[PATCH stable-2.15 28/37] Disable pylint unsubscriptable-object warning

In these cases result is subscriptable. Signed-off-by: Brian Foley --- lib/cmdlib/instance_create.py| 8 lib/cmdlib/instance_migration.py | 3 ++- lib/cmdlib/instance_storage.py | 2 +- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git

[PATCH stable-2.15 31/37] Quell trailing newline

Signed-off-by: Brian Foley --- autotools/build-rpc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/autotools/build-rpc b/autotools/build-rpc index d3ccad6..1d59aae 100755 --- a/autotools/build-rpc +++ b/autotools/build-rpc @@ -212,7 +212,7 @@ def main():

[PATCH stable-2.15 35/37] Quell cell-var-from-loop warning

Signed-off-by: Brian Foley --- autotools/build-bash-completion | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/autotools/build-bash-completion b/autotools/build-bash-completion index 4a064ba..e0ad9b2 100755 --- a/autotools/build-bash-completion

[PATCH stable-2.15 23/37] Disable pylint unpacking-non-sequence warning

For these cases, self.ia_result actually will have a tuple assigned to it, it's just hard for static analysis tools to see it. Signed-off-by: Brian Foley --- lib/cmdlib/instance.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/cmdlib/instance.py

[PATCH stable-2.15 22/37] Disable pylint misplaced-comparison-constant warning

It's correct, and easier to read a < x < b in this case. Signed-off-by: Brian Foley --- lib/utils/__init__.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/utils/__init__.py b/lib/utils/__init__.py index bdd9761..d3a71d5 100644 ---

[PATCH stable-2.15 33/37] Quell too-many-boolean-expressions

Signed-off-by: Brian Foley --- tools/move-instance | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/move-instance b/tools/move-instance index 4ca6327..84b5972 100755 --- a/tools/move-instance +++ b/tools/move-instance @@ -943,9 +943,9 @@ def

[PATCH stable-2.15 24/37] Disable incorrect pylint not-callable warning

REQ_REQUEST will be a parameter validating function from ht, but pylint can't statically determine this. Signed-off-by: Brian Foley --- lib/masterd/iallocator.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/masterd/iallocator.py

[PATCH stable-2.15 29/37] Quell consider-using-enumerate warning

Signed-off-by: Brian Foley --- lib/config/verify.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/config/verify.py b/lib/config/verify.py index e53b384..2d8b99e 100644 --- a/lib/config/verify.py +++ b/lib/config/verify.py @@ -107,8 +107,8 @@ def

[PATCH stable-2.15 30/37] Quell bad-whitespace warning

Signed-off-by: Brian Foley --- lib/objects.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/objects.py b/lib/objects.py index 6235947..13e5ee2 100644 --- a/lib/objects.py +++ b/lib/objects.py @@ -1254,7 +1254,7 @@ class Instance(TaggableObject):

[PATCH stable-2.15 26/37] Disable unwanted pylint wrong-import-position warnings

For these, we want to import some modules in a particular order, so disable the pylint warnings. Signed-off-by: Brian Foley --- autotools/check-imports | 3 ++- lib/build/sphinx_ext.py | 3 +++ lib/ovf.py | 3 ++- 3 files changed, 7 insertions(+), 2 deletions(-)

[PATCH stable-2.15 20/37] Disable pylint eval-used warning

Signed-off-by: Brian Foley --- lib/build/sphinx_ext.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/build/sphinx_ext.py b/lib/build/sphinx_ext.py index 6b30894..37f1574 100644 --- a/lib/build/sphinx_ext.py +++ b/lib/build/sphinx_ext.py @@ -288,7

[PATCH stable-2.15 37/37] Quell pylint socket.timeout warning

socket.timeout doensn't have string or errno attributes. Signed-off-by: Brian Foley --- lib/hypervisor/hv_kvm/monitor.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/hypervisor/hv_kvm/monitor.py b/lib/hypervisor/hv_kvm/monitor.py index

[PATCH stable-2.15 25/37] Disable pylint unused-wildcard-import warning

+++ b/lib/cli.py @@ -56,7 +56,7 @@ from ganeti import pathutils from ganeti import serializer import ganeti.cli_opts # Import constants -from ganeti.cli_opts import * # pylint: disable=W0401 +from ganeti.cli_opts import * # pylint: disable=W0401,W0614 from ganeti.runtime import (Get

