Re: [pve-devel] [PATCH] add hugepages option v2

Hi Alexandre, > vm configuration > > hugepages: (any|2|1024) the patch looks good to me, but it is quite big. So I need more time to test before we can include that in the stable codebase. I also wonder if we can split the code into a separate file, because the file

Re: [pve-devel] [PATCH docs] describe begin/end markers and .pve-ignore.

applied (+cleanups) ___ pve-devel mailing list pve-devel@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [PATCH manager 1/2] fix #1000: add backup checkbox for mountpoints

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

Re: [pve-devel] [PATCH manager 2/2] fix reset button with bind mounts

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

Re: [pve-devel] [RFC pve-common] Add user to run_command

My be we can fork and set the id. > Dietmar Maurer hat am 18. Mai 2016 um 18:11 geschrieben: > > > > I discuss with Wolfgang and we will change some things. > > Set Home Dir. > > Check if setuid and setguid worked. > > Check User. > > > > So I will send a patch V2. > > >

Re: [pve-devel] [RFC pve-common] Add user to run_command

> I discuss with Wolfgang and we will change some things. > Set Home Dir. > Check if setuid and setguid worked. > Check User. > > So I will send a patch V2. > > We can set the ID back on the end of the function. > I think this make sense. But it is not trivial. Please read: # man setuid If

Re: [pve-devel] [RFC pve-common] Add user to run_command

> I discuss with Wolfgang and we will change some things. > Set Home Dir. > Check if setuid and setguid worked. > Check User. BTW, we already have such code in PVE/Daemon.pm ___ pve-devel mailing list pve-devel@pve.proxmox.com

Re: [pve-devel] [RFC pve-common] Add user to run_command

> I also wonder why we have that afterfork option, and that > $orig_pid checks, because we do not fork at all? Ignore me. That afterfork is used in QemuServer.pm, and $orig_pid is just a safety check. I assume it is now useless, but it does not harm.

Re: [pve-devel] [PATCH manager 1/2] fix #1000: add backup checkbox for mountpoints

> excluding bind mounts what if I want to backup my bind mount? > and rootfs ___ pve-devel mailing list pve-devel@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [RFC pve-common] Add user to run_command

I also wonder why we have that afterfork option, and that $orig_pid checks, because we do not fork at all? Is that a leftover from old code? > On May 18, 2016 at 5:10 PM Wolfgang Link wrote: > > > I discuss with Wolfgang and we will change some things. > Set Home Dir. >

Re: [pve-devel] [PATCH pve-docs] Add chapter for thin provisioning

applied, with small modifications: > +Thin provisioning > +- I made this a sub-section of "Storage Types"... ___ pve-devel mailing list pve-devel@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [RFC pve-common] Add user to run_command

I discuss with Wolfgang and we will change some things. Set Home Dir. Check if setuid and setguid worked. Check User. So I will send a patch V2. We can set the ID back on the end of the function. I think this make sense. On 05/18/2016 05:00 PM, Dietmar Maurer wrote: > And when do you change

Re: [pve-devel] [RFC pve-common] Add user to run_command

And when do you change this back? > + POSIX::setuid((getpwnam($user))[2]); > + POSIX::setgid((getgrnam($user))[3]); > + > eval { > local $ENV{LC_ALL} = $lang; > ___ pve-devel mailing list pve-devel@pve.proxmox.com

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

Thanks! On Wed, May 18, 2016 at 5:50 PM, Dietmar Maurer wrote: > OK, added this manually. > >> On May 18, 2016 at 1:47 PM Igor Vlasenko wrote: >> >> >> > Added it as attachment. >> I see, attachments are prohibited. >> I'll repost it. In any case, the

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

OK, added this manually. > On May 18, 2016 at 1:47 PM Igor Vlasenko wrote: > > > > Added it as attachment. > I see, attachments are prohibited. > I'll repost it. In any case, the change is just > -my $bond_modes = > +our $bond_modes =

