where
    preseed/early_command runs, reopen-console has already found
    the consoles and added d-i jobs for each of them to /etc/inittab.
    However, it doesn't send the HUP to init until just after the
    scripts including preseed/early_command are run.

    Yank the tty0 job out from inittab just in the nick of time.
    This should be safe for old distros because we have always
    forced the desired console to hvc0; while the problem that
    necessitates this kludge didn't exist in the past, removing
    any inittab job for tty0 shouldn't have any noticeable effect
    on them.

 -- Mitchell Berger <mitchb@mit.edu>  Wed, 01 Jan 2020 03:43:00 -0500

invirt-xen-config (0.0.59) unstable; urgency=low

  * Fix the fact that we can't bump RAM quotas above 2048M
    (and acknowledge the reality that 2048M really isn't the rarest
    thing to need these days)

  Quoting from history:

  """
     invirt-xen-config (0.0.15) unstable; urgency=low

       * Set the max memory to a really high number so we can bump it for users
       * Pass the correct networking device to vif-invirtroute

      -- Evan Broder <broder@mit.edu>  Fri, 14 Nov 2008 00:03:47 -0500
  """

  i,i "640K ought to be enough for anyone."

 -- Mitchell Berger <mitchb@mit.edu>  Tue, 23 Jul 2019 18:50:00 -0400

invirt-xen-config (0.0.58) unstable; urgency=low

  * Support kickstart-based Fedora autoinstalls.

 -- Quentin Smith <quentin@mit.edu>  Sat, 15 Jun 2019 20:59:54 -0400

invirt-xen-config (0.0.57) unstable; urgency=low

  * invirt-database: Fix the autoinstall logic so that if we've
    increased the memory temporarily for the autoinstall above
    the machine's database setting, we also bring maxmem up to
    the same level.  Otherwise the machine doesn't turn on
    because it's requesting to start with more memory than maxmem,
    and this failure mode is silent to the user.

 -- Mitchell Berger <mitchb@mit.edu>  Sat, 02 Feb 2019 16:52:00 -0500

invirt-xen-config (0.0.56) unstable; urgency=low

  * vif-invirtroute: Kill arpspoof with SIGKILL when we're done.  The
    version in this distro (which is the latest beta from the
    unmaintained upstream) tries to "clean up" when you normally
    terminate it.  This involves trying to perform an ARP lookup
    without interfering to see what the MAC should've been and then
    sending replies for that.  This appears to cause a race depending
    on whether the entry for the machine's old domain (if it's being
    rebooted) is listed as <incomplete> in the ARP table or is totally
    flushed.  If you lose the race, it sends further gratuitous ARPs
    telling the router that your MAC is all-zeros because of a bug
    where it fails to check for an incomplete ARP entry.  This is
    believed to cause the bug that has plagued XVM for years where
    you sometimes lose off-network connectivity on reboot.

 -- Mitchell Berger <mitchb@mit.edu>  Mon, 25 Jun 2018 22:04:00 -0400

invirt-xen-config (0.0.55) unstable; urgency=low

  * invirt-database: bump up autoinstall memory to make modern releases
    work again.

 -- Mitchell Berger <mitchb@mit.edu>  Fri, 01 Jun 2018 02:17:00 -0400

invirt-xen-config (0.0.54) unstable; urgency=low

  * invirt-database: handle several possible actions to take with an
    "other" set of network parameters.  Renumber a machine by swapping
    the sets of netparams, DNAT from an old address by passing through
    additional parameters to vif-invirtroute, and teardown an IP
    transition by clearing all the "other" values.  Also, use the
    netmask and gateway for the VM when autoinstalling, rather than
    parameters from the global cluster configuration.
  * vif-invirtroute: if there is an "other" IP address, DNAT it to the
    primary IP, set up routing to get it to the VM, and arp for it.
    Also, cut out the nonsense with backgrounding arpspoof, sleeping, and
    killing it; we have the technology for 'timeout'.  Finally, don't
    pass a "src" argument to "ip route del" to ensure a match even if
    the host's "main IP" has changed and caused the route to be inserted
    with a different src.
  * templatize-xend-config: explicitly turn on the relocation server
  * qemu-ifup.invirt: make HVMs work and not only if your eth0 just happens
    to have an IPv4 network address sitting on it

 -- Mitchell Berger <mitchb@mit.edu>  Sun, 27 May 2018 01:45:00 -0400

invirt-xen-config (0.0.53) unstable; urgency=low

  * Fix machine UUID assignment after SQLAlchemy started returning Unicode
    objects.

