---
www/manager6/ceph/Pool.js | 25 +++--
1 file changed, 7 insertions(+), 18 deletions(-)
diff --git a/www/manager6/ceph/Pool.js b/www/manager6/ceph/Pool.js
index df43be44..bc08c075 100644
--- a/www/manager6/ceph/Pool.js
+++ b/www/manager6/ceph/Pool.js
@@ -162,32 +162,21 @@ E
On 05/08/2017 12:31 PM, Dominik Csapak wrote:
> we improve the icons in the tree and the resource grid by
> differentiating between cluster online/offline status and no rrd data
>
> when we have no rrd data from a node/storage, instead of showing a
> red x (which is scary) even if the node is reac
, etc ... (verb + name)
* useful nitpicks from Thomas about code formatting
v3 changes:
* when command lines parameters are mandatory, do not hide them behind switches
Emmanuel Kasper (3):
Refactor the resolution of target disk format in own sub
Add ImportDisk module to import external disk
Each disk passed as paramater is add as 'unused[n]' in the vm.conf
(the default) or ide[n]|scsi[n]|sata[n]
We rely on qemu-img(1) convert heuristics to detect the file type,
as this works in most case.
---
PVE/QemuServer/ImportDisk.pm | 98
PVE/QemuServ
This will allow use to reuse the code in ImportDisk.pm.
---
PVE/QemuServer.pm | 33 ++---
1 file changed, 22 insertions(+), 11 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 9dbb17d..7e91ac6 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
The VM must be already existing, and the syntax is
qm importdisk 421 minix204.img pve4tank
where 421 is an already existing VM
---
PVE/CLI/qm.pm | 51 +++
1 file changed, 51 insertions(+)
diff --git a/PVE/CLI/qm.pm b/PVE/CLI/qm.pm
index 44439dd..f
Following OVF parameters will be extracted:
* VM name
* Memory
* Number of cores
* disks and their associated controllers
---
PVE/QemuServer/Makefile | 1 +
PVE/QemuServer/OVF.pm | 236
2 files changed, 237 insertions(+)
create mode 100644
Currently the following extracted parameters are used to create a VM:
* VM name
* Memory
* Number of cores
* Disks
---
PVE/CLI/qm.pm | 99 +++
1 file changed, 99 insertions(+)
diff --git a/PVE/CLI/qm.pm b/PVE/CLI/qm.pm
index f6fe4a2..971bc9c
This will allow code reuse for qm importovf
---
PVE/API2/Qemu.pm | 20
PVE/QemuServer.pm | 20
2 files changed, 24 insertions(+), 16 deletions(-)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index 54db5ab..f8e767d 100644
--- a/PVE/API2/Qemu.pm
+++ b/
, so we test
them
with the according regexp.
* Test if the disk image the OVF is refering to is contained in the current
directory
to prevent symlinks attacks.
Emmanuel Kasper (6):
Force overwriting existing compressed man pages and symbolic links
Add a build dependency to libxml
XML::LibXML is being actively developed, and has 91 reverse
depencies in Debian Stretch, no it should not go away.
---
control.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/control.in b/control.in
index d0601ac..7834e31 100644
--- a/control.in
+++ b/control.in
@@ -3,7 +3,7
This allows calling the 'make install' target twice in a row.
