Renaming the 'block' directory to 'storage', because I plan to
place code there that is related to file storage and leaving
it named 'block' would be misleading.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

Renaming 'storage.py' to 'container.py'. It will be moved into the new
'storage' directory, which will come in later patches to avoid clashes of
notation.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

The order of evaluation of the conditions is changed, so it's easier to add
more (foreseen) checks for exclusive_storage.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Prereq checks relative disks are grouped together and moved in a separate
method. This reduces the clutter in CheckPrereq().

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

In DRBD 8.4, Ganeti does no longer use indexed external meta data but
flexible external metadata. Therefore, there is no longer a meta data
index in the output of `drbdsetup show`.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

This check is performed during gnt-cluster verify and outputs warnings
if at least two different DRBD versions are found within a node group.
In such a case, all nodes with their installed DRBD version are
displayed.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

All functionality specific to a single DRBD8 devide is now in DRBD8Dev,
whereas functionality which is valid for the whole DRBD "installation"
on a device is collected in DRBD8.

This makes it possible to remove a couple of FIXME's and clarifies the
code in some areas.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

Right now the DRBD8 class has multiple responsibilities: a) it
reprensents a device which can be set up, grown, etc. and b) it
represents the whole DRBD system on a node which has a usermode helper,
which knows how to shut down _all_ devices, etc. Therefore, the DRBD8Dev
class will only be responsible for a device, whereas the soon to be
extraced DRBD8 class is responsible for the DRBD system.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

gnt-cluster verify should issue a warning if there are multiple DRBD
versions present in a node group. In order to do so, the DRBD version
has to be queryable from noded.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

The way disk templates were displayed in instance policies became confusing
since the introduction of the enabled_disk_templates cluster parameter.

Fix issue 440.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

DRBD 8.4.3 does not support specifying volume ids directly while
creating the meta device. It requires a slightly different syntax (using
flex-external as volume id) which is fortunately also supported by (at
least) DRBD 8.4.2.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

This fixes issue 432: when setting the volume group name, the presence
of the volume group was checked on all nodes. This patch now restricts
the check only to the vm-capable ones.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

Currently, when creating an instance with file storage fails,
the error message is rather general. This patch adds a hint
on what to do as a possible fix.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

-enable-kvm and -disable-kvm are deprecated in favor of accel=kvm passed
as a machine specification. This fixes compatibility with newer qemu
versions.

Signed-off-by: Guido Trotter <ultrotter@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

We had parameter validation when the runtime is generated, but in the
middle of the function. Move up, before any parameter is accessed.

Signed-off-by: Guido Trotter <ultrotter@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

Newer versions of kvm (from 1.4) don't support -nographic with
-daemonize. When -display exists, use -display none instead.

This fixes Issue 389.

Signed-off-by: Guido Trotter <ultrotter@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

This patch fixes issue 247. It empowers the user to specify
explicitly the vif type of a nic configuration of a Xen HVM
instance. This includes the option of setting it to '' on the
commandline, causing Ganeti to leave out the 'type' parameter
in the vif configuration completely.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

The DRBD84CmdGenerator class, which generates commands suited for DRBD
8.4, is added. A common baseclass for DRBD83CmdGenerator and
DRBD84CmdGenerator is introduced as well.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

In order to easily support the DRBD 8.4 command syntax, this patch
extracts all command generation logic (i.e. where calls to `drbdsetup`
are assembled) into a new class DRBD83CmdGenerator.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

Now the DRBD8 class actually checks which version of DRBD is running on
the system and chooses the right class to parse the `drbdsetup show`
output.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

Common functionality between the DRBD 8.3 and DRBD 8.4 parser has been
extracted into BaseShowInfo. A test which verifies the behaviour is
included, but the DRBD84ShowInfo class is not yet used within the DRBD8
class.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

Parsing the status (/proc/drbd) and `drbdsetup show` output takes quite
a bit of code, which has to be augmented by even more with the DRBD 8.4
support. So extract all the related classes into their own file.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

