From: Wolfgang Bumiller <w.bumiller@proxmox.com>
To: Dominik Csapak <d.csapak@proxmox.com>
Cc: pve-devel@lists.proxmox.com
Subject: Re: [pve-devel] [PATCH container 1/2] add old config and unprivileged to check_ct_modify_config_perm
Date: Wed, 4 Aug 2021 10:45:53 +0200 [thread overview]
Message-ID: <20210804084553.nt5ajmlyr7p5jf2m@wobu-vie.proxmox.com> (raw)
In-Reply-To: <20210803122954.2641138-2-d.csapak@proxmox.com>
On Tue, Aug 03, 2021 at 02:29:51PM +0200, Dominik Csapak wrote:
> we'll need that for checking the features more granularly
>
> Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
> ---
> src/PVE/API2/LXC.pm | 6 ++++--
> src/PVE/API2/LXC/Config.pm | 9 ++++++---
> src/PVE/LXC.pm | 2 +-
> 3 files changed, 11 insertions(+), 6 deletions(-)
>
> diff --git a/src/PVE/API2/LXC.pm b/src/PVE/API2/LXC.pm
> index b929481..e16ce6c 100644
> --- a/src/PVE/API2/LXC.pm
> +++ b/src/PVE/API2/LXC.pm
> @@ -254,7 +254,7 @@ __PACKAGE__->register_method({
> my $ostemplate = extract_param($param, 'ostemplate');
> my $storage = extract_param($param, 'storage') // 'local';
>
> - PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, $pool, $param, []);
> + PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, $pool, undef, $param, [], $unprivileged);
>
> my $storage_cfg = cfs_read_file("storage.cfg");
>
> @@ -1679,7 +1679,9 @@ __PACKAGE__->register_method({
>
> die "no options specified\n" if !scalar(keys %$param);
>
> - PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $param, []);
> + my $conf = PVE::LXC::Config->load_config($vmid);
> +
> + PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $conf, $param, [], $conf->{unprivileged});
3 lines down we have a locked section where we load the config and use
it for a digest check, so if we need the config for the check already,
we should probably move the check itself down into the locked section
now.
Alternatively with this being the resize API call which shouldn't affect the config much,
we could also add a comment and explicitly *not* pass the old config,
but that's probably not as nice.
>
> my $storage_cfg = cfs_read_file("storage.cfg");
>
> diff --git a/src/PVE/API2/LXC/Config.pm b/src/PVE/API2/LXC/Config.pm
> index 73fec36..ab136c0 100644
> --- a/src/PVE/API2/LXC/Config.pm
> +++ b/src/PVE/API2/LXC/Config.pm
> @@ -135,6 +135,9 @@ __PACKAGE__->register_method({
> my $node = extract_param($param, 'node');
> my $vmid = extract_param($param, 'vmid');
>
> + my $conf = PVE::LXC::Config->load_config($vmid);
Similar issue as above, but with a lot more work happening in between.
AFAICT it's nothing too expensive though, so we could extend the locked
section here, too. Alternatively remember this digest and add a 2nd
digest check in the locked section below further down.
> + my $unprivileged = $conf->{unprivileged};
> +
> my $digest = extract_param($param, 'digest');
>
> die "no options specified\n" if !scalar(keys %$param);
> @@ -144,8 +147,8 @@ __PACKAGE__->register_method({
> my $revert_str = extract_param($param, 'revert');
> my @revert = PVE::Tools::split_list($revert_str);
>
> - PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, {}, [@delete]);
> - PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, {}, [@revert]);
> + PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $conf, {}, [@delete], $unprivileged);
> + PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $conf, {}, [@revert], $unprivileged);
>
> foreach my $opt (@revert) {
> raise_param_exc({ revert => "unknown option '$opt'" })
> @@ -166,7 +169,7 @@ __PACKAGE__->register_method({
> if grep(/^$opt$/, @revert);
> }
>
> - PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $param, []);
> + PVE::LXC::check_ct_modify_config_perm($rpcenv, $authuser, $vmid, undef, $conf, $param, [], $unprivileged);
>
> my $storage_cfg = PVE::Storage::config();
>
> diff --git a/src/PVE/LXC.pm b/src/PVE/LXC.pm
> index 139f901..32a2127 100644
> --- a/src/PVE/LXC.pm
> +++ b/src/PVE/LXC.pm
> @@ -1242,7 +1242,7 @@ sub template_create {
> }
>
> sub check_ct_modify_config_perm {
> - my ($rpcenv, $authuser, $vmid, $pool, $newconf, $delete) = @_;
> + my ($rpcenv, $authuser, $vmid, $pool, $oldconf, $newconf, $delete, $unprivileged) = @_;
>
> return 1 if $authuser eq 'root@pam';
> my $storage_cfg = PVE::Storage::config();
> --
> 2.30.2
next prev parent reply other threads:[~2021-08-04 8:45 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-03 12:29 [pve-devel] [PATCH container/manager] default nesting for unpriv containers in ui Dominik Csapak
2021-08-03 12:29 ` [pve-devel] [PATCH container 1/2] add old config and unprivileged to check_ct_modify_config_perm Dominik Csapak
2021-08-04 8:45 ` Wolfgang Bumiller [this message]
2021-08-04 8:47 ` Fabian Ebner
2021-08-04 8:49 ` Fabian Ebner
2021-08-03 12:29 ` [pve-devel] [PATCH container 2/2] allow nesting to be changed for VM.Allocate on unprivileged containers Dominik Csapak
2021-08-04 8:53 ` Wolfgang Bumiller
2021-08-04 8:57 ` Fabian Ebner
2021-08-03 12:29 ` [pve-devel] [PATCH manager 1/2] ui: lxc/Options: allow opening features window for VM.Allocate Dominik Csapak
2021-08-03 12:29 ` [pve-devel] [PATCH manager 2/2] ui: lxc/CreateWizard: add a 'nesting' checkbox and enable it by default Dominik Csapak
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210804084553.nt5ajmlyr7p5jf2m@wobu-vie.proxmox.com \
--to=w.bumiller@proxmox.com \
--cc=d.csapak@proxmox.com \
--cc=pve-devel@lists.proxmox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox