[pve-devel] applied: [pmg-devel] [PATCH pmg-gui 1/1] overwrite run_editor of base class

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [pmg-devel] [PATCH manager 1/1] overwrite the built-in 'run_editor' function

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH widget-toolkit 2/2] add a checkbox to edit windows for advanced options

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH widget-toolkit 1/2] add advanced options to the input panel

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [PATCH widget-toolkit 2/2] add a checkbox to edit windows for advanced options

[Thomas Lamprecht]

Am 04/05/2018 um 04:03 PM schrieb Dominik Csapak:

if the inputpanel has advanced options, show a checkbox to
show/hide them

Signed-off-by: Dominik Csapak 
---
  window/Edit.js | 39 +++
  1 file changed, 39 insertions(+)

diff --git a/window/Edit.js b/window/Edit.js
index f72bee0..8d5aa19 100644
--- a/window/Edit.js
+++ b/window/Edit.js
@@ -274,6 +274,23 @@ Ext.define('Proxmox.window.Edit', {
var dirty = form.isDirty();
submitBtn.setDisabled(!valid || !(dirty || me.isCreate));
resetBtn.setDisabled(!dirty);
+
+   if (inputPanel && inputPanel.hasAdvanced) {
+   // we want to show the advanced options
+   // as soon as some of it is not valid
+   var advancedItems = 
me.down('#advancedContainer').query('field');
+   var valid = true;
+   advancedItems.forEach(function(field) {
+   if (!field.isValid()) {
+   valid = false;
+   }
+   });


Do you want to run isValid on all fields, as else only the first invalid
will be shown?

If so include this info in the comment, e.g., something like:
// tell user why the panel is invalid, show all invalid advanced items 
always


else use:
valid = advancedItems.every(function(f) { return f.isValid() });


+
+   if (!valid) {
+   inputPanel.setAdvancedVisible(true);
+   me.down('#advancedcb').setValue(true);
+   }
+   }
};
  
  	form.on('dirtychange', set_button_status);

@@ -297,6 +314,28 @@ Ext.define('Proxmox.window.Edit', {
me.buttons = [ submitBtn, resetBtn ];
}
  
+	if (inputPanel && inputPanel.hasAdvanced) {

+   var sp = Ext.state.Manager.getProvider();
+   var advchecked = sp.get('proxmox-advanced-cb');
+   inputPanel.setAdvancedVisible(advchecked);
+   me.buttons.unshift(
+  {
+  xtype: 'proxmoxcheckbox',
+  itemId: 'advancedcb',
+  boxLabelAlign: 'before',
+  boxLabel: gettext('Advanced'),
+  stateId: 'proxmox-advanced-cb',
+  value: advchecked,
+  listeners: {
+  change: function(cb, val) {
+  inputPanel.setAdvancedVisible(val);
+  sp.set('proxmox-advanced-cb', val);
+  }
+  }
+  }
+   );
+   }
+
var onlineHelp = me.onlineHelp;
if (!onlineHelp && inputPanel && inputPanel.onlineHelp) {
onlineHelp = inputPanel.onlineHelp;




___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [RFC qemu-server] API/create: move locking inside worker

[Thomas Lamprecht]

Move the locking inside worker, so that the process doing the actual
work (create or restore) holds the lock, and can call functions which
do locking without deadlocking.

This mirrors the behaviour we use for containers, and allows to add
an 'autostart' parameter which starts the VM after successful
creation. vm_start needs the lock and as not the worker but it's
parents held it, it couldn't know that it was actually save to
continue...

Signed-off-by: Thomas Lamprecht 
---

I discussed this with Fabian a few months ago and have something in
mind that this shouldn't be that easy, but I cannot remember what
exactly that reason was, so RFC. :)

 PVE/API2/Qemu.pm | 9 ++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index 0f27d29..8695e94 100644
--- a/PVE/API2/Qemu.pm
+++ b/PVE/API2/Qemu.pm
@@ -560,7 +560,7 @@ __PACKAGE__->register_method({
# ensure no old replication state are exists
PVE::ReplicationState::delete_guest_states($vmid);
 
-   return $rpcenv->fork_worker('qmrestore', $vmid, $authuser, 
$realcmd);
+   return PVE::QemuConfig->lock_config_full($vmid, 1, $realcmd);
};
 
my $createfn = sub {
@@ -607,10 +607,13 @@ __PACKAGE__->register_method({
PVE::AccessControl::add_vm_to_pool($vmid, $pool) if $pool;
};
 
-   return $rpcenv->fork_worker('qmcreate', $vmid, $authuser, $realcmd);
+   return PVE::QemuConfig->lock_config_full($vmid, 1, $realcmd);
};
 
-   return PVE::QemuConfig->lock_config_full($vmid, 1, $archive ? 
$restorefn : $createfn);
+   my $worker_name = $archive ? 'qmrestore' : 'qmcreate';
+   my $code = $archive ? $restorefn : $createfn;
+
+   return $rpcenv->fork_worker($worker_name, $vmid, $authuser, $code);
 }});
 
 __PACKAGE__->register_method({
-- 
2.14.2


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH manager] merge cores, cpulimit and cpuunits for containers

[Dominik Csapak]

like we do for qemu vms

Signed-off-by: Dominik Csapak 
---
 www/manager6/lxc/Resources.js | 41 +++--
 1 file changed, 23 insertions(+), 18 deletions(-)

diff --git a/www/manager6/lxc/Resources.js b/www/manager6/lxc/Resources.js
index c4ae2db8..676ad340 100644
--- a/www/manager6/lxc/Resources.js
+++ b/www/manager6/lxc/Resources.js
@@ -59,32 +59,37 @@ Ext.define('PVE.lxc.RessourceView', {
defaultValue: '',
tdCls: 'pve-itype-icon-processor',
renderer: function(value) {
-   if (value) { return value; }
-   return gettext('unlimited');
-   }
-   },
-   cpulimit: {
-   header: gettext('CPU limit'),
-   editor: caps.vms['VM.Config.CPU'] ? 'PVE.lxc.CPUEdit' : 
undefined,
-   defaultValue: 0,
-   tdCls: 'pve-itype-icon-processor',
-   renderer: function(value) {
-   if (value > 0) { return value; }
-   return gettext('unlimited');
+   var cpulimit = me.getObjectValue('cpulimit');
+   var cpuunits = me.getObjectValue('cpuunits');
+   var res;
+   if (value) {
+   res = value;
+   } else {
+   res = gettext('unlimited');
+   }
+
+   if (cpulimit) {
+   res += ' [cpulimit=' + cpulimit + ']';
+   }
+
+   if (cpuunits) {
+   res += ' [cpuunits=' + cpuunits + ']';
+   }
+   return res;
}
},
-   cpuunits: {
-   header: gettext('CPU units'),
-   editor: caps.vms['VM.Config.CPU'] ? 'PVE.lxc.CPUEdit' : 
undefined,
-   defaultValue: 1024,
-   tdCls: 'pve-itype-icon-processor'
-   },
rootfs: {
header: gettext('Root Disk'),
defaultValue: Proxmox.Utils.noneText,
editor: mpeditor,
tdCls: 'pve-itype-icon-storage'
},
+   cpulimit: {
+   visible: false
+   },
+   cpuunits: {
+   visible: false
+   },
unprivileged: {
visible: false
}
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH manager 1/1] overwrite the built-in 'run_editor' function