[PATCH stable-2.15 27/37] Disable pylint bare-except warning

This is actually dead code which wasn't removed until Ganeti 2.16 Signed-off-by: Brian Foley <bpfo...@google.com> --- lib/server/masterd.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/server/masterd.py b/lib/server/masterd.py index be98f69..f8787d4 100644 ---

[PATCH stable-2.15 21/37] Disable incorect pylint simplify-if-statement warning

Signed-off-by: Brian Foley --- lib/cmdlib/instance_create.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/cmdlib/instance_create.py b/lib/cmdlib/instance_create.py index aec9d9f..1b1abf8 100644 --- a/lib/cmdlib/instance_create.py +++

[PATCH stable-2.15 32/37] Remove pylint tests removed in pylint 2.0

Signed-off-by: Brian Foley --- pylintrc | 2 -- pylintrc-test | 2 -- 2 files changed, 4 deletions(-) diff --git a/pylintrc b/pylintrc index 47be2b8..342eeff 100644 --- a/pylintrc +++ b/pylintrc @@ -20,7 +20,6 @@ evaluation = 10.0 - ((float(5 * error + warning +

[PATCH stable-2.15 34/37] Use default value lambda param to avoid cell-var-from-loop

Signed-off-by: Brian Foley --- lib/tools/burnin.py | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/lib/tools/burnin.py b/lib/tools/burnin.py index 3054d37..a32e4db 100755 --- a/lib/tools/burnin.py +++ b/lib/tools/burnin.py @@ -677,8 +677,10 @@ class

[PATCH stable-2.15 36/37] Quell the pylint wrong-import-order warnings

pycurl import random import re import shutil @@ -60,8 +61,8 @@ import stat import tempfile import time import zlib -import contextlib -import collections + +import pycurl from ganeti import errors from ganeti import http diff --git a/lib/cli.py b/lib/cli.py index 8622738..f01d8d9 100644 ---

[PATCH stable-2.15 19/37] Disable pylint invalid-name warning

Signed-off-by: Brian Foley --- lib/cli_opts.py | 2 +- lib/compat.py | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/cli_opts.py b/lib/cli_opts.py index ae58ede..6a7a33a 100644 --- a/lib/cli_opts.py +++ b/lib/cli_opts.py @@ -551,7 +551,7 @@

[PATCH stable-2.15 16/37] Disable pylint warning about use of deprecated pragmas

Autogenerated RPC code like that in metad.py uses pylint: disable-all to disable unimportant warnings. This option has been deprecated in favour of the new skip-file option in pylint >= 0.27, but we want to keep the old option for backward compatibility with earlier pylints. Signed-off-by: Brian

[PATCH stable-2.15 18/37] Disable pylint import-self warning

In this case it's a deliberate test to make sure that the contents of lib/ have been correctly installed in a dir called ganeti. Signed-off-by: Brian Foley <bpfo...@google.com> --- lib/__init__.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/__init__.py b/lib/__

[PATCH stable-2.15 17/37] Disable some pylint unused-import warnings

import new as md5_hash # pylint: disable=W0611 import sha sha1 = sha sha1_hash = sha.new diff --git a/lib/rapi/testutils.py b/lib/rapi/testutils.py index 3c09db2..40989dd 100644 --- a/lib/rapi/testutils.py +++ b/lib/rapi/testutils.py @@ -49,8 +49,8 @@ import ganeti.rpc.client as rpccl from

[PATCH stable-2.15 15/37] Remove pylint's star-args warning

ock.py | 2 +- tools/confd-client| 3 --- tools/ganeti-listrunner | 2 +- 31 files changed, 25 insertions(+), 50 deletions(-) diff --git a/Makefile.am b/Makefile.am index abaa325..d139d8b 100644 --- a/Makefile.am +++ b/Makefile.am @@ -2686,9 +2

[PATCH stable-2.15 12/37] Cleanup: Remove unused/duplicate module/fn import

nNodes + CheckImageValidity, EnsureKvmdOnNodes import ganeti.masterd.instance diff --git a/lib/cmdlib/instance_utils.py b/lib/cmdlib/instance_utils.py index 20cc3d4..3037d9e 100644 --- a/lib/cmdlib/instance_utils.py +++ b/lib/cmdlib/instance_utils.py @@ -44,7 +44,7 @@ from ganeti import pathutils

[PATCH stable-2.15 14/37] Disable pylint warning about unrecognised --disable

Allows us to disable warnings that are present in some versions of pylint but not others. Signed-off-by: Brian Foley --- Makefile.am | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/Makefile.am b/Makefile.am index 8b859c1..abaa325 100644 ---