 -- Quentin Smith <quentin@mit.edu>  Mon, 06 Jan 2014 22:02:11 -0500

invirt-xen-config (0.0.52) UNRELEASED; urgency=low

  * Fix PVHVM networking

 -- Adam Glasgall <glasgall@mit.edu>  Mon, 02 Sep 2013 21:16:01 -0400

invirt-xen-config (0.0.51) unstable; urgency=low

  * Fixed code for NIC model options.

 -- Steven Valdez <dvorak42@mit.edu>  Wed, 31 Jul 2013 22:28:39 -0400

invirt-xen-config (0.0.50) unstable; urgency=low

  * Adding NIC model option to DB.

 -- Steven Valdez <dvorak42@mit.edu>  Fri, 26 Jul 2013 22:28:47 -0400

invirt-xen-config (0.0.49) precise; urgency=low

  * Change HVM emulated network card to pcnet, because the emulated
    realtek hardware is as bad as the real realtek hardware.

 -- Adam Glasgall <glasgall@mit.edu>  Tue, 23 Jul 2013 02:52:06 -0400

invirt-xen-config (0.0.48) precise; urgency=low

  * Disable PVHVM for now until we figure out the best way to do it

 -- Adam Glasgall <glasgall@mit.edu>  Mon, 22 Jul 2013 01:08:40 -0400

invirt-xen-config (0.0.47) unstable; urgency=low

  * Blame glasgall for the wasted version number.

 -- Quentin Smith <quentin@mit.edu>  Sun, 21 Jul 2013 14:55:00 -0400

invirt-xen-config (0.0.46) unstable; urgency=low

  * Non-hotplug events apparently also use vif-setup.

 -- Quentin Smith <quentin@mit.edu>  Sun, 21 Jul 2013 14:49:54 -0400

invirt-xen-config (0.0.45) unstable; urgency=low

  * Neuter vif-setup (from xl, perhaps?)
  * Update vif-invirtroute for Xen 4's scripts.

 -- Quentin Smith <quentin@mit.edu>  Sun, 21 Jul 2013 14:28:58 -0400

invirt-xen-config (0.0.44) unstable; urgency=low

  * Disable populate-on-demand for HVMs.

 -- Quentin Smith <quentin@mit.edu>  Sun, 21 Jul 2013 13:14:53 -0400

invirt-xen-config (0.0.43+nmu1) precise; urgency=low

  * Use config-package-dev.

 -- Alexander Chernyakhovsky <achernya@mit.edu>  Sun, 21 Jul 2013 12:48:47 -0400

invirt-xen-config (0.0.43) precise; urgency=low

  * fix the diversion too.

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 21 Jul 2013 01:14:41 -0400

invirt-xen-config (0.0.42) precise; urgency=low

  * qemu-ifup moved from /etc/xen to /etc/xen/scripts

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 21 Jul 2013 01:06:27 -0400

invirt-xen-config (0.0.41) precise; urgency=low

  * no-change rebuild

 -- Adam Glasgall <glasgall@mit.edu>  Sat, 20 Jul 2013 16:54:56 -0400

invirt-xen-config (0.0.40) precise; urgency=low

  * Updating version for precise.

 -- Steven Valdez <dvorak42@mit.edu>  Sat, 20 Jul 2013 16:46:47 -0400

invirt-xen-config (0.0.39) unstable; urgency=low

  * Treat config.hosts.internal_ips as an array

 -- Peter Iannucci <iannucci@mit.edu>  Mon, 01 Jul 2013 21:51:00 -0400

invirt-xen-config (0.0.38~glasgall15) precise; urgency=low

  * Teach qemu-dm-invirt that /usr/lib/xen moved to /usr/lib/xen-4.1

 -- Adam Glasgall <glasgall@mit.edu>  Thu, 06 Jun 2013 21:36:37 -0400

invirt-xen-config (0.0.38~glasgall14achernya1) precise; urgency=low

  * Switch to dh7
  * Bump standards version

 -- Alexander Chernyakhovsky <achernya@mit.edu>  Fri, 31 May 2013 21:16:20 -0400

invirt-xen-config (0.0.38~glasgall14) precise; urgency=low

  * /usr/lib/xen moved to /usr/lib/xen-4.1. teach invirt-database this.