---
Makefile | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/Makefile b/Makefile
index b8b7d4b..d815c2a 100644
--- a/Makefile
+++ b/Makefile
@@ -81,12 +81,12 @@ install: ${PKGSOURCES}
install -D -m 06
This includes:
* a test script
* the manifests generated by exports from a VmWare Workstation
* disk images are generated from qemu-img, with a 2KB size
(it is possible to inspect the disk images with od -bc
they contain a VMDK header and the rest are null characters)
---
test/Makefile
---
qm.adoc | 68 +
1 file changed, 68 insertions(+)
diff --git a/qm.adoc b/qm.adoc
index d8ab448..87071b2 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -645,6 +645,74 @@ NOTE: It is not possible to start templates, because this
would modif
---
www/manager6/ceph/Pool.js | 23 +++
1 file changed, 7 insertions(+), 16 deletions(-)
diff --git a/www/manager6/ceph/Pool.js b/www/manager6/ceph/Pool.js
index df43be44..dff6ecc4 100644
--- a/www/manager6/ceph/Pool.js
+++ b/www/manager6/ceph/Pool.js
@@ -162,32 +162,23 @@ Ext
---
www/manager6/Utils.js | 1 +
www/manager6/window/SafeDestroy.js | 6 --
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/www/manager6/Utils.js b/www/manager6/Utils.js
index d1ada771..25d78f7b 100644
--- a/www/manager6/Utils.js
+++ b/www/manager6/Utils.js
@@ -613,
changes since v1:
* don't forge's to set the button to disabled by default, and pass the button
the selection Model, so the button is only activated when the user previously
select a pool in the pool grid
Emmanuel Kasper (2):
Adapt SafeDestroy component to be able to delete Ceph P
+1 for the idea as the ouput of this command is not totally clear, and
we use it in pvereport
minor nitpick downwards
On 05/24/2017 12:45 PM, Thomas Lamprecht wrote:
> Add column names at top of output, this allows easier understanding
> of what each column means.
>
> Use leading spaces on the p
(using local storage as the qm config example a bit below also uses that)
---
qm.adoc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/qm.adoc b/qm.adoc
index d8ab448..025a7a9 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -660,7 +660,7 @@ CLI Usage Examples
Create a new VM with 4 GB
In a previous commit, the inputField of SafeDestroy.js was converted
from a numberfield to a textfield to allow ceph pools names to be
entered.
This means the type of 'value' changed from number to string and
the strong equal test === was failing against a numeric id.
To solve this, we now always s
On 06/13/2017 05:04 PM, Dietmar Maurer wrote:
> I though this is disabled on a default installation? (local-lvm or local-zfs
> are enabled)?
on defaults installation the "local" storage is set to only allow
templates, backups and iso but it looks qm create does not check the
allowed content types
uses default storages as they are created during install process
---
qm.adoc | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/qm.adoc b/qm.adoc
index d8ab448..2a87484 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -658,9 +658,10 @@ create and delete virtual disks.
CLI Usage Example
On 06/14/2017 01:04 PM, Dietmar Maurer wrote:
>> but it looks qm create does not check the
>> allowed content types so qm create ... local:4 worked in all case
>
> I guess this is a bug we should fix?
I created https://bugzilla.proxmox.com/show_bug.cgi?id=1417
so we don't forget
_
use this to avoid code duplication.
Emmanuel Kasper (1):
Add run_or_get_killed utility
src/PVE/Tools.pm | 43 +++
1 file changed, 43 insertions(+)
--
2.11.0
___
pve-devel mailing list
pve-devel
This runs subroutine in a forked process
and kills it after a timeout
---
src/PVE/Tools.pm | 43 +++
1 file changed, 43 insertions(+)
diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm
index da7da5d..0bf94ae 100644
--- a/src/PVE/Tools.pm
+++ b/src/PVE/Tools.p
Wrap the -d test with run_or_get_killed sub this test
can make pvestatd hang on I/O wait when a nfsd process is stopped
---
PVE/Storage/Plugin.pm | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/PVE/Storage/Plugin.pm b/PVE/Storage/Plugin.pm
index cc4c7e7..3742258 100644
API calls return dates in seconds where the JS Date() constructor
wants millisecond.
For the correct computation use render_timestamp_human_readable.
A similar fix was commited in aaa89d0b880ea9ab97927872e64d4427bbe2232d
for VMs but Containers missed the fix.
---
18th of January 1970 was nice but
Otherwise the date is truncated by default
---
www/manager6/lxc/SnapshotTree.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/lxc/SnapshotTree.js b/www/manager6/lxc/SnapshotTree.js
index df5b31cf..c80fb439 100644
--- a/www/manager6/lxc/SnapshotTree.js
+++ b/www/m
serie looks good to me
On 06/19/2017 04:25 PM, Dominik Csapak wrote:
> to see the full date, and save space from the other columns
>
> Signed-off-by: Dominik Csapak
> ---
> www/manager6/grid/Replication.js | 6 +-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/www/manage
---
NB: 'pending' has not been translated yet. We have only 'pending changes'.