[pve-devel] [PATCH manager 1/2] fix #1000: add backup checkbox for mountpoints

excluding bind mounts and rootfs Signed-off-by: Dominik Csapak --- www/manager6/lxc/ResourceEdit.js | 16 1 file changed, 16 insertions(+) diff --git a/www/manager6/lxc/ResourceEdit.js b/www/manager6/lxc/ResourceEdit.js index c5d29f0..fd731ef 100644 ---

[pve-devel] [PATCH manager 2/2] fix reset button with bind mounts

if we do not disable the storageselector with bind mounts, it automatically sets the first storage which can hold containers (because of allowBlank: false) and triggers the dirty flag but clicking on reset does not delete the value there Signed-off-by: Dominik Csapak ---

[pve-devel] [PATCH pve-docs] Add chapter for thin provisioning

--- pvesm.adoc | 17 + 1 file changed, 17 insertions(+) diff --git a/pvesm.adoc b/pvesm.adoc index 36be46e..8bf043e 100644 --- a/pvesm.adoc +++ b/pvesm.adoc @@ -83,6 +83,23 @@ snapshots and clones. TIP: It is possible to use LVM on top of an iSCSI storage. That way you get a

[pve-devel] RFC Add chapter expling thin provisioning in pve manual

Hi I've notived we mention thin provisioning in the ceph rados and zfs chapters, without explaining what we exactly mean. I would also need this to explain in the qemu chapter the use of the 'option' for hard disks. ___ pve-devel mailing list

[pve-devel] [RFC pve-common] Add user to run_command

This patch allow to run a command with an alternative user. At the moment we run all commands as root. --- src/PVE/Tools.pm | 7 +++ 1 file changed, 7 insertions(+) diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm index 8c7f373..5a69daa 100644 --- a/src/PVE/Tools.pm +++ b/src/PVE/Tools.pm @@

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

> Added it as attachment. I see, attachments are prohibited. I'll repost it. In any case, the change is just -my $bond_modes = +our $bond_modes = diff --git a/src/PVE/INotify.pm b/src/PVE/INotify.pm index c34659f..1a34df1 100644 --- a/src/PVE/INotify.pm +++ b/src/PVE/INotify.pm @@ -665,7 +665,7

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

On Wed, May 18, 2016 at 1:54 PM, Dietmar Maurer wrote: > error: patch failed: src/PVE/INotify.pm:665 > error: src/PVE/INotify.pm: patch does not apply I am sorry, I did not set up git send-mail and am sending from gmail, may be it corrupted the spaces. Added it as

Re: [pve-devel] [PATCH docs] set destdir

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

Re: [pve-devel] [PATCH common] typo fix

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

[pve-devel] [PATCH docs] describe begin/end markers and .pve-ignore.

Added the description right between the list of the changes we make and the paragraph about how they are os specific which contains the bit about unmanaged containers which I'm also making a bit more explicit. --- pct.adoc | 23 +-- 1 file changed, 21 insertions(+), 2

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

I am unable to apply this patch - I get: error: patch failed: src/PVE/INotify.pm:665 error: src/PVE/INotify.pm: patch does not apply Patch failed at 0001 network interfaces code is split into syntax dependent and independent parts not sure why? > > Signed-off-by: Igor Vlasenko

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

On Wed, May 18, 2016 at 12:42 PM, Dietmar Maurer wrote: >> Let us start with the following patch below: >> It is really short, trivial, passes tests, > Again, I still see no reason to change that code. > Simply put your code into a separate sub. We can decide later > if it is

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

> Well, I was wrong to propose changes in big chunks. > I agree that the change is too big and is not trivial. > > Let us start with the following patch below: > It is really short, trivial, passes tests, Again, I still see no reason to change that code. Simply put your code into a separate

Re: [pve-devel] [PATCH manager] allow tagging of vlan 1 in gui

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

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

On Wed, May 18, 2016 at 12:09 PM, Dietmar Maurer wrote: >> In fact, systemd-networkd is also another reason to accept refactoring >> network configuration >> code to be more abstract. >> The patch I proposed is the most trivial. > Sorry, but for me this is not trivial, and I