[Dominik Csapak]

this is needed for the 'enter' handler

Signed-off-by: Dominik Csapak 
---
 www/manager6/lxc/DNS.js   | 1 +
 www/manager6/qemu/HardwareView.js | 1 +
 2 files changed, 2 insertions(+)

diff --git a/www/manager6/lxc/DNS.js b/www/manager6/lxc/DNS.js
index 3287ae91..4eab6b11 100644
--- a/www/manager6/lxc/DNS.js
+++ b/www/manager6/lxc/DNS.js
@@ -229,6 +229,7 @@ Ext.define('PVE.lxc.DNS', {
url: "/api2/json/nodes/" + nodename + "/lxc/" + vmid + "/config",
selModel: sm,
cwidth1: 150,
+   run_editor: run_editor,
tbar: [ edit_btn ],
rows: rows,
listeners: {
diff --git a/www/manager6/qemu/HardwareView.js 
b/www/manager6/qemu/HardwareView.js
index dcdea14c..8017f84a 100644
--- a/www/manager6/qemu/HardwareView.js
+++ b/www/manager6/qemu/HardwareView.js
@@ -496,6 +496,7 @@ Ext.define('PVE.qemu.HardwareView', {
url: '/api2/json/' + 'nodes/' + nodename + '/qemu/' + vmid + 
'/pending',
interval: 5000,
selModel: sm,
+   run_editor: run_editor,
tbar: [ 
{
text: gettext('Add'),
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH widget-toolkit/manager/pmg-gui] allow pressing enter in ObjectGrids

[Dominik Csapak]

this series makes it possible to press enter to
open the editor of (pending)ObjectGrids

proxmox-widget-toolkit

Dominik Csapak (1):
  allow pressing enter in ObjectGrids to edit a field

 grid/ObjectGrid.js | 8 
 node/DNSView.js| 1 +
 node/TimeView.js   | 1 +
 3 files changed, 10 insertions(+)

pve-manager

Dominik Csapak (1):
  overwrite the built-in 'run_editor' function

 www/manager6/lxc/DNS.js   | 1 +
 www/manager6/qemu/HardwareView.js | 1 +
 2 files changed, 2 insertions(+)

pmg-gui

Dominik Csapak (1):
  overwrite run_editor of base class

 js/ActionList.js   | 1 +
 js/MyNetworks.js   | 1 +
 js/ObjectGroup.js  | 1 +
 js/RelayDomains.js | 1 +
 js/Transport.js| 1 +
 5 files changed, 5 insertions(+)

-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH widget-toolkit 1/1] allow pressing enter in ObjectGrids to edit a field

[Dominik Csapak]

for this we need to overwrite the 'run_editor' function
of the ObjectGrid if we use a custom one

Signed-off-by: Dominik Csapak 
---
 grid/ObjectGrid.js | 8 
 node/DNSView.js| 1 +
 node/TimeView.js   | 1 +
 3 files changed, 10 insertions(+)

diff --git a/grid/ObjectGrid.js b/grid/ObjectGrid.js
index bd294c8..68937ce 100644
--- a/grid/ObjectGrid.js
+++ b/grid/ObjectGrid.js
@@ -224,6 +224,14 @@ Ext.define('Proxmox.grid.ObjectGrid', {
return value;
 },
 
+listeners: {
+   itemkeyup: function(view, record, item, index, e) {
+   if (e.getKey() === e.ENTER) {
+   this.run_editor();
+   }
+   }
+},
+
 initComponent : function() {
var me = this;
 
diff --git a/node/DNSView.js b/node/DNSView.js
index 2df2dac..b0f0973 100644
--- a/node/DNSView.js
+++ b/node/DNSView.js
@@ -20,6 +20,7 @@ Ext.define('Proxmox.node.DNSView', {
url: "/api2/json/nodes/" + me.nodename + "/dns",
cwidth1: 130,
interval: 1000,
+   run_editor: run_editor,
rows: {
search: { header: 'Search domain', required: true },
dns1: { header: gettext('DNS server') + " 1", required: true },
diff --git a/node/TimeView.js b/node/TimeView.js
index 0cf68eb..27de02d 100644
--- a/node/TimeView.js
+++ b/node/TimeView.js
@@ -26,6 +26,7 @@ Ext.define('Proxmox.node.TimeView', {
url: "/api2/json/nodes/" + me.nodename + "/time",
cwidth1: 150,
interval: 1000,
+   run_editor: run_editor,
rows: {
timezone: { 
header: gettext('Time zone'), 
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH pmg-gui 1/1] overwrite run_editor of base class