However new features might need new text to be understood.
PVE/CLI/pvesr.pm | 2 +-
www/manager6/grid/Replication.js | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/PVE/CLI/pvesr.
---
PVE/API2/Storage/Status.pm | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/PVE/API2/Storage/Status.pm b/PVE/API2/Storage/Status.pm
index 81ba67f..c6e3b1a 100644
--- a/PVE/API2/Storage/Status.pm
+++ b/PVE/API2/Storage/Status.pm
@@ -353,8 +353,8 @@ __PACKAGE__->register_m
On 06/23/2017 11:36 AM, Dominik Csapak wrote:
> we used a static emptytext at creation, which is wrong after editing
> now copied from qemu/Options.js (the name), but instead of deleting the
> hostname on the backend on an empty field, we set it to CT
> (this is also the default in the wizard)
>
emptyText is not submitted, because it does not count as value when
getSubmitValue() is
called on it in PVE.panel.InputPanel.getValues()
---
www/manager6/lxc/DNS.js | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/www/manager6/lxc/DNS.js b/www/manager6/lxc/DNS.js
index 2c087a
This will allow to disable replication, when for instance we
add a disk on a non-replicatable storage.
The option is hidden in the wizard, because at that time no VM
replication has been set.
---
www/manager6/qemu/HDEdit.js | 14 ++
1 file changed, 14 insertions(+)
diff --git a/www/ma
---
qm.adoc | 6 ++
1 file changed, 6 insertions(+)
diff --git a/qm.adoc b/qm.adoc
index 2a87484..b458d54 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -186,6 +186,12 @@ This provides a good balance between safety and speed.
If you want the {pve} backup manager to skip a disk when doing a backup of
---
www/manager6/lxc/ResourceEdit.js | 15 +++
1 file changed, 15 insertions(+)
diff --git a/www/manager6/lxc/ResourceEdit.js b/www/manager6/lxc/ResourceEdit.js
index bf7c76f1..41bd25cb 100644
--- a/www/manager6/lxc/ResourceEdit.js
+++ b/www/manager6/lxc/ResourceEdit.js
@@ -263,6 +263
---
www/manager6/lxc/ResourceEdit.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/lxc/ResourceEdit.js b/www/manager6/lxc/ResourceEdit.js
index 41bd25cb..dcf85f6e 100644
--- a/www/manager6/lxc/ResourceEdit.js
+++ b/www/manager6/lxc/ResourceEdit.js
@@ -506,7 +506,
---
qm.adoc | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/qm.adoc b/qm.adoc
index fce34aa..02a4555 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -40,7 +40,7 @@ devices, and runs as it were running on real hardware. For
instance you can pass
an iso image as a parameter to
---
pct.adoc | 20
1 file changed, 20 insertions(+)
diff --git a/pct.adoc b/pct.adoc
index 4ca8d7e..ab529a9 100644
--- a/pct.adoc
+++ b/pct.adoc
@@ -320,6 +320,26 @@ ACLs allow you to set more detailed file ownership than
the traditional user/
group/others model.
+Backu
Wrap the -d test with run_or_get_killed sub this test
can make pvestatd hang on I/O wait when a nfsd process is stopped
This might help with other file based storages (GlusterFS ?) too
---
PVE/Storage/Plugin.pm | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/PVE/Stora
This runs subroutine in a forked process
and kills it after a timeout
---
changes since V1:
* add a second pipe to catch die() causes in the passed sub
* disable pending alarms until we reaped the child
* disable signals until we reaped the child ( both to avoid Zombies)
* rename the subroutine
---
vzdump.adoc | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/vzdump.adoc b/vzdump.adoc
index 5c6804c..e56bdcb 100644
--- a/vzdump.adoc
+++ b/vzdump.adoc
@@ -125,9 +125,11 @@ NOTE: `snapshot` mode requires that all backed up volumes
are on a storage that
supports sn
Modal Windows disable the Workspace behind them, so it was not possible
to get onlineHelp when starting a oneshot backup or configuring a backup job.