Re: [pve-devel] [PATCH container] allow VLAN 1 tag in containers

applied (rebased) ___ pve-devel mailing list pve-devel@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] [PATCH qemu-server] allow VLAN 1 tag in qemu-kvm vms

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

Re: [pve-devel] [PATCH common] show warning when using VLAN 1 on a VLAN aware bridge

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

[pve-devel] [PATCH common] typo fix

--- src/PVE/INotify.pm | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/PVE/INotify.pm b/src/PVE/INotify.pm index c34659f..b6749d3 100644 --- a/src/PVE/INotify.pm +++ b/src/PVE/INotify.pm @@ -340,7 +340,7 @@ sub register_file { sub register_regex { my ($dir, $regex,

[pve-devel] [PATCH qemu-server] allow VLAN 1 tag in qemu-kvm vms

Signed-off-by: Dominik Csapak --- PVE/QemuServer.pm | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index 3921c09..fe19d7f 100644 --- a/PVE/QemuServer.pm +++ b/PVE/QemuServer.pm @@ -601,7 +601,7 @@ my $net_fmt = {

[pve-devel] [PATCH container] allow VLAN 1 tag in containers

Signed-off-by: Dominik Csapak --- src/PVE/LXC/Config.pm | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/PVE/LXC/Config.pm b/src/PVE/LXC/Config.pm index bf10b1c..191f413 100644 --- a/src/PVE/LXC/Config.pm +++ b/src/PVE/LXC/Config.pm @@ -483,7 +483,7

[pve-devel] [PATCH common] show warning when using VLAN 1 on a VLAN aware bridge

Signed-off-by: Dominik Csapak --- src/PVE/Network.pm | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/PVE/Network.pm b/src/PVE/Network.pm index 0a984ad..71af2b8 100644 --- a/src/PVE/Network.pm +++ b/src/PVE/Network.pm @@ -179,6 +179,8 @@ my $bridge_add_interface =

[pve-devel] [PATCH manager] allow tagging of vlan 1 in gui

Signed-off-by: Dominik Csapak --- www/manager6/form/VLanField.js | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/www/manager6/form/VLanField.js b/www/manager6/form/VLanField.js index 402427b..c699907 100644 --- a/www/manager6/form/VLanField.js +++

[pve-devel] [PATCH] reenable vlan 1

we previously disabled vlan 1, altough it worked correctly now we reenable it, but add a warning for vlan aware bridges since the default pvid of the bridge is vlan 1 ___ pve-devel mailing list pve-devel@pve.proxmox.com

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

> In fact, systemd-networkd is also another reason to accept refactoring > network configuration > code to be more abstract. > > The patch I proposed is the most trivial. Sorry, but for me this is not trivial, and I am quite unsure it it make things easier. I would keep that code separate.

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

On Wed, May 18, 2016 at 11:10 AM, Dietmar Maurer wrote: >> In short, etcnet is declarative type network configuration system, >> Where declarative is the same idea as > > It would make more sense if altlinux supports systemd-networkd instead? Well, etcnet is a bit older.

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

> In short, etcnet is declarative type network configuration system, > Where declarative is the same idea as It would make more sense if altlinux supports systemd-networkd instead? ___ pve-devel mailing list pve-devel@pve.proxmox.com

Re: [pve-devel] Maximum Bind-Mounts

On Wed, May 18, 2016 at 5:42 AM, Dietmar Maurer wrote: > Not sure, but maybe http://bindfs.org/ can help here? > Could be, but fuse is not known for its speed. I'll stick with the normal bind mounts. ___ pve-devel mailing list

Re: [pve-devel] [PATCH common] network interfaces code is split into syntax dependent and independent parts

On Wed, May 18, 2016 at 7:57 AM, Dietmar Maurer wrote: >> I am trying to port PVE to ALT Linux. It use etcnet as network >> configuration subsystem >> instead of /etc/network/interfaces. > > Where can I find reasonable documentation for that? I only found: >