[Dominik Csapak]

so that every call lands in the custom run_editor

Signed-off-by: Dominik Csapak 
---
 js/ActionList.js   | 1 +
 js/MyNetworks.js   | 1 +
 js/ObjectGroup.js  | 1 +
 js/RelayDomains.js | 1 +
 js/Transport.js| 1 +
 5 files changed, 5 insertions(+)

diff --git a/js/ActionList.js b/js/ActionList.js
index 213afb2..af5e2e3 100644
--- a/js/ActionList.js
+++ b/js/ActionList.js
@@ -121,6 +121,7 @@ Ext.define('PMG.ActionList', {
}
 
Ext.apply(me, {
+   run_editor: run_editor,
columns: [
{
header: gettext('Name'),
diff --git a/js/MyNetworks.js b/js/MyNetworks.js
index 955b3fa..6ef0022 100644
--- a/js/MyNetworks.js
+++ b/js/MyNetworks.js
@@ -114,6 +114,7 @@ Ext.define('PMG.MyNetworks', {
Ext.apply(me, {
store: store,
tbar: tbar,
+   run_editor: run_editor,
viewConfig: {
trackOver: false
},
diff --git a/js/ObjectGroup.js b/js/ObjectGroup.js
index 1e76ef3..0de4da3 100644
--- a/js/ObjectGroup.js
+++ b/js/ObjectGroup.js
@@ -238,6 +238,7 @@ Ext.define('PMG.ObjectGroup', {
Proxmox.Utils.monStoreErrors(me, me.store, true);
 
Ext.apply(me, {
+   run_editor: run_editor,
listeners: {
itemdblclick: run_editor,
activate: reload
diff --git a/js/RelayDomains.js b/js/RelayDomains.js
index 8945b2d..26560c1 100644
--- a/js/RelayDomains.js
+++ b/js/RelayDomains.js
@@ -114,6 +114,7 @@ Ext.define('PMG.RelayDomains', {
Ext.apply(me, {
store: store,
tbar: tbar,
+   run_editor: run_editor,
viewConfig: {
trackOver: false
},
diff --git a/js/Transport.js b/js/Transport.js
index b5e20d8..f125835 100644
--- a/js/Transport.js
+++ b/js/Transport.js
@@ -133,6 +133,7 @@ Ext.define('PMG.Transport', {
Ext.apply(me, {
store: store,
tbar: tbar,
+   run_editor: run_editor,
viewConfig: {
trackOver: false
},
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH widget-toolkit] better default focus selection

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [RFC qemu-server] API/create: move locking inside worker

[Fabian Grünbichler]

On Fri, Apr 06, 2018 at 11:54:03AM +0200, Thomas Lamprecht wrote:
> Move the locking inside worker, so that the process doing the actual
> work (create or restore) holds the lock, and can call functions which
> do locking without deadlocking.
> 
> This mirrors the behaviour we use for containers, and allows to add
> an 'autostart' parameter which starts the VM after successful
> creation. vm_start needs the lock and as not the worker but it's
> parents held it, it couldn't know that it was actually save to
> continue...
> 
> Signed-off-by: Thomas Lamprecht 
> ---
> 
> I discussed this with Fabian a few months ago and have something in
> mind that this shouldn't be that easy, but I cannot remember what
> exactly that reason was, so RFC. :)

there is one issue - if somebody holds the flock and you only realize it
after you have forked, you did a fork for nothing (and instead of a
rather fast "timeout" error message, you have to check the task log.
this is not nice from a usability perspective, although it should not
cause problems from a technical/lockdep one ;)

the clean solution is

flock {
  set_config_lock
}

fork {
  do stuff
  flock {
(re)read config
check_config_lock
modify/write config
  }
  do some more stuff
  flock {
(re)read config
check_config_lock
remove_config_lock
final modify/write config
  }
}

of course, this is much more involved and harder to get all corner cases
right ;) I would like to move all the create/restore/clone API paths to
this flow in the long run, but I am not opposed to switching the
fork/lock order in places where we need the flock sooner/now.