---
www/manager6/dc/Backup.js | 1 +
www/manager6/window/Backup.js | 8 +++-
2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/w
This runs subroutine in a forked process
and kills it after a timeout
---
src/PVE/Tools.pm | 66
1 file changed, 66 insertions(+)
diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm
index da7da5d..fce2b1c 100644
--- a/src/PVE/Tools.pm
+++ b/sr
t/pve/nfsbsd' does not
exist or is unreachable
nfsbsdnfs 0 0 0 0 100.00%
Emmanuel Kasper (1):
Add run_fork_with_timeout utility
src/PVE/Tools.pm | 66
1 file changed, 66 in
Wrap the -d test with run_or_get_killed sub this test
can make pvestatd hang on I/O wait when a nfsd process is stopped
This might help with other file based storages, for instance
directory storages on unplugged USB devices.
---
PVE/Storage/Plugin.pm | 9 +++--
1 file changed, 7 insertions(+
This fixes the problem that a restarted container was seen as migrating
in the UI.
Background:
The check for migration happens on a 'disconnected' event, but can very well
come for other reasons, like entering reboot on the LXC console.
Since we have no idea about what have caused the disconnectio
This fixes a "Use of uninitialized value in multiplication (*) "
warning when doing a migration
---
PVE/QemuMigrate.pm | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm
index cb1639b..f995584 100644
--- a/PVE/QemuMigrate.pm
+++ b/PVE
On 07/02/2017 08:30 AM, Alexandre Derumier wrote:
> If online is checked && vm is a qemu machine,
>
> we display 2 new field:
>
> -use-local-disk checkbox (unchecked by default)
> -target storage option (disabled by default, enabled is localdisk is checked)
do you plan to update the documentat
This string is ugly and unfit for human consumption for 99% of our users.
The same information is available in a number of place in the GUI via Node
Summary, Package Versions, Subscription (Report)
---
www/manager6/Workspace.js | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/
This patch serie tries to iron out some rough aspects of the UI.
From the usability perspective the first one is the most important IMHO
Emmanuel Kasper (5):
Add a "Run task in background" button to make users feel safer
Do not display the abbreviated commit SHA-1 in the workspa
It is not oobvious when closing a Task Window, especially
one than popped up from itself ( template download, or backup), that
closing will not interrupt the action.
Instead of leaving the user in doubt, a "Run task in background button"
let the user work with the rest of the UI while the task run
This will allow reuse of the class for other inlined buttons
---
www/css/ext6-pve.css | 2 +-
www/manager6/button/HelpButton.js | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/www/css/ext6-pve.css b/www/css/ext6-pve.css
index 67bc865d..8a5f2af8 100644
--- a/www/
* group the action buttons on the left, to distinguish them from the modal
'close' button
* display the action buttons as grey instead of blue, to be consistent
with the rest of the GUI
---
www/manager6/window/Settings.js | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git
---
www/manager6/window/Migrate.js| 6 --
www/manager6/window/TaskViewer.js | 3 ++-
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/www/manager6/window/Migrate.js b/www/manager6/window/Migrate.js
index dfc8da33..17ef8f67 100644
--- a/www/manager6/window/Migrate.js
+++ b/www
This prepares for the future, in case we want to infer the
OS Type from the installation media name or content.
---
www/manager6/qemu/CreateWizard.js | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/www/manager6/qemu/CreateWizard.js
b/www/manager6/qemu/CreateWizard.j
This allow to declare all the components inside the same items
array, and separate behaviour from view better.
No functional changes.
---
www/manager6/qemu/CreateWizard.js | 71 ++-
1 file changed, 40 insertions(+), 31 deletions(-)
diff --git a/www/manager6/qe
This prepares the way for detecting the OStype from the install media,
without making the detection itself yet.
Emmanuel Kasper (3):
Use camel case for InputPanels aliases
Use a View Controller for making components react to each other
Select OS Type after selecting the Installation Media
Camel Case names for component alias follow Ext recommended practises
and are used otherwise everyelse in the code base.
No functionnal changes, aliases for these components were not used
anyway.