This class parses the `drbdsetup show` output and represents it in an
easily accessible format. It got extracted so that 1) the DRBD8 class
can focus more on DRBD logic rather than parsing and 2) it's easier to
adapt the parser to new formats.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

In the new context of the DRBD8Info class, the name
_JoinProcDataPerMinor doesn't fit any more. _JoinLinesPerMinor reflects
more precisely what's done, and doesn't imply that the data comes from
/proc/drbd.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

As the DRBD8 class got bigger due to the previous merge of BaseDRBD, now
parts of it are ripped out into DRBD8Info. This new class parses
/proc/drbd and exposes the information in an easily accessible way. This
allowed to simplify some methods in DRBD8 and do make the tests more
concise.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

BaseDRBD was probably useful when DRBD 0.7 and DRBD 8 were supported.
However, there is only one subclass of BaseDRBD remaining (DRBD8), and
the separation of responsibilities between those two classes was rather
randomly chosen.

The unification into one class also eases the introduction of
DRBD 8.4 support, as the responsibilities of the class(es) will be
distributed anew.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

That's what the method actually does, so there is no reason why we
should use a funny but unreadable name.

Signed-off-by: Thomas Thrainer <thomasth@google.com>
Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Michele Tartara <mtartara@google.com>

Command line options accept multiple min/max specifications in instance
policies. The output of show-ispecs-cmd is updated accordingly.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

Now instance policies can contain more than one min/max specs.  This is the
main element of the "Constrained instance sizes" section in the
"Partitioned Ganeti" design doc.

This is a big patch, but changing the type of a configuration item requires
to change all the code that handles it.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

This is needed to be able to validate the std spec against multiple min/max
spec pairs (appearing in next patches).

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

The current error message ends up saying something like
'192.168.5.5' != 'network1', which is sort of non-intuitive (yes, of
course they are different). Fix this by explaining better what the
erroneous condition actually is.

Signed-off-by: Guido Trotter <ultrotter@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

This patch makes the constant DEFAULT_ENABLED_DISK_TEMPLATES
a list instead of a set, because the order of the elements actually matters
as the first one is planned to be the default disk template for instance
creation. Also, this constant is always used like a list and thus providing it
as a set complicates the code that uses it.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

Was accidentally introduced by commit 912737ba.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

This patch does three things:

1. It deprecates the option '--no-lvm-storage' of 'gnt-cluster modify'
and 'gnt-cluster init'. Technically, it is not fully removed, but kept in order
to warn the user that it is no longer supported and that she should use
--enabled-disk-templates instead.

2. The consistency check between '--no-lvm-storage' and '--vg-name' is replaced
by checks between '--enabled-disk-templates' and '--vg-name'. There are these
cases:
- vg name, lvm disk template enabled = ok
- no vg name, lvm disk template enabled = error
- vg name, no lvm enabled = warning
- no vg name, no lvm enabled = ok
I added quite a lot of tests for all these and the transitions from each case
to another to the QA.

3. The check whether or not the volume group is available on all nodes is now
done only in these cases:
- the volume group name gets set and lvm is already enabled
- lvm is getting enabled and the volume group was set before

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

This patch add a couple of utility functions dealing with
storage types and disk templates.

Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

Currently, if converting an instance from plain to drbd fails after
renaming the original LVs, the instance is left in an inconsistent
state. This commit tries to undo the renaming if a failure occurs
on assembling a DRBD device, e.g., when device minor numbers are
exhausted. (Issue 229)

Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Thomas Thrainer <thomasth@google.com>

Opcode-specific implementation of the reason trail for the instance
startup operation.

Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

Opcode-specific implementation of the reason trail for the instance shutdown
operation.

Signed-off-by: Michele Tartara <mtartara@google.com>
Reviewed-by: Guido Trotter <ultrotter@google.com>

Command line options are brought in line with the specs change of previous
patch. Old options are still allowed in gnt-cluster init, where the
semantic will remain non-ambiguous even after introducing multiple specs.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

Specs used to specify limits in instance policies are always complete, even
at group level (only the whole limit set can be overridden). This is in
preparation for introducing multiple limits.

Signed-off-by: Bernardo Dal Seno <bdalseno@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>