> 
>  PVE/API2/Qemu.pm | 9 ++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
> index 0f27d29..8695e94 100644
> --- a/PVE/API2/Qemu.pm
> +++ b/PVE/API2/Qemu.pm
> @@ -560,7 +560,7 @@ __PACKAGE__->register_method({
>   # ensure no old replication state are exists
>   PVE::ReplicationState::delete_guest_states($vmid);
>  
> - return $rpcenv->fork_worker('qmrestore', $vmid, $authuser, 
> $realcmd);
> + return PVE::QemuConfig->lock_config_full($vmid, 1, $realcmd);
>   };
>  
>   my $createfn = sub {
> @@ -607,10 +607,13 @@ __PACKAGE__->register_method({
>   PVE::AccessControl::add_vm_to_pool($vmid, $pool) if $pool;
>   };
>  
> - return $rpcenv->fork_worker('qmcreate', $vmid, $authuser, $realcmd);
> + return PVE::QemuConfig->lock_config_full($vmid, 1, $realcmd);
>   };
>  
> - return PVE::QemuConfig->lock_config_full($vmid, 1, $archive ? 
> $restorefn : $createfn);
> + my $worker_name = $archive ? 'qmrestore' : 'qmcreate';
> + my $code = $archive ? $restorefn : $createfn;
> +
> + return $rpcenv->fork_worker($worker_name, $vmid, $authuser, $code);
>  }});
>  
>  __PACKAGE__->register_method({
> -- 
> 2.14.2
> 
> 
> ___
> pve-devel mailing list
> pve-devel@pve.proxmox.com
> https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH manager] merge cores, cpulimit and cpuunits for containers

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH widget-toolkit 1/1] allow pressing enter in ObjectGrids to edit a field

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH manager] better focus of lxc/MPEdit

[Dominik Csapak]

this gives the focus to either the 'path' field
or to the 'X' tool (if it is a rootfs)

Signed-off-by: Dominik Csapak 
---
 www/manager6/lxc/MPEdit.js | 1 +
 1 file changed, 1 insertion(+)

diff --git a/www/manager6/lxc/MPEdit.js b/www/manager6/lxc/MPEdit.js
index 37f725ea..cf8d6f02 100644
--- a/www/manager6/lxc/MPEdit.js
+++ b/www/manager6/lxc/MPEdit.js
@@ -324,6 +324,7 @@ Ext.define('PVE.lxc.MountPointEdit', {
 
Ext.apply(me, {
subject: subject,
+   defaultFocus: me.confid !== 'rootfs' ? 'textfield[name=mp]' : 
'tool',
items: ipanel
});
 
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH widget-toolkit] better default focus selection

[Dominik Csapak]

we do not want to focus on hidden/disabled fields,
because that focus gets lost and prevents some things
e.g. cancelling with ESC

Signed-off-by: Dominik Csapak 
---
 window/Edit.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/window/Edit.js b/window/Edit.js
index 8d5aa19..9548e21 100644
--- a/window/Edit.js
+++ b/window/Edit.js
@@ -34,7 +34,7 @@ Ext.define('Proxmox.window.Edit', {
 defaultButton: 'submitbutton',
 
 // finds the first form field
-defaultFocus: 'field',
+defaultFocus: 'field[disabled=false][hidden=false]',
 
 showProgress: false,
 
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [RFC qemu-server] API/create: move locking inside worker

[Thomas Lamprecht]

Am 04/06/2018 um 12:28 PM schrieb Fabian Grünbichler:

On Fri, Apr 06, 2018 at 11:54:03AM +0200, Thomas Lamprecht wrote:

Move the locking inside worker, so that the process doing the actual
work (create or restore) holds the lock, and can call functions which
do locking without deadlocking.