---
www/manager6/qemu/CDEdit.js | 2 +-
www/manager6/qemu/HDEdit.js | 2 +-
www/manager6/qe
Hi Alexandre
When migrating local disks and ide I get the following error:
drive-ide0: transferred: 52428800 bytes remaining: 0 bytes total:
52428800 bytes progression: 100.00 % busy: 0 ready: 1
all mirroring jobs are ready
drive-ide0: Completing block job...
drive-ide0: Completed successfully.
d
On 07/10/2017 11:06 AM, Alexandre DERUMIER wrote:
> Hi,
>
> I don't remember to have implemedted this device_del.
>
> I don't see why need device_del on source
> (That mean that we can't live migration disk without hotplug support)
>
>
> if nbd reference on source is a problem for vm stop,
On 07/10/2017 11:09 AM, Dominik Csapak wrote:
> On 07/06/2017 02:09 PM, Emmanuel Kasper wrote:
>> Camel Case names for component alias follow Ext recommended practises
>> and are used otherwise everyelse in the code base.
>>
>> No functionnal changes, aliases for the
This prepares for the future, in case we want to infer the
OS Type from the installation media name or content.
---
www/manager6/qemu/CreateWizard.js | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/www/manager6/qemu/CreateWizard.js
b/www/manager6/qemu/CreateWizard.j
This prepares the way for detecting the OStype from the install media,
without making the detection itself yet.
changes since V1:
* remove useless getView() call, since the ViewController can lookup
reference by itself
Emmanuel Kasper (3):
Use camel case for InputPanels aliases
Use a View
This allow to declare all the components inside the same items
array, and separate behaviour from view better.
No functional changes.
---
www/manager6/qemu/CreateWizard.js | 71 ++-
1 file changed, 40 insertions(+), 31 deletions(-)
diff --git a/www/manager6/qe
Camel Case names for component alias follow Ext recommended practises
and are used otherwise everyelse in the code base.
No functionnal changes, aliases for these components were not used
anyway.
---
www/manager6/qemu/CDEdit.js | 2 +-
www/manager6/qemu/HDEdit.js | 2 +-
www/manager6/qe
On 07/12/2017 07:12 AM, Thomas Lamprecht wrote:
> Hi,
>
> On 07/11/2017 10:06 PM, Stefan Priebe - Profihost AG wrote:
>> Hello,
>>
>> Am 11.07.2017 um 16:40 schrieb Thomas Lamprecht:
>>> commit 85909c04c49879f5fffa366fc3233eee2b157e97 switched from cirrus
>>> to vga for non windows OSs.
>>>
>>>
This was fixed in c46366fd494c7dd0cd46bf133146e6f7a98e32a4
for 'qm create', but 'qm importdisk' has the same issue
---
PVE/CLI/qm.pm | 6 ++
1 file changed, 6 insertions(+)
diff --git a/PVE/CLI/qm.pm b/PVE/CLI/qm.pm
index f6fe4a2..e192153 100755
--- a/PVE/CLI/qm.pm
+++ b/PVE/CLI/qm.pm
@@ -420
This was fixed in c46366fd494c7dd0cd46bf133146e6f7a98e32a4
for 'qm create', but 'qm importdisk' has the same issue
---
PVE/CLI/qm.pm | 6 ++
1 file changed, 6 insertions(+)
diff --git a/PVE/CLI/qm.pm b/PVE/CLI/qm.pm
index f6fe4a2..e192153 100755
--- a/PVE/CLI/qm.pm
+++ b/PVE/CLI/qm.pm
@@ -420
On packages removal (!= purge) systemd units are masked.
The postinst script has then to reenable this units at the
beginning of the script.
Our other packages are doing this manually, or automatically
when the dh_systemd_enable helpers generated a postinst,
but this was missing here.
---
debian/
On packages removal (!= purge) systemd units are masked.
The postinst script has then to reenable this units at the
beginning of the 'configure' step.
Our other packages are doing this manually, or automatically
when the dh_systemd_enable helpers generated a postinst,
but this was missing here.
--
When calling chown on a setuid/setgid executable, the
setuid/setgid bits are reset to 0. So ordering matters
here. See chown(2).