 -- Adam Glasgall <glasgall@mit.edu>  Wed, 08 May 2013 00:06:25 -0400

invirt-xen-config (0.0.38~glasgall13) precise; urgency=low

  * Undo enabling network-bridge; we want to manage our own bridge config
    now. Oh, irony.

 -- Adam Glasgall <glasgall@mit.edu>  Mon, 06 May 2013 19:14:41 -0400

invirt-xen-config (0.0.38~glasgall12) precise; urgency=low

  * xm was expecting a string and was getting a unistring for machine
    name, which resulted in the VM name always being
    'invirt-database'. Use passed-in machine_name instead, since that will
    be a string. IBTSOCS.

 -- Adam Glasgall <glasgall@mit.edu>  Mon, 06 May 2013 19:14:23 -0400

invirt-xen-config (0.0.38~glasgall11) precise; urgency=low

  * Add dependency on ethtool.

 -- Adam Glasgall <glasgall@mit.edu>  Fri, 26 Apr 2013 22:36:10 -0400

invirt-xen-config (0.0.38~glasgall10) precise; urgency=low

  * SQLAlchemy API updates in invirt-database xen config

 -- Adam Glasgall <glasgall@mit.edu>  Fri, 26 Apr 2013 21:14:25 -0400

invirt-xen-config (0.0.38~glasgall9) precise; urgency=low

  * _really_  correct regexp this time.

 -- Adam Glasgall <glasgall@mit.edu>  Sat, 13 Apr 2013 17:57:11 -0400

invirt-xen-config (0.0.38~glasgall8) precise; urgency=low

  * fix typo in regexp

 -- Adam Glasgall <glasgall@mit.edu>  Sat, 13 Apr 2013 17:49:08 -0400

invirt-xen-config (0.0.38~glasgall7) precise; urgency=low

  * Correct regexp in s/// over xend-config.sxp

 -- Adam Glasgall <glasgall@mit.edu>  Sat, 13 Apr 2013 17:46:43 -0400

invirt-xen-config (0.0.38~glasgall6) precise; urgency=low

  * Enable network-bridge in xend-config.sxp, since Precise doesn't by 
    default anymore.

 -- Adam Glasgall <glasgall@mit.edu>  Sat, 13 Apr 2013 17:40:33 -0400

invirt-xen-config (0.0.38~glasgall5) precise; urgency=low

  * s/// out the "${XEN_CONFIG_DIR}" reference that was making mako choke

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 24 Mar 2013 20:31:31 -0400

invirt-xen-config (0.0.38~glasgall4) precise; urgency=low

  * Oops, no, we don't actually want to uninstall grub. Depend on grub-pc
    instead.

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 24 Mar 2013 17:49:24 -0400

invirt-xen-config (0.0.38~glasgall3) precise; urgency=low

  * ubuntu-xen-server bit the dust several releases ago. replace it by
    explicitly listing the xen bits we need in Depends.

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 24 Mar 2013 17:44:24 -0400

invirt-xen-config (0.0.38~glasgall2) precise; urgency=low

  * Bump xen-utils dependency
  * Update xend-config.sxp transformer for newer default config

 -- Adam Glasgall <glasgall@mit.edu>  Sun, 24 Mar 2013 17:37:05 -0400

invirt-xen-config (0.0.38~glasgall1) precise; urgency=low

  * Bump version for precise

 -- Adam Glasgall <glasgall@mit.edu>  Thu, 21 Mar 2013 23:24:57 -0700

invirt-xen-config (0.0.37) unstable; urgency=low

  * Support multiple NICs on HVMs.

 -- Mitchell Berger <mitchb@mit.edu>  Sat, 29 Oct 2011 21:53:00 -0400

invirt-xen-config (0.0.36) unstable; urgency=low

  * Version bump to recover from invirtibuilder version error. No changes.

