GitLab

One of our Haskell tests asserts that the Python and Haskell user and
group constants match. This patch fixes the order in which the mock
Python code outputs the users and groups to match the order of the
Haskell-side enumeration.
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

- Fix the test for setting file_storage_dir, which didn't check if the
  value was really set.
- Add tests for shared_file_storage_dir, which were missing completely.
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

With the call missing, it wasn't possible to change the directory after
cluster initialization. Fixes #1101.
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

The options --syslog, --no-user-checks, --no-voting,
and --yes-do-it were already described in the description.
Add them to the synopsis as well.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

The -b option was already described in the description of the daemon,
add it to the synopsis as well. While there, sort the synopsis to
reflect the order in which the options are described; this is also
consistent with the argument order in the synopsis of other Ganeti
daemons.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

The options -p, -b, --syslog, and --no-user-check were already described
in the description; add them to the synopsis as well.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

* stable-2.11
  (no changes)

* stable-2.10
  Substitute 'suffix' for 'revision'
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Helga Velroyen <helgav@google.com>

...so that the implicit retry on connection failure
is handled gracefully.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Add tests that verify the defining properties of safeUpdateLocksWaiting.

1.) If the state contains no pending request by the requester, then
    updateLocksWaiting and safeUpdateLocksWaiting coincide.

2.) safeUpdateLocksWaiting is idempotent on all states.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

With our timeouts on connections, we have to deal with connections
being interrupted at any time. Therefore, we provide a repeatable
version of updateLocksWaiting that gracefully ignores requests that
have already been recorded.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

...so that it can be repeated, if necessary.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

In order to have update requests repeatable, always
accept requests that do not require any change to the
state. Note that this is not implied by the current
definition, as the request might ask for two locks at
different level, and thus the repetition would violate
lock order.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Normally, we should not place instances on nodes that do
not have enough disks. However, there is one exception: if
we failover an instance from an offline node, that node can
well be secondary of that instance---the fact that it is currently
primary proves it has enough disks space. The reason why we have
to handle that case special is that if a node is offline, we sometimes
cannot determine the amount of disk available, hence the conservative
estimation is 0.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

* stable-2.10
  Substitute 'suffix' for 'revision'
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Avoiding lint errors on the onehand, and code complexity
on the other.
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

This patch fixes a bug in the comparison of the
'ssconf_master_candidate_certs' and disables the test
for vcluster.
Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

Currently, there is not test for just running renew-crypto
to only renew the cluster certificate (and not the node
certificates). This patches adds the test.
Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

In order to hunt down some flakiness, this patch adds
and additional check to the QA of renew-crypto to ensure
the state of ssconf_master_candidate_certs.
Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Currently the cluster is only verified after a series of
renew-crypto operations is carried out. This makes it hard
to trace errors back to originating call. This patch adds
a verifcation step after each renew-crypto call.
Signed-off-by: Helga Velroyen <helgav@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Signed-off-by: Lisa Velden <velden@google.com>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

This patch, extends the 'ganeti.utils.storage_unittest.py' unit test
with a new test for the Gluster storage type. Also, the current unit
tests are updated accordingly to correspond to the latest changes.
Signed-off-by: Dimitris Bliablias <bl.dimitris@gmail.com>
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Cherry-picked from 42fdf9d0Signed-off-by: Lisa Velden <velden@google.com>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

This patch, extends the file-based storage type verification checks to
include the new Gluster storage type. Those modifications include a
missing check for the gluster storage directory at cluster
initialization time, and extends the 'gnt-cluster verify' command to
also verify the gluster file-storage path.
Signed-off-by: Dimitris Bliablias <bl.dimitris@gmail.com>
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Cherry-picked from 604c8be2Signed-off-by: Lisa Velden <velden@google.com>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

This patch, adds the 'gluster' storage type to the set of storage types
for which full node storage reporting is available. This set is used by
the 'LUNodeQueryStorage' logical unit for getting information on storage
units on node(s).
Signed-off-by: Dimitris Bliablias <bl.dimitris@gmail.com>
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Cherry-picked from 7ec47185Signed-off-by: Lisa Velden <velden@google.com>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Currently, the 'gluster' and 'sharedfile' disk templates are both mapped
to the Shared File storage type. This compromise causes the Gluster block
devices to be interpreted in the config as 'sharedfile' devices, instead
of 'gluster', and subsequently to be wrongly handled as Shared File
disks. This behavior, currently makes the 'gluster' disk template not
functional.