---
debian/rules | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/debian/rules b/debian/rules
index 9ce61388..dc9375ab 100755
--- a/debian/rules
+++ b
This fixes the warning:
dpkg-shlibdeps: warning: package could avoid a useless dependency if
debian/pve-manager/usr/bin/pvemailforward was not linked against libdl.so.2 (it
uses none of the library's symbols)
---
bin/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/bi
On 07/10/2017 11:19 AM, Dominik Csapak wrote:
> On 07/06/2017 02:01 PM, Emmanuel Kasper wrote:
>> This patch serie tries to iron out some rough aspects of the UI.
>> From the usability perspective the first one is the most important IMHO
>>
>> Emmanuel Kasper (
serie looks good to me
On 07/20/2017 04:16 PM, Dominik Csapak wrote:
> this series adapts the gui to the luminous json api changes
> since they moved some stuff around and removed some things
>
> this should only be applied to master, since we only support luminous
> on pve 5, for pve 4 i will se
---
pve-network.adoc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pve-network.adoc b/pve-network.adoc
index b8eb7a0..f2f5103 100644
--- a/pve-network.adoc
+++ b/pve-network.adoc
@@ -301,7 +301,7 @@ iface eno1 inet manual
iface eno2 inet manual
auto bond0
-iface bond0 ine
---
pve-network.adoc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/pve-network.adoc b/pve-network.adoc
index dd86029..b8eb7a0 100644
--- a/pve-network.adoc
+++ b/pve-network.adoc
@@ -83,8 +83,8 @@ Default Configuration using a Bridge
---
pve-network.adoc | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/pve-network.adoc b/pve-network.adoc
index 315ba63..dd86029 100644
--- a/pve-network.adoc
+++ b/pve-network.adoc
@@ -246,8 +246,13 @@ slaves in the single logical bonded interface such that
different
It can happen that the qmp connection gets lost while mirroring a disk.
In that case the current block job get cancelled, but the real cause of the
failure
is lost, becase we die() at a later step with the generic message
"die "$job: mirroring has been cancelled\n"
example:
...
drive-scsi0: trans
It can happen that the qmp connection gets lost while mirroring a disk.
In that case the current block job get cancelled, but the real cause of the
failure
is lost, becase we die() at a later step with the generic message
"die "$job: mirroring has been cancelled\n"
example:
...
drive-scsi0: trans
This should not be needed since we call 'block-job-complete' before
in qemu_drive_mirror_monitor(), and after benchmarking it does not
appear to be needed nor provide a measurable improvement when shutting
down the source.
---
PVE/QemuMigrate.pm | 5 -
1 file changed, 5 deletions(-)
diff --gi
---
PVE/QemuMigrate.pm | 2 --
1 file changed, 2 deletions(-)
diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm
index ac2884b..3169b7a 100644
--- a/PVE/QemuMigrate.pm
+++ b/PVE/QemuMigrate.pm
@@ -224,8 +224,6 @@ sub sync_disks {
# local volumes which have been copied
$self->{volumes}
Currently when qmp fails to report a query-block-jobs,
we exit with the following log:
drive-scsi0: transferred: 2950692864 bytes remaining: 3491758080 bytes total:
6442450944 bytes progression: 45.80 % busy: 1 ready: 0
drive-scsi0: transferred: 2950692864 bytes remaining: 3491758080 bytes total:
On 07/31/2017 03:00 PM, Fabian Grünbichler wrote:
> On Thu, Jul 27, 2017 at 11:25:41AM +0200, Emmanuel Kasper wrote:
>> It can happen that the qmp connection gets lost while mirroring a disk.
>> In that case the current block job get cancelled, but the real cause of the
>&g
disks from local storage migration.
On 07/27/2017 02:32 PM, Emmanuel Kasper wrote:
> This should not be needed since we call 'block-job-complete' before
> in qemu_drive_mirror_monitor(), and after benchmarking it does not
> appear to be needed nor provide a measurable improv
On 08/24/2017 08:30 AM, Fabian Grünbichler wrote:
> On Thu, Aug 24, 2017 at 07:05:35AM +0200, Thomas Lamprecht wrote:
>> On 08/23/2017 07:15 PM, Alexandre DERUMIER wrote:
>>> for me, this patch is ok.
>>>
>>> if the job is complete, we don't need to unplug. (and that mean that vm
>>> need to sup
This checkbox had not effect whatsoever:
* if the VM was online and offline was selected, the migration would fail
with the message that the --online flag is needed for running VMs
* if the the VM was offline and online was selected, the migration would
happen offline anyway
---
www/manager6/w
config
object.