This mirrors the behaviour we use for containers, and allows to add
an 'autostart' parameter which starts the VM after successful
creation. vm_start needs the lock and as not the worker but it's
parents held it, it couldn't know that it was actually save to
continue...

Signed-off-by: Thomas Lamprecht 
---

I discussed this with Fabian a few months ago and have something in
mind that this shouldn't be that easy, but I cannot remember what
exactly that reason was, so RFC. :)


there is one issue - if somebody holds the flock and you only realize it
after you have forked, you did a fork for nothing (and instead of a
rather fast "timeout" error message, you have to check the task log.
this is not nice from a usability perspective, although it should not
cause problems from a technical/lockdep one ;)



Ah, yeah, I could add a simple check if the VM is already locked before
forking the worker, I mean that's obviously racy but should catch most
cases, and we do that elsewhere too, AFAIK.

I'll try to add it and just send it out, if it's accetable to you then
we could apply it as a temporary improvement :)


the clean solution is

flock {
   set_config_lock
}

fork {
   do stuff
   flock {
 (re)read config
 check_config_lock
 modify/write config
   }
   do some more stuff
   flock {
 (re)read config
 check_config_lock
 remove_config_lock
 final modify/write config
   }
}

of course, this is much more involved and harder to get all corner cases
right ;) I would like to move all the create/restore/clone API paths to
this flow in the long run, but I am not opposed to switching the
fork/lock order in places where we need the flock sooner/now.



Ah yes I remember now, thanks!
I'll try to give that a look in one and a half week, after my vacation.

cheers,
Thomas

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [RFC qemu-server] API/create: move locking inside worker

[Thomas Lamprecht]

Am 04/06/2018 um 01:24 PM schrieb Thomas Lamprecht:

Am 04/06/2018 um 12:28 PM schrieb Fabian Grünbichler:

On Fri, Apr 06, 2018 at 11:54:03AM +0200, Thomas Lamprecht wrote:

Move the locking inside worker, so that the process doing the actual
work (create or restore) holds the lock, and can call functions which
do locking without deadlocking.

This mirrors the behaviour we use for containers, and allows to add
an 'autostart' parameter which starts the VM after successful
creation. vm_start needs the lock and as not the worker but it's
parents held it, it couldn't know that it was actually save to
continue...

Signed-off-by: Thomas Lamprecht 
---

I discussed this with Fabian a few months ago and have something in
mind that this shouldn't be that easy, but I cannot remember what
exactly that reason was, so RFC. :)


there is one issue - if somebody holds the flock and you only realize it
after you have forked, you did a fork for nothing (and instead of a
rather fast "timeout" error message, you have to check the task log.
this is not nice from a usability perspective, although it should not
cause problems from a technical/lockdep one ;)



Ah, yeah, I could add a simple check if the VM is already locked before
forking the worker, I mean that's obviously racy but should catch most
cases, and we do that elsewhere too, AFAIK.

I'll try to add it and just send it out, if it's accetable to you then
we could apply it as a temporary improvement :)




OK, scratch it, without calling flock myself (here or in a new method) I 
cannot do a timeout-less 'check_lock', AFAIS, so better just do the

right thing (after vacantion) ;)

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH manager] better focus of lxc/MPEdit

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] [PATCH manager] fix editor and set_button_status for cloudinit

[Dominik Csapak]

cloudinit images do not have an editor so return here
and value has to be a string to match
also set the remove button text correctly when selecting a cloudinit disk