This patch, fixes this issue by introducing the Gluster storage type,
which will be exclusively used by the 'gluster' disk template.
Signed-off-by: Dimitris Bliablias <bl.dimitris@gmail.com>
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

Cherry-picked from ba550291Signed-off-by: Lisa Velden <velden@google.com>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

* stable-2.11
  (no changes)

* stable-2.10
  Check for gnt-cluster before running gnt-cluster upgrade
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

* stbale-2.10
  Check for gnt-cluster before running gnt-cluster upgrade
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

With Ganeti imposing no requirement on Xen versions, the configuration
file that should be present varies greatly on the choice of toolstack
and Xen version. xend-config.sxp was considered obligatory, but in
higher versions and with xl, it is superseded by xl.conf and considered
deprecated. This patch makes it have the same status as xl.conf - Ganeti
will distribute it but not require its presence.
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

When ganeti is removed (not purged) `/etc/cron.d/ganeti` is not deleted,
thus after a reboot cron tries to execute gnt-cluster upgrade and fails.

The same pattern is used on all other cron entries.
Signed-off-by: Christos Trochalakis <christos@skroutz.gr>
Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

..  with the latest bugfixes and known issues
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

After cherry-picking the changes needed for GHC7.8, the former isn't
used any more.
Signed-off-by: Petr Pudlak <pudlak@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

to ensure we are not reporting stray volumes on other VGs as orphans.
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Prior to patch eeda5882, Ganeti checked all the LVs on nodes under
its control, and dutifuly declared any stray volumes as orphans in
warnings emitted during cluster-verify. After the patch, the nodes
returned information related only to the LV set as *the* Ganeti VG,
bypassing this problem.

Unfortunately, this led Ganeti to report the disks of instances created
on a different VG as non-existent. This patch fixes things by reverting
patch eeda5882 and making Ganeti warn about orphan volumes only if
they belong to the Ganeti-administered VG.
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

When opening domain sockets for communication, the Haskell daemons did
not set any permissions for the sockets, defaulting to 0700. This was
fine when all of them ran as root, but was bound to cause trouble in a
split-user setup. The first issue is RAPI access after master-failover,
where RAPI could not send make any inquiries until the watcher restored
the desired permissions of the socket.

This patch modifies Luxid to use a g+rw socket, and leaves other servers
to their default of 0600.
Signed-off-by: Hrvoje Ribicic <riba@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Signed-off-by: Lisa Velden <velden@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

It turns out that GHC 7.8's -dynamic-too is forbidden not only for
profiling builds object files, but also for those using HPC coverage.

This commit accordingly renames HFLAGS_NOPROF to HFLAGS_DYNAMIC
(since it now is not conditional on profiling any more),
and makes sure that it is not used for profiling or HPC.

This way we achieve that, for profiling+coverage builds
  - .dyn_o files are available for use in TH in the following 3 cases
  - .o files are for the normal binaries
  - .hpc_o files are for coverage-enabled test binaries
  - .prof_o files are for profiling binaries

We make .hpc_o files depend on the .o files because the creation of
the .o files will also create the .so files needed for TH.
This was already in place for .prof_o files.

This requires that HFLAGS now also contains `-itest/hs` since
the rule for .o files is now also responsible for building .o/.so files
in test/hs.
Signed-off-by: Niklas Hambuechen <niklash@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Cherry-picked-from: 1ad14f38Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

This fixes
  ./configure --enable-haskell-tests && make hs-tests
builds on GHC 7.8.

It changes `-osuf .ext` to `-osuf ext`; the old form was allowed in
GHC <= 7.6, but GHC 7.8 seems to have problems with it:
  https://ghc.haskell.org/trac/ghc/ticket/9760

The form without dot works in all GHC versions, and we don't have
dots in other -osuf arguments in the same file either, so this
was probably an accident.
Signed-off-by: Niklas Hambuechen <niklash@google.com>
Reviewed-by: Klaus Aehlig <aehlig@google.com>

Cherry-picked-from: 9664aff9Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>

GHC 7.8 switched to dynamic linking being used for ghci, which requires
that .so file are being built for modules that are loaded for TH.

For this reason, GHC >= 7.8 has a -dynamic-too flag, which we now use.

However, -dynamic-too must not be enabled for profiling builds,
which is why this commit introduces a HFLAGS_NOPROF variable
(currently only containing -dynamic-too) that must be passed to
all GHC invocations that are not creating profiling output.
Signed-off-by: Niklas Hambuechen <niklash@google.com>
Reviewed-by: Hrvoje Ribicic <riba@google.com>

Cherry-picked-from: 083776b9Signed-off-by: Klaus Aehlig <aehlig@google.com>
Reviewed-by: Petr Pudlak <pudlak@google.com>