[PATCH stable-2.15 13/37] Fix pylint >1.4 pycurl no-member warnings

Add pycurl to the whitelist of C extensions that can be loaded for the purposes of attribute checking. This works around a security feature introduced in pylint 1.4 to avoid arbitrary code injection. Signed-off-by: Brian Foley --- Makefile.am | 4 configure.ac | 4

[PATCH stable-2.15 09/37] Cleanup: Iterate dict rather than key list

The later explicitly constructs a new list of the dictionary's keys, which we don't need here. Quietens pylint's consider-iterating-dictionary warning. Signed-off-by: Brian Foley --- autotools/check-imports | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

[PATCH stable-2.15 10/37] Cleanup: Remove some unneeded pylint disables

Signed-off-by: Brian Foley --- lib/cmdlib/cluster/verify.py | 1 - lib/cmdlib/test.py | 1 - lib/server/noded.py | 1 - 3 files changed, 3 deletions(-) diff --git a/lib/cmdlib/cluster/verify.py b/lib/cmdlib/cluster/verify.py index dc5bac7..fe9e55e 100644

[PATCH stable-2.15 04/37] Cleanup: del is a statement not a function

Remove unnecessary parens to make pylint happy. Signed-off-by: Brian Foley --- lib/hypervisor/hv_kvm/monitor.py | 2 +- qa/qa_config.py | 2 +- test/py/ganeti.hypervisor.hv_kvm_unittest.py | 2 +- 3 files changed, 3 insertions(+), 3

[PATCH stable-2.15 11/37] Cleanup: Fix unidiomatic-typecheck

Signed-off-by: Brian Foley --- lib/cli.py | 2 +- lib/ovf.py | 2 +- lib/rpc_defs.py | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/cli.py b/lib/cli.py index c8d8d1d..e29bc02 100644 --- a/lib/cli.py +++ b/lib/cli.py @@ -2796,7 +2796,7 @@

[PATCH stable-2.15 06/37] Cleanup: Fix for/else with no break in AddAuthorizedKeys

After a refactoring in c178396, the only break statement in the for/else loop in AddAuthorizedKeys was removed. This means the else block is always executed, so remove the else: to quell a pylint warning. No functional change. Signed-off-by: Brian Foley --- lib/ssh.py | 14

[PATCH stable-2.15 07/37] Cleanup: StartInstance and RebootInstance return None

Quell assignment-from-none warning Signed-off-by: Brian Foley --- lib/backend.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/lib/backend.py b/lib/backend.py index 82aa493..3cd07d1 100644 --- a/lib/backend.py +++ b/lib/backend.py @@ -2988,9 +2988,8

[PATCH stable-2.15 08/37] Cleanup: Remove unused format key

No functional change, but quietens pylint unused-format-string-argument warning. Signed-off-by: Brian Foley --- lib/objects.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/lib/objects.py b/lib/objects.py index 96e7092..29dbbbc 100644 ---

[PATCH stable-2.15 02/37] Cleanup: Simplify boolean assignment

No functional change. Found by pylint's simplifiable-if-statement. Signed-off-by: Brian Foley --- lib/client/gnt_cluster.py | 7 +-- tools/cluster-merge | 5 + 2 files changed, 2 insertions(+), 10 deletions(-) diff --git a/lib/client/gnt_cluster.py

[PATCH stable-2.15 00/37] Cleanup for pylint 1.6.4

Ganeti's python code passes the pylint checks in the version of pylint included with Debian Jessie. Unfortunately this is a really old pylint (0.26 from 2012) and the latest stable pylint (1.6.4 from 2016) has a lot of additional checks. Ganeti produces many many warnings and errors when run

[PATCH stable-2.15 05/37] Cleanup: Replace map/filters with list comprehensions

No functional change, but more pythonic and quells pylint's deprecated-lambda warning. Signed-off-by: Brian Foley --- lib/cli.py | 14 ++ lib/cmdlib/cluster/verify.py | 6 +++--- lib/cmdlib/instance_storage.py | 2 +-

[PATCH stable-2.15 03/37] Cleanup: Use FOO not in BAR instead of not FOO in BAR

No functional change, but more pythonic, and makes pylint's unneeded-not warning happy. Signed-off-by: Brian Foley --- lib/utils/io.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/utils/io.py b/lib/utils/io.py index 9e74d68..49bab0d 100644 ---