 -- Joshua Oreman <oremanj@mit.edu>  Tue, 26 Apr 2011 16:30:22 -0400

invirt-xen-config (0.0.33) unstable; urgency=low

  * Explicitly disable proxy_arp on the "all" interface; it seems to be
    enabled on bootup if we don't.

 -- Joshua Oreman <oremanj@mit.edu>  Tue, 26 Apr 2011 16:12:52 -0400

invirt-xen-config (0.0.32) unstable; urgency=low

  * Don't enable proxy arp on every interface, just the public one (fixes
    problems with reachability on the backend network!)

 -- Quentin Smith <quentin@mit.edu>  Wed, 09 Mar 2011 18:18:06 -0500

invirt-xen-config (0.0.31) unstable; urgency=low

  * Only run arpspoof when NICs are brought online (this potentially fixes
    a race condition that occurs when migrating VMs)
  * Remove the advertised gso-tcpv4 feature from VIFs which causes the
    guest to incorrectly believe that Xen supports TCP segmentation
    offload. This supercedes our previous fix of disabling TSO in guest
    kernels.

 -- Quentin Smith <quentin@mit.edu>  Mon, 27 Dec 2010 01:19:58 -0500

invirt-xen-config (0.0.30) unstable; urgency=low

  * Support debian-installer based autoinstallation.

 -- Quentin Smith <quentin@mit.edu>  Sun, 26 Dec 2010 17:10:54 -0500

invirt-xen-config (0.0.29) unstable; urgency=low

  * Merged in r2821,2824,2827 from hvirt branch.

 -- Greg Brockman <gdb@mit.edu>  Sun, 31 Jan 2010 18:44:53 -0500

invirt-xen-config (0.0.28) unstable; urgency=low

  * Restarting xend is just a *really bad* idae.

 -- Evan Broder <broder@mit.edu>  Tue, 26 Jan 2010 23:32:47 -0500

invirt-xen-config (0.0.27) unstable; urgency=low

  [ Greg Brockman ]
  * Nuked the pinning of libxen3
  * Changed xen.network -> xen

  [ Evan Broder ]
  * Don't try to start sysvms in the invirt-xen-config init script. The
    code has never been reliable.
  * Don't create iptables rules in vif-invirtroute - they don't do anything.
  * For simplicity, and because we've never used it, tear the IPv6
    "support" out of vif-invirtroute.

  [ Quentin Smith ]
  * Increase dom0's weight to improve IO performance.

 -- Greg Brockman <gdb@mit.edu>  Sun, 17 Jan 2010 23:32:03 -0500

invirt-xen-config (0.0.26) unstable; urgency=low

  * Kill the moocow password for the VNC server. Not setting this variable
    causes VNC clients to not prompt for a password instead.

 -- Evan Broder <broder@mit.edu>  Mon, 28 Sep 2009 21:20:21 -0400

invirt-xen-config (0.0.25) unstable; urgency=low

  * Some of the autoinstallers want a lot of RAM. Give it to them.

 -- Evan Broder <broder@mit.edu>  Sun, 26 Jul 2009 20:35:13 -0700

invirt-xen-config (0.0.24) unstable; urgency=low

  * Prepare for a Xen 3.3 world by allowing either 3.2 or 3.3 packages.

 -- Evan Broder <broder@mit.edu>  Fri, 20 Feb 2009 17:36:23 -0500

invirt-xen-config (0.0.23) unstable; urgency=low

  * revert 0.0.22
  * use run-parts --list to get a tighter set of sysvm config files

 -- Greg Price <price@mit.edu>  Tue, 03 Feb 2009 00:20:03 -0500

invirt-xen-config (0.0.22) unstable; urgency=low

  * Correctly calculate both the list of sysvms to boot and the list of
    Mako templates to render in the init script.

 -- Evan Broder <broder@mit.edu>  Mon, 02 Feb 2009 17:30:51 -0500

invirt-xen-config (0.0.21) unstable; urgency=low

  * Move sysvms/* into xvm-prodconfig from invirt-xen-config
  * genericize sysvm startup to glob /etc/xen/sysvms/s_*
  * eth2 -> config.xen.network.iface

 -- Greg Price <price@mit.edu>  Sat, 31 Jan 2009 18:40:20 -0500

invirt-xen-config (0.0.20) unstable; urgency=low

  * Add Provides and Conflicts lines to debian/control for
    config-package-dev

 -- Evan Broder <broder@mit.edu>  Sat, 24 Jan 2009 20:16:06 -0500

invirt-xen-config (0.0.19) unstable; urgency=low

  * prune tempfile, subprocess imports from invirt-database
    (subprocess!  wow, that was a hack.  that was also over a year ago, r223.)