Signed-off-by: Dominik Csapak 
---
 www/manager6/qemu/HardwareView.js | 8 +---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/www/manager6/qemu/HardwareView.js 
b/www/manager6/qemu/HardwareView.js
index e726cb70..b597e172 100644
--- a/www/manager6/qemu/HardwareView.js
+++ b/www/manager6/qemu/HardwareView.js
@@ -248,7 +248,9 @@ Ext.define('PVE.qemu.HardwareView', {
var editor = rowdef.editor;
if (rowdef.tdCls == 'pve-itype-icon-storage') {
var value = me.getObjectValue(rec.data.key, '', true); 
-   if (value.match(/media=cdrom/)) {
+   if (value.match(/vm-.*-cloudinit/)) {
+   return;
+   } else if (value.match(/media=cdrom/)) {
editor = 'PVE.qemu.CDEdit';
}
}
@@ -518,12 +520,12 @@ Ext.define('PVE.qemu.HardwareView', {
rowdef.tdCls == 'pve-itype-icon-storage' &&
(value && !value.match(/media=cdrom/));
 
-   var isCloudInit = (value && value.match(/vm-.*-cloudinit/));
+   var isCloudInit = (value && 
value.toString().match(/vm-.*-cloudinit/));
 
var isEfi = (key === 'efidisk0');
 
remove_btn.setDisabled(rec.data['delete'] || (rowdef.never_delete 
=== true));
-   remove_btn.setText(isUsedDisk ? remove_btn.altText : 
remove_btn.defaultText);
+   remove_btn.setText((isUsedDisk && !isCloudInit) ? 
remove_btn.altText : remove_btn.defaultText);
 
edit_btn.setDisabled(rec.data['delete'] || !rowdef.editor || 
isCloudInit);
 
-- 
2.11.0


___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH docs] Add documentation for CIFS Storage Plugin.

[Thomas Lamprecht]

thanks, applied, with cleanup followup

Am 04/05/2018 um 02:08 PM schrieb Wolfgang Link:

---
  pve-intro.adoc|  3 +-
  pve-storage-cifs.adoc | 99 +++
  pvesm.adoc|  5 +++
  qm.adoc   |  2 +-
  vzdump.adoc   |  2 +-
  5 files changed, 108 insertions(+), 3 deletions(-)
  create mode 100644 pve-storage-cifs.adoc

diff --git a/pve-intro.adoc b/pve-intro.adoc
index 1188e77..f0b0d1e 100644
--- a/pve-intro.adoc
+++ b/pve-intro.adoc
@@ -106,6 +106,7 @@ We currently support the following Network storage types:
  * LVM Group (network backing with iSCSI targets)
  * iSCSI target
  * NFS Share
+* CIFS Share
  * Ceph RBD
  * Directly use iSCSI LUNs
  * GlusterFS
@@ -125,7 +126,7 @@ running Containers and KVM guests. It basically creates an 
archive of
  the VM or CT data which includes the VM/CT configuration files.
  
  KVM live backup works for all storage types including VM images on

-NFS, iSCSI LUN, Ceph RBD or Sheepdog. The new backup format is
+NFS, CIFS, iSCSI LUN, Ceph RBD or Sheepdog. The new backup format is
  optimized for storing VM backups fast and effective (sparse files, out
  of order data, minimized I/O).
  