[PATCH stable-2.15 01/37] Cleanup: Remove some unnecessary if (...) parens

No functional change, but makes pylint 1.1.0 a little happier, and is more pythonic. Signed-off-by: Brian Foley --- lib/cli.py| 4 ++-- lib/cmdlib/instance.py| 2 +- lib/hypervisor/hv_kvm/__init__.py | 4 ++-- lib/uidpool.py

Re: [PATCH stable-2.15] Fix invalid variable error for file-based disks

On Fri, Dec 02, 2016 at 08:28:13PM +0100, Iustin Pop wrote: > On 2016-12-02 18:03:33, Ganeti Development List wrote: > > Introduced by 94e252a33. Found by pylint's undefined-loop-variable > > warning. > > [snip] > Good pylint! LGTM. Thanks. Submitted. robb...@gent

Re: [PATCH stable-2.15] Fix coexistence of location tags and non-DRBD instances

Pop <ius...@google.com> LGTM, and thank you. I'm going to do a merge forward of the recent (and a batch of upcoming) 2.15 patches in the next few days, and this is a nice fix to include. Cheers, Brian. > --- > src/Ganeti/HTools/Cluster.hs | 4 ++-- > src/Ganeti/HTool

Re: Issue 1185 in ganeti: hbal: IntMap.!: key -1 is not an element of the map

Comment #17 on issue 1185 by gannef...@googlemail.com: hbal: IntMap.!: key -1 is not an element of the map https://code.google.com/p/ganeti/issues/detail?id=1185 The good news is great, and I saw your patch flow by on the -devel list, thanks for that. Guess it gets into 2.16 release

Re: Issue 1185 in ganeti: hbal: IntMap.!: key -1 is not an element of the map

Comment #16 on issue 1185 by iust...@gmail.com: hbal: IntMap.!: key -1 is not an element of the map https://code.google.com/p/ganeti/issues/detail?id=1185 Actually, I forgot the exact details since I last looked this up—it's expected to fail with extstorage, basically with anything other

[PATCH stable-2.15] Fix invalid variable error for file-based disks

Introduced by 94e252a33. Found by pylint's undefined-loop-variable warning. Signed-off-by: Brian Foley --- lib/cmdlib/instance_storage.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/cmdlib/instance_storage.py b/lib/cmdlib/instance_storage.py index

Re: [PATCH stable-2.15 1/2] FIX: Only set INSTANCE_NICn_NETWORK_NAME if net name is defined

This looks like a nasty bug! Thanks for fixing it. LGTM. On Fri, Dec 2, 2016 at 5:08 PM, 'Brian Foley' via ganeti-devel < ganeti-devel@googlegroups.com> wrote: > Found by pylint's using-constant-test warning. 'network' is a module, > 'net' is the network name for this NIC. &g

Re: [PATCH stable-2.15 2/2] FIX: Refactor DiagnoseOS to use a loop, not an inner fn

LGTM, thanks! On Fri, Dec 2, 2016 at 5:08 PM, 'Brian Foley' via ganeti-devel < ganeti-devel@googlegroups.com> wrote: > Found by pylint's cell-var-from-loop. > > The inner function tries to access a dictionary that's local to the > outer function, assuming that Python has simil

[PATCH stable-2.15 2/2] FIX: Refactor DiagnoseOS to use a loop, not an inner fn

Found by pylint's cell-var-from-loop. The inner function tries to access a dictionary that's local to the outer function, assuming that Python has similar lexical scoping rules to most languages. Unfortunately it doesn't. See

[PATCH stable-2.15 1/2] FIX: Only set INSTANCE_NICn_NETWORK_NAME if net name is defined

b/lib/cmdlib/instance_utils.py index b703c1e..20cc3d4 100644 --- a/lib/cmdlib/instance_utils.py +++ b/lib/cmdlib/instance_utils.py @@ -38,7 +38,6 @@ from ganeti import errors from ganeti import ht from ganeti import locking from ganeti.masterd import iallocator -from ganeti import network from

Re: [PATCH stable-2.16 3/7] Make FinalizeMigration{Src,Dst} more robust

On Fri, Dec 02, 2016 at 05:02:27PM +, 'Viktor Bachraty' via ganeti-devel wrote: > Migration is supported on KVM/Xen hypervisors only. Finalize methods are > called in both success or suspected failure (the master node can't > certainly know about failures - e.g. in case of timeou