 -- Greg Price <price@mit.edu>  Tue, 09 Dec 2008 00:25:56 -0500

invirt-xen-config (0.0.18) unstable; urgency=low

  * Explicitly grab the short form of the hostname to assemble the backend
    hostname

 -- Evan Broder <broder@mit.edu>  Wed, 26 Nov 2008 01:27:07 -0500

invirt-xen-config (0.0.17) unstable; urgency=low

  * Pin libxen3 to the hardy repository so it doesn't get pulled from
    hardy-backports until we're ready to upgrade to Xen 3.3

 -- Evan Broder <broder@mit.edu>  Sat, 22 Nov 2008 19:19:07 -0500

invirt-xen-config (0.0.16) unstable; urgency=low

  * Pick the kernel and initrd for the autoinstaller based on the kernel
    the dom0 is currently running

 -- Evan Broder <broder@mit.edu>  Sat, 15 Nov 2008 23:45:30 -0500

invirt-xen-config (0.0.15) unstable; urgency=low

  * Set the max memory to a really high number so we can bump it for users
  * Pass the correct networking device to vif-invirtroute

 -- Evan Broder <broder@mit.edu>  Fri, 14 Nov 2008 00:03:47 -0500

invirt-xen-config (0.0.14) unstable; urgency=low

  * Throw away any output from LVM commands while starting sysvms
  * Correctly delete the lock LVs

 -- Evan Broder <broder@mit.edu>  Mon, 10 Nov 2008 03:57:09 -0500

invirt-xen-config (0.0.13) unstable; urgency=low

  * Use a lock LV when attempting to start sysvms since exclusive LV
    activation has useless semantics

 -- Evan Broder <broder@mit.edu>  Mon, 10 Nov 2008 03:15:31 -0500

invirt-xen-config (0.0.12) unstable; urgency=low

  * Add missing close paren to init script

 -- Evan Broder <broder@mit.edu>  Fri, 31 Oct 2008 09:50:47 -0400

invirt-xen-config (0.0.11) unstable; urgency=low

  * Automatically startup sysvms if they're not running already

 -- Evan Broder <broder@mit.edu>  Fri, 31 Oct 2008 09:47:15 -0400

invirt-xen-config (0.0.10) unstable; urgency=low

  * Generate config files for sysvms

 -- Evan Broder <broder@mit.edu>  Fri, 31 Oct 2008 07:40:13 -0400

invirt-xen-config (0.0.9) unstable; urgency=low

  * don't try to disable tx checksums for HVMs; it errors out

 -- Greg Price <price@mit.edu>  Wed, 29 Oct 2008 01:26:07 -0400

invirt-xen-config (0.0.8) unstable; urgency=low

  * sipb-xen-base -> invirt-base

 -- Evan Broder <broder@mit.edu>  Tue, 28 Oct 2008 04:23:38 -0400

invirt-xen-config (0.0.7) unstable; urgency=low

  * Clarify description to distinguish from invirt-host-master

 -- Greg Price <price@mit.edu>  Sat, 25 Oct 2008 21:16:51 -0400

invirt-xen-config (0.0.6) unstable; urgency=low

  * sipb-xen-database-common -> invirt-database

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 21:03:33 -0400

invirt-xen-config (0.0.5) unstable; urgency=low

  * Give the init script a less weird sounding description

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 19:11:28 -0400

invirt-xen-config (0.0.4) unstable; urgency=low

  * Another typo in the init script

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 19:07:32 -0400

invirt-xen-config (0.0.3) unstable; urgency=low

  * Some small fixes in the init script

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 19:03:12 -0400

invirt-xen-config (0.0.2) unstable; urgency=low

  * Whoops - add the dependencies for the old sipb-xen-database-client

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 18:38:56 -0400

invirt-xen-config (0.0.1) unstable; urgency=low

  * Rename sipb-xen-dom0 -> invirt-xen-config
  * Rearrange the source package
  * Divert /etc/xen/qemu-ifup, since it existed and we've been scribbling
    over it
  * Generate /etc/xen/xend-config.sxp
  * Move the invirt-database Xen config script into this package

 -- Evan Broder <broder@mit.edu>  Sat, 25 Oct 2008 17:52:39 -0400

sipb-xen-dom0 (2.10) unstable; urgency=low

  * Fix DHCP checksum error for ParaVMs

 -- Evan Broder <broder@mit.edu>  Sun, 19 Oct 2008 22:12:09 -0400

sipb-xen-dom0 (2.9.1) unstable; urgency=low

  * Whoops - was slightly too vicious with the cuts to qemu-dm-sipb

 -- Evan Broder <broder@mit.edu>  Sun, 19 Oct 2008 01:55:41 -0400

sipb-xen-dom0 (2.9) unstable; urgency=low

  * Move the qemu-ifup script to the location where it's now expected to
    be.
  * Clear out some debugging code from qemu-dm-sipb

 -- Evan Broder <broder@mit.edu>  Sun, 19 Oct 2008 01:39:43 -0400

sipb-xen-dom0 (2.8) unstable; urgency=low

  * Apply Sam's fix to vif-sipbroute for the prod cluster (the interface
    should eventually be configurized)

 -- Evan Broder <broder@mit.edu>  Tue, 07 Oct 2008 01:18:37 -0400

sipb-xen-dom0 (2.7) unstable; urgency=low

  * Update qemu-dm-sipb to reflect the new location of qemu-dm

 -- Evan Broder <broder@mit.edu>  Mon, 06 Oct 2008 18:55:59 -0400

sipb-xen-dom0 (2.6) unstable; urgency=low

  * depend on sipb-xen-base

 -- Greg Price <price@mit.edu>  Wed, 01 Oct 2008 20:28:21 -0400

sipb-xen-dom0 (2.5) unstable; urgency=low

  * Don't rely on exported ROOT.

 -- Anders Kaseorg <andersk@mit.edu>  Fri, 02 May 2008 02:41:43 -0400

sipb-xen-dom0 (2.4) unstable; urgency=low

  * Add sipb-xen-losetup, a half-sane wrapper around the insane losetup.

 -- Greg Price <price@mit.edu>  Thu,  1 May 2008 01:13:28 -0400

sipb-xen-dom0 (2.3) unstable; urgency=low

  * Depend on generic metapackages/provides instead of particular 
    versions.

 -- Anders Kaseorg <andersk@sipb-xen.mit.edu>  Sat, 27 Oct 2007 21:23:00 -0400

sipb-xen-dom0 (2.2) unstable; urgency=low

  * Add qemu-ifup script that depends on the device model script and
    invokes vif-sipbroute.    As a consequence, we don't really support
    bridged hvms any more.  It's not clear how to get enough information
    out of qemu to do this.
  * Call arpspoof with 18.181.0.1's address.  This is unfortunate in two
    ways.  First, if we renumber we'll need to update the script.  Second,
    it is possible that someone on 18.181 besides the router may have an
    address cached.  Unfortunately, it doesn't work right if we don't
    send the spoofed arp directly to the router.

 -- Sam Hartman <hartmans@debian.org>  Sun, 26 Aug 2007 18:57:17 -0400

sipb-xen-dom0 (2.1) unstable; urgency=low

  * Fix typo in init script 
  * depend  on dsniff for arpspoof

 -- Sam Hartman <hartmans@debian.org>  Fri, 24 Aug 2007 17:22:20 -0400

sipb-xen-dom0 (2) unstable; urgency=low

  * Add qemu-dm-sipb written by andersk to get us the  domain ID in qemu-ifup 
  * Add vif-sipbroute, a version of vif-route that does better netwwork
    isolation and has initial but useless ipv6 support.  This version also
    uses arpspoof to take over an address for domain migrations.   
  * Add init script to enable rp_filter, proxy_arp and forwarding for the network config.

 -- Sam Hartman <hartmans@debian.org>  Fri, 24 Aug 2007 17:17:23 -0400

sipb-xen-dom0 (1) unstable; urgency=low

  *  First version

 -- Sam Hartman <hartmans@debian.org>  Sat,  4 Aug 2007 20:20:32 -0400