Emmanuel Kasper (3):
Allow to set the onlineHelp via a config Object
Rework Migrate Window with declarative synthax
Turn the offline/online checkbox of the Migrate Windows into a
displayfield
www/manager6/button/HelpButton.js | 6 +
www/manager6/window/Migrate.js| 231
Also group all qemu vs LXC tests in a single place.
---
www/manager6/window/Migrate.js | 217 ++---
1 file changed, 116 insertions(+), 101 deletions(-)
diff --git a/www/manager6/window/Migrate.js b/www/manager6/window/Migrate.js
index 17ef8f67..1434d40c 100644
For cases when we need to set the online help block id of a modal window after
the help
button has been configured.
---
www/manager6/button/HelpButton.js | 6 ++
1 file changed, 6 insertions(+)
diff --git a/www/manager6/button/HelpButton.js
b/www/manager6/button/HelpButton.js
index 8585e4ad
makes sense, looks good to me
On 07/26/2017 03:49 PM, Thomas Lamprecht wrote:
> It isn't needed as its obvious that the table below show the
> configuration. It just takes away space.
>
> Signed-off-by: Thomas Lamprecht
> ---
> www/manager6/lxc/CreateWizard.js | 1 -
> www/manager6/qemu/Create
On 07/26/2017 03:49 PM, Thomas Lamprecht wrote:
> Add a text area to the end of the creation wizard, to allow
> adding some notes to this guest.
> We can set this through the API already on creation so no big changes
> needed.
not sure about this one, adding options is always a trade off between
f
On 07/26/2017 03:49 PM, Thomas Lamprecht wrote:
> The current size was chosen with the older, more compact, classic
> theme. The Crisp theme, now in use, has a bit more spaces and bigger
> font sizes to adapt to the increased pixel density of current
> displays.
> Increase the size a bit (~13%) but
If not the following could happen:
* user has a VM with two disks attached to a LSI controller
* adding a third disk, he notices the iothread option, and clicks it
* the scsihw type for all disks is changed now to virtio-scsi-single
* on next cold restart, STOP 0x007B INACCESSABLE_BOOT_DEVI
On 07/19/2017 03:45 PM, Dominik Csapak wrote:
> we parse and save the storage type infos in PVE.Utils.storageformats
> and can get them with PVE.Utils.getStorageFormat(type)
> so that we do not have to hardcode the values
Thank you for bringing this forward, especially the refactoring in the
new D
some coments inline
> +++ b/www/manager6/form/DiskStorageSelector.js
> @@ -0,0 +1,109 @@
> +Ext.define('PVE.form.DiskStorageSelector', {
> +extend: 'Ext.container.Container',
> +alias: 'widget.pveDiskStorageSelector',
> +
> +layout: 'fit',
> +defaults: {
> + margin: '0 0 5 0'
> +setStorageFormats: function(types) {
> + /*jslint confusion: true*/
> + // format/select are functions elsewhere
> + var me = this;
> + var type,sf;
> + me.storageformats = {};
> + for (type in types) {
> + if (types.hasOwnProperty(type)) {
> + var
patches 7 to 12 tested OK, looks good to me
On 07/19/2017 03:45 PM, Dominik Csapak wrote:
> Signed-off-by: Dominik Csapak
> ---
> www/manager6/lxc/MPEdit.js | 119
> -
> 1 file changed, 32 insertions(+), 87 deletions(-)
>
> diff --git a/www/manager6/
Patch 1 and 2 have no functional changes
Patch 3 has the option changed.
This could be useful for instance to track the actions of a user / node /
status completion.
Emmanuel Kasper (3):
Rewrite Log Panel and Task Panel with declarative synthax
White and indentation fixes
Make Cluster
This allow for instance to sort by Completion Status,
Username, Node.
It is possible to revert to the default sorting by using
the Reset Layout option in the User preferences.
---
www/manager6/dc/Log.js | 2 +-
www/manager6/dc/Tasks.js | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
d
101 - 200 of 931 matches
Mail list logo