[PATCH stable-2.16 3/7] Make FinalizeMigration{Src,Dst} more robust

Migration is supported on KVM/Xen hypervisors only. Finalize methods are called in both success or suspected failure (the master node can't certainly know about failures - e.g. in case of timeouts). Thus these functions need to be as robust as possible in order to be able to recover from failures

Re: [PATCH stable-2.16 3/7] Make FinalizeMigration{Src,Dst} more robust

make the code generic. I think removing a suffix from the domain name isn't a dangerous operation so we can safely do it on XL as well. On Fri, Dec 2, 2016 at 3:34 PM, Brian Foley <bpfo...@google.com> wrote: > On Fri, Dec 02, 2016 at 01:59:49PM +, 'Viktor Bachraty' via > ganeti

Re: Issue 1194 in ganeti: 'gnt-cluster verify' cannot detect drbd module version such as 8.4.8-1

Updates: Status: Fixed Comment #1 on issue 1194 by bpfo...@google.com: 'gnt-cluster verify' cannot detect drbd module version such as 8.4.8-1 https://code.google.com/p/ganeti/issues/detail?id=1194 Fix landed in stable-2.15 with 49980c6983469bab73b3076d44ae36ead639d972 -- You

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Updates: Status: Fixed Comment #7 on issue 1193 by bpfo...@google.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 (No comment was entered for this change.) -- You received this message because this project is configured

Re: [PATCH stable-2.16 5/7] Make migrate --cleanup more robust

On Fri, Dec 02, 2016 at 02:04:00PM +, 'Viktor Bachraty' via ganeti-devel wrote: > In case of Xen migrations, the most common failure case is when the > instance fails to freeze so the migration fails with domains running on > both target and source node. This patchs allows migrate

Re: [PATCH stable-2.16 7/7] StartInstance restores instance state if running

On Fri, Dec 02, 2016 at 02:02:21PM +, 'Viktor Bachraty' via ganeti-devel wrote: > Add a side effect to StartInstance on Xen hypervisor that restores the > Xen configuration in case it is missing. This is a common failure > scenario during migrations. The generic backend checks if the

Re: [PATCH stable-2.16 6/7] Allow migrate --cleanup to adopt an instance

On Fri, Dec 02, 2016 at 02:01:37PM +, 'Viktor Bachraty' via ganeti-devel wrote: > In some failure modes, Ganeti state of record may desync from actual > state of world. This patch allows migrate --cleanup to adopt an instance > if it is detected running on an unexpected node. >

Re: [PATCH stable-2.16 3/7] Make FinalizeMigration{Src,Dst} more robust

On Fri, Dec 02, 2016 at 01:59:49PM +, 'Viktor Bachraty' via ganeti-devel wrote: > Migration is supported on KVM/Xen hypervisors only. Finalize methods are > called in both success or suspected failure (the master node can't > certainly know about failures - e.g. in case of timeou

Re: [PATCH master 2/6] Support OS params removal in gnt-instance modify

expansion is worth the effort. Do you think you or anyone else might have a recurrent use case for this that might warrant its introduction? Cheers, Morg. On 2 December 2016 at 15:10, 'Brian Foley' via ganeti-devel < ganeti-devel@googlegroups.com> wrote: > On Fri, Dec 02, 2016 at 0

Re: Issue 1185 in ganeti: hbal: IntMap.!: key -1 is not an element of the map

Comment #15 on issue 1185 by gannef...@googlemail.com: hbal: IntMap.!: key -1 is not an element of the map https://code.google.com/p/ganeti/issues/detail?id=1185 Hi would be nice if that can get into 2.16 before its finally released, as I would love to use the nlocation tags, to ensure

Re: [PATCH master 2/6] Support OS params removal in gnt-instance modify

On Fri, Dec 02, 2016 at 04:53:51PM +0200, Yiannis Tsiouris wrote: > On 12/02, Brian Foley wrote: > > On Thu, Dec 01, 2016 at 12:34:11PM +0200, Yiannis Tsiouris wrote: > > > This patch extends 'gnt-instance modify' by allowing a user to remove a > > > list of public/private OS parameters from an

Re: [PATCH master 6/6] Add OS params removal in gnt-instance reinstall

On Thu, Dec 01, 2016 at 12:34:15PM +0200, Yiannis Tsiouris wrote: > This patch adds '--remove-os-parameters' and > '--remove-os-parameters-private' options to 'gnt-instance reinstall'. > Similarly to 'gnt-instance modify', the new options can be used to > perform reinstalls to OS providers that