diff --git a/pve-storage-cifs.adoc b/pve-storage-cifs.adoc

new file mode 100644
index 000..696d809
--- /dev/null
+++ b/pve-storage-cifs.adoc
@@ -0,0 +1,99 @@
+CIFS Backend
+---
+ifdef::wiki[]
+:pve-toplevel:
+:title: Storage: CIFS
+endif::wiki[]
+
+Storage pool type: `cifs`
+
+The CIFS backend is based on the directory backend, so it shares most
+properties. The directory layout and the file naming conventions are
+the same. The main advantage is that you can directly configure the
+CIFS server, so the backend can mount the share automatically in
+the hole cluster. There is no need to modify `/etc/fstab`. The backend
+can also test if the server is online, and provides a method to query
+the server for exported shares.
+
+Configuration
+~
+
+The backend supports all common storage properties, except the shared
+flag, which is always set. Additionally, the following properties are
+used to configure the CIFS server:
+
+server::
+
+Server IP or DNS name. To avoid DNS lookup delays, it is usually
+preferable to use an IP address instead of a DNS name - unless you
+have a very reliable DNS server, or list the server in the local
+`/etc/hosts` file.
+
+share::
+
+CIFS share (as listed by `pvesm cifsscan`).
+
+Optional properties:
+
+username::
+
+If not presents, "guest" is used.
+
+password::
+
+The user password.
+It will be saved in a private directory (/etc/pve/priv/.cred).
+
+domain::
+
+sets the domain (workgroup) of the user
+
+smbversion::
+
+SMB protocol Version (default is `3`).
+SMB1 is not supported due to security issues.
+
+path::
+
+The local mount point (defaults to `/mnt/pve//`).
+
+.Configuration Example (`/etc/pve/storage.cfg`)
+
+cifs: backup
+   path /mnt/pve/backup
+   server 10.0.0.11
+   share VMData
+   content backup
+   username anna
+   smbversion 3
+
+
+
+Storage Features
+
+
+CIFS does not support snapshots, but the backend uses `qcow2` features
+to implement snapshots and cloning.
+
+.Storage features for backend `nfs`
+[width="100%",cols="m,m,3*d",options="header"]
+|==
+|Content types |Image formats |Shared |Snapshots 
|Clones
+|images rootdir vztmpl iso backup  |raw qcow2 vmdk subvol |yes|qcow2 
|qcow2
+|==
+
+Examples
+
+
+You can get a list of exported CIFS shares with:
+
+ # pvesm cifsscan  [--username ] [--password]
+
+ifdef::wiki[]
+
+See Also
+
+
+* link:/wiki/Storage[Storage]
+
+endif::wiki[]
diff --git a/pvesm.adoc b/pvesm.adoc
index 62d190e..1d55d59 100644
--- a/pvesm.adoc
+++ b/pvesm.adoc
@@ -71,6 +71,7 @@ snapshots and clones.
  |ZFS (local)|zfspool |file  |no|yes  |yes
  |Directory  |dir |file  |no|no^1^|yes
  |NFS|nfs |file  |yes   |no^1^|yes
+|CIFS   |cifs|file  |yes   |no^1^|yes
  |GlusterFS  |glusterfs   |file  |yes   |no^1^|yes
  |LVM|lvm |block |no^2^ |no   |yes
  |LVM-thin   |lvmthin |block |no|yes  |yes
@@ -370,6 +371,8 @@ See Also
  
  * link:/wiki/Storage:_NFS[Storage: NFS]
  
+* link:/wiki/Storage:_CIFS[Storage: CIFS]

+
  * link:/wiki/Storage:_RBD[Storage: RBD]
  
  * link:/wiki/Storage:_ZFS[Storage: ZFS]

@@ -386,6 +389,8 @@ include::pve-storage-dir.adoc[]
  
  include::pve-storage-nfs.adoc[]
  
+include::pve-storage-cifs.adoc[]

+
  include::pve-storage-glusterfs.adoc[]
  
  include::pve-storage-zfspool.adoc[]

diff --git a/qm.adoc b/qm.adoc
index 154c5c1..5fba463 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -163,7 +163,7 @@ On each controller you attach a number 

[pve-devel] applied: [PATCH manager 01/11] add advanced checkbox to the wizard

[Dietmar Maurer]

applied all (11) patches

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH manager] fix editor and set_button_status for cloudinit

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH docs 1/2] fix wording for new memory dialog for qemu

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

[pve-devel] applied: [PATCH docs 2/2] mention that NAT mode is not available on the WebUI

[Dietmar Maurer]

applied

___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel