the VM UUID conflicts as well.
Also, remove the previous_restart_time node when explicitly rebooting a guest --
the user ought to be able to drive reboots faster than the safety catch.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Added a recreate call to ImageHandler, allowing the subclasses of that to
hook into the code that runs when xend restarts. This allows us in particular
to reregister the watches for HVM shutdown, and read the PID of qemu-dm from
the store.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
The 'builder' (eg. image type) should either be 'hvm' or 'linux' as
image.py only supports these at the moment.
Signed-off-by: Alastair Tse <atse@xensource.com>
Add function to list all vbds attached to a VM.
Update VM config builder to specify the default as 'linux'
Signed-off-by: Alastair Tse <atse@xensource.com>
Get rid of the case skewing in XendAPI.
Remove _to_xml methods because it has been removed from the latest API
docs.
Signed-off-by: Alastair Tse <atse@xensource.com>
Attempt to get rid of most SXP dependency in Xend except in
configuration parsing and certain places where it is
required.
XendConfig now stores configuration options similar to Xen API VM
and devices attributes but exports an SXP version to certain things
like xm and XendCheckpoint. All access to VM and device attributes
should now use Python dictionaries rather than the SXP parser.
Signed-off-by: Alastair Tse <atse@xensource.com>
domains_lock. This makes the VM.start call blocking, but allows xm list to
work while it is blocked. Move the handling of start_paused out of
XendDomainInfo.start as part of this.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
unpause the VM on start by default. This should fix problems seen recently
whereby devices are not attached to the VM by the time they boot, as
xm create was expecting to be able to wait for the devices before unpausing.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
this handled by a new message (xend.domains_with_state) and make xend.domains
equivalent to xend.domains_with_state(running). This restores the Xen 3.0.3
semantics of xend.domains.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
to use PAM, or to be turned off entirely, on a listener by listener basis.
Listen on a different unix domain socket for the Xen-API server, so that it
can co-exist with the others.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
code from the relocation protocol handling and some plumbing. Add a new
configuration entry for the Xen-API server, including use of this mechanism.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Also simplify the HVM builder interface by doing more work in
the python wrapper, and fix mapping of shared_info page after the
change to map foreign pages by GMFN rather than MFN.
Signed-off-by: Keir Fraser <keir@xensource.com>
Add a resumed domain to the self.domains list after XendCheckpoint.restore,
to ensure that it does not get recreated immediately afterwards.
Perform domain_resume under the domains_lock, given that the domain needs to
be added.
These fixes together should fix xm suspend and xm resume.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
shutting down -- they don't have exit handling themselves and ought to just
be made daemon threads.
Added some logging to dialog shutdown problems.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
The Xen IPI spec states that destroy, hard_shutdown, pause, resume, start, suspend, and unpause return void. This patch makes it so and permits proper parsing of the response by clients (libxen) that follow the spec.
Signed-off-by: Jim Fehlig <jfehlig@novell.com>
devices' state when a VM configuration file is destroyed. Currently only
the vTPM device's state needs to be explicitly destroyed.
I am also surrounding the saving of the managed domain's configuration
with a try-catch.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
that it logs everything written to stderr by xc_save and xc_restore as
errors whereas in fact the vast majority of this output is
information/debug (and all actual errors are marked by the string ERROR:
at the start of the message) -- this is confusing to folks looking at
the logs and makes automated log analysis tricky.
Fix is to scan for the ERROR: string and log anything without it using
log.info instead.
Signed-off by: Simon Graham <Simon.Graham@stratus.com>
Alastair Tse