Re: [PATCH master 5/6] Support clear OS params in gnt-instance reinstall

On Thu, Dec 01, 2016 at 12:34:14PM +0200, Yiannis Tsiouris wrote: > This extends 'gnt-instance reinstall' to support > '--clear-os-parameters' and '--clear-os-parameters-private' options from > 'gnt-instance modify'. E.g.: > > $ gnt-instance reinstall --clear-os-parameters -o \ > -O

Re: [PATCH master 4/6] Test new options of gnt-instance modify in RAPI

On Thu, Dec 01, 2016 at 12:34:13PM +0200, Yiannis Tsiouris wrote: > This patch adds tests to RAPI for gnt-instance modify options: > - '--clear-os-parameters'/'--clear-os-parameters-private' > - '--remove-os-parameters'/'--remove-os-parameters-private' So... this is just testing that these

Re: [PATCH master 2/6] Support OS params removal in gnt-instance modify

D'oh! To reply to myself: On Fri, Dec 02, 2016 at 02:26:21PM +, Brian Foley wrote: > Maybe add an else: print warning about removing an option that's not set. > > > + > > + for osp in remove_osparams_private: > > +if osp in private_parms: > > + raise

Re: [PATCH master 2/6] Support OS params removal in gnt-instance modify

On Thu, Dec 01, 2016 at 12:34:11PM +0200, Yiannis Tsiouris wrote: > This patch extends 'gnt-instance modify' by allowing a user to remove a > list of public/private OS parameters from an instance. This can be > useful before performing a reinstall to a new OS provider. Example > usage: > > $

Re: [PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

reschi' via > ganeti-devel wrote: > > Following issue #1194, this patch allows Ganeti to correctly > > parse drbd versions that also include a dash in their k-fix > > version component. > > > > This technically makes Ganeti treat x.y.z.k the same as x.y

Re: [PATCH master 1/6] Add clear OS params options in gnt-instance modify

On Fri, Dec 02, 2016 at 11:25:08AM +0200, Yiannis Tsiouris wrote: > This patch adds two new options in 'gnt-instance modify' which allow the > user to clear all current public/private OS parameters of an instance. > This might be useful for OS providers that consider no parameters as > valid or,

[PATCH stable-2.16 5/7] Make migrate --cleanup more robust

In case of Xen migrations, the most common failure case is when the instance fails to freeze so the migration fails with domains running on both target and source node. This patchs allows migrate --cleanup to recover by running AbortMigrate() in case the instance is running on both the source and

[PATCH stable-2.16 7/7] StartInstance restores instance state if running

drive_uri) +@param block_devices: blockdevices assigned to this instance +""" +startup_memory = self._InstanceStartupMemory(instance) +self._MakeConfigFile(instance, startup_memory, block_devices) + def StartInstance(self, instance, block_device

[PATCH stable-2.16 6/7] Allow migrate --cleanup to adopt an instance

In some failure modes, Ganeti state of record may desync from actual state of world. This patch allows migrate --cleanup to adopt an instance if it is detected running on an unexpected node. Signed-off-by: Viktor Bachraty <vbachr...@google.com> --- lib/cmdlib/instance_migration.p

[PATCH stable-2.16 3/7] Make FinalizeMigration{Src,Dst} more robust

Migration is supported on KVM/Xen hypervisors only. Finalize methods are called in both success or suspected failure (the master node can't certainly know about failures - e.g. in case of timeouts). Thus these functions need to be as robust as possible in order to be able to recover from failures

Re: Issue 1185 in ganeti: hbal: IntMap.!: key -1 is not an element of the map

Comment #14 on issue 1185 by iust...@gmail.com: hbal: IntMap.!: key -1 is not an element of the map https://code.google.com/p/ganeti/issues/detail?id=1185 Hi Joerg, that's excellent information, I'll take this into account when I get back to fixing this, thanks! -- You received

Re: Issue 1185 in ganeti: hbal: IntMap.!: key -1 is not an element of the map

Comment #13 on issue 1185 by gannef...@googlemail.com: hbal: IntMap.!: key -1 is not an element of the map https://code.google.com/p/ganeti/issues/detail?id=1185 Hi I just experienced this error too. I'm neither using plain nor lvm nor drbd storage, I have exclusively extstorage. What

Re: [PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

On Fri, Dec 02, 2016 at 11:20:53AM +, 'Federico Morg Pareschi' via ganeti-devel wrote: > Following issue #1194, this patch allows Ganeti to correctly > parse drbd versions that also include a dash in their k-fix > version component. > > This technically makes Ganeti treat x

[PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

Following issue #1194, this patch allows Ganeti to correctly parse drbd versions that also include a dash in their k-fix version component. This technically makes Ganeti treat x.y.z.k the same as x.y.z-k with regards to compatibility. For practicality we can assume that these versions

Re: Issue 1195 in ganeti: Wrong GANETI_CLUSTER and GANETI_OBJECT_TYPE values in hook environment for master-ip-turnup-post

Comment #1 on issue 1195 by xal...@gmail.com: Wrong GANETI_CLUSTER and GANETI_OBJECT_TYPE values in hook environment for master-ip-turnup-post https://code.google.com/p/ganeti/issues/detail?id=1195 The OBJECT_TYPE and CLUSTER values are OK for an instance failover operation. -- You

Issue 1195 in ganeti: Wrong GANETI_CLUSTER and GANETI_OBJECT_TYPE values in hook environment for master-ip-turnup-post

Status: New Owner: New issue 1195 by xal...@gmail.com: Wrong GANETI_CLUSTER and GANETI_OBJECT_TYPE values in hook environment for master-ip-turnup-post https://code.google.com/p/ganeti/issues/detail?id=1195 What software version are you running? Please provide the output of &quo

Re: [PATCH master 0/6] Clear/remove public/private OS params in gnt-instance modify/reinstall

4 +++-- > lib/cmdlib/instance_operation.py | 16 > lib/cmdlib/instance_set_params.py | 45 - > lib/rapi/rlib2.py | 16 +++- > man/gnt-instance.rst | 25 > src/Ganeti/HTools/Rep

Re: [PATCH master 3/6] Add cmdlib tests for removing instance OS params

On 1 December 2016 at 10:34, Yiannis Tsiouris wrote: > From: Nikos Skalkotos > > Add LUInstanceSetParams tests for removing individual public & private > parameters, as well as, for clearing out public & private parameters. > > Signed-off-by: Nikos Skalkotos

Re: [PATCH master 1/6] Add clear OS params options in gnt-instance modify

ce modify --clear-os-parameters-private > > Signed-off-by: Yiannis Tsiouris <ts...@grnet.gr> > --- > lib/cli_opts.py | 15 +++ > lib/client/gnt_instance.py| 8 +++- > lib/cmdlib/instance_set_params.py | 23 ++-

Re: [PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

On Thu, Dec 01, 2016 at 04:09:56AM -0800, Ganeti Development List wrote: > > On Thursday, December 1, 2016 at 11:53:43 AM UTC, Iustin Pop wrote: > > > > On Thu, Dec 01, 2016 at 11:38:35AM +, Ganeti Development List wrote: > > > Following issue #1194, this patch

Re: [PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

On Thursday, December 1, 2016 at 11:53:43 AM UTC, Iustin Pop wrote: > > On Thu, Dec 01, 2016 at 11:38:35AM +, Ganeti Development List wrote: > > Following issue #1194, this patch allows Ganeti to correctly > > parse drbd versions that also include a dash in their

Re: [PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

On Thu, Dec 01, 2016 at 11:38:35AM +, Ganeti Development List wrote: > Following issue #1194, this patch allows Ganeti to correctly > parse drbd versions that also include a dash in their k-fix > version component. This means 8.4.8-1 and 8.4.8.1 will be treated the same. Is this th

[PATCH stable-2.15] Fix for incorrect parsing of DRBD versions

Following issue #1194, this patch allows Ganeti to correctly parse drbd versions that also include a dash in their k-fix version component. Signed-off-by: Federico Morg Pareschi <m...@google.com> --- lib/storage/drbd_info.py| 17 +++-- t

Issue 1194 in ganeti: 'gnt-cluster verify' cannot detect drbd module version such as 8.4.8-1

Status: New Owner: New issue 1194 by ndgoogl...@netdirect.fr: 'gnt-cluster verify' cannot detect drbd module version such as 8.4.8-1 https://code.google.com/p/ganeti/issues/detail?id=1194 gnt-cluster (ganeti v2.14.2-177-g003cd9a) 2.15.2 Software version: 2.15.2 Internode protocol

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #6 on issue 1193 by yiannis@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 Fix landed in a789817c6eba26658d7b31ff5c276a8cc07b999a. Thanks! -- You received this message because this project is configured to send

Re: [PATCH master] Fix for instance reinstall not updating config

This patch has been pushed to 2.16, thanks Viktor for reviewing it. On Tuesday, November 29, 2016 at 10:07:52 AM UTC, Viktor Bachraty wrote: > > I wouldn't mind backporting up to 2.16, since we haven't released stable > (yet). > > On Tue, Nov 29, 2016 at 9:59 AM, 'Federico Pares

Re: [PATCH master] Fix for instance reinstall not updating config

I wouldn't mind backporting up to 2.16, since we haven't released stable (yet). On Tue, Nov 29, 2016 at 9:59 AM, 'Federico Pareschi' via ganeti-devel < ganeti-devel@googlegroups.com> wrote: > I also think that originally this behavior was intentional, however it is > very non-int

Re: [PATCH master] Fix for instance reinstall not updating config

this be backported to 2.17 maybe? Morg. On Monday, November 28, 2016 at 10:56:05 PM UTC, Iustin Pop wrote: > > On 2016-11-28 14:10:39, Ganeti Development List wrote: > > From the modified log line it looks like it was intentional, but as we > > discussed I don't see a good reason for thi

Re: [PATCH master] Fix for instance reinstall not updating config

lso is it happening only on master ? Other than that LGTM. On Mon, Nov 28, 2016 at 1:58 PM, 'Federico Morg Pareschi' via ganeti-devel < ganeti-devel@googlegroups.com> wrote: > As per issue #1193, instance reinstalls with modified OS parameters > did not persist among reinstalls and

[PATCH master] Fix for instance reinstall not updating config

As per issue #1193, instance reinstalls with modified OS parameters did not persist among reinstalls and were not saved to the cluster's config data. With this patch, this problem should be fixed. Signed-off-by: Federico Morg Pareschi --- lib/cmdlib/instance_operation.py | 8

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #5 on issue 1193 by yiannis@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 I've already checked the code in 2.10, 2.12, 2.15, 2.16, latest and it 's like this everywhere. My guess is that noone really used

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #4 on issue 1193 by skalk...@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 Well, reinstalling an instance means keep the hardware intact and allow the OS provider (which may be different from the original one

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #3 on issue 1193 by m...@google.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 This is an interesting one... I'm not entirely sure if this is a bug or a "feature". It depends on what we expect from

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #2 on issue 1193 by yiannis@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 Possibly relates to #913 -- You received this message because this project is configured to send all issue notifications

Re: Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Comment #1 on issue 1193 by yiannis@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 s/early_change2/tsiou in the command output section. Sorry for that! :) -- You received this message because this project is configured

Issue 1193 in ganeti: gnt-instance reinstall doesn't update config.data

Status: New Owner: New issue 1193 by yiannis@gmail.com: gnt-instance reinstall doesn't update config.data https://code.google.com/p/ganeti/issues/detail?id=1193 What software version are you running? Please provide the output of "gnt- cluster --version", "gnt-

Re: Issue 1178 in ganeti: Live migration fails with permission error on /var/run/ganeti/kvm-hypervisor

Comment #4 on issue 1178 by bcandler...@googlemail.com: Live migration fails with permission error on /var/run/ganeti/kvm-hypervisor https://code.google.com/p/ganeti/issues/detail?id=1178 I am seeing this problem again, also in a cluster mid-way through upgrade. root@wrn-vm2:~# gnt-instance

Issue 1192 in ganeti: Too old dependencies

Status: New Owner: New issue 1192 by actionmysti...@gmail.com: Too old dependencies https://code.google.com/p/ganeti/issues/detail?id=1192 What software version are you running? Please provide the output of "gnt- cluster --version", "gnt-cluster version", a

Re: Issue 695 in ganeti: support kvm run-time memory changes (ballooning)

Comment #15 on issue 695 by bcandler...@googlemail.com: support kvm run-time memory changes (ballooning) https://code.google.com/p/ganeti/issues/detail?id=695 See also: https://lists.gnu.org/archive/html/qemu-devel/2015-10/msg02280.html -- You received this message because this project

Re: Issue 695 in ganeti: support kvm run-time memory changes (ballooning)

Comment #14 on issue 695 by bcandler...@googlemail.com: support kvm run-time memory changes (ballooning) https://code.google.com/p/ganeti/issues/detail?id=695 but what if i want an instance to have for example 2G per default (on startup) but the ability to increase it to a max of 10G

<    1   2   3   4   5   6   7   8   9   10   >