Signed-off-by: Mira Limbeck
---
v2:
- add `api: ` prefix to commit msg
PVE/API2/APT.pm | 1 +
1 file changed, 1 insertion(+)
diff --git a/PVE/API2/APT.pm b/PVE/API2/APT.pm
index 19f0baca0..4095e790f 100644
--- a/PVE/API2/APT.pm
+++ b/PVE/API2/APT.pm
@@ -774,6 +774,7 @@ __PACKAGE__
On 4/24/24 13:20, Mira Limbeck wrote:
> Signed-off-by: Mira Limbeck
> ---
> PVE/API2/APT.pm | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/PVE/API2/APT.pm b/PVE/API2/APT.pm
> index 19f0baca0..4095e790f 100644
> --- a/PVE/API2/APT.pm
> +++ b/PVE/API
Signed-off-by: Mira Limbeck
---
PVE/API2/APT.pm | 1 +
1 file changed, 1 insertion(+)
diff --git a/PVE/API2/APT.pm b/PVE/API2/APT.pm
index 19f0baca0..4095e790f 100644
--- a/PVE/API2/APT.pm
+++ b/PVE/API2/APT.pm
@@ -774,6 +774,7 @@ __PACKAGE__->register_method({
libpve-network-p
On 4/18/24 17:45, Alexander Zeidler wrote:
> On Thu, 2024-04-18 at 12:43 +0200, Mira Limbeck wrote:
>> On 4/18/24 11:16, Alexander Zeidler wrote:
>>> Successful boots which crashed somehow and sometime afterwards, will
>>> show the same "until" value (&quo
` and `pvesh get /cluster/resources` next to each other, and `lscpu`
to the hardware section where we have all the `dmidecode` output, and
`lspci`?
Except for the `dmidecode -t16,17` command needing a space for
consistency, I'm fine with any of the addition
On 4/18/24 11:16, Alexander Zeidler wrote:
> Successful boots which crashed somehow and sometime afterwards, will
> show the same "until" value ("still running" or timestamp) as the next
> following boot(s). The most recent boot from such a sequence of
> duplicated "until" lines, has not been
On 4/18/24 11:16, Alexander Zeidler wrote:
> While using keywords (-t bios,...) would be possible, depending on the
> server it also bloats the report with uninteresting information,
> hiding the relevant.
>
> Therefore the non-grouped, specific number types are used. Where we
> only need
On 4/18/24 11:16, Alexander Zeidler wrote:
> * to see if a RAM upgrade is slot/capacity-wise possible
> * to spot added/replaced RAM that may now be causing issues
>
> Maximum Capacity: 2 TB
> Size: 16 GB Part Number: 18ASF2G72PZ-2G6D1
> Size: 16 GB Part Number:
disp.setValue("Terms of service not fetched yet");
> + field.setValue(undefined);
> + checkbox.setValue(undefined);
> + checkbox.setHidden(true);
> +},
> +
> });
>
> Ext.define('PVE.node.ACMEAccountView', {
Tested this patch series in a Debian 12 container with Pebble 2.4
installed via the Debian Bookworm repositories.
The behavior of the combobox changed with this patch. The combobox now
shows an `x` to clear it, but doesn't actually clear the selection.
Instead it always sets it to `Custom`.
Account creation worked like a charm using a custom directory and EAB
(with the 2nd patch).
Other than the combobox issue above, consider this and the 2nd patch:
Tested-by: Mira Limbeck
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
also log stderr of remote command with
insecure storage migration")
Signed-off-by: Mira Limbeck
---
v3:
- added log prefix for remote error logs
- fixed style issues
v2:
- incorporated Fiona's suggestions
- added `Fixes: ...` to commit message
- kept old ip/port matching including
ecure storage migration")
Signed-off-by: Mira Limbeck
---
v2:
- incorporated Fiona's suggestions
- added `Fixes: ...` to commit message
- kept old ip/port matching including # untaint comments
- added logging for all messages in STDERR
- simplified branches
src/PVE/Stora
in a migration or
replication failing.
the bare open3 call is replaced by the run_command wrapper from
pve-common to use a safe wrapper around open3 with the same
functionality.
STDERR is now read separately from STDOUT and the last line of STDERR
is kept in case of errors.
Signed-off-by: Mira Limbeck
;$dev_sysfs_dir/remove", "1") if -e
> $dev_sysfs_dir;
> + if (-e $dev_sysfs_dir) {
> + warn "cleaning up mediated device $uuid\n";
> + PVE::SysFSTools::file_write("$dev_sysfs_dir/remove", "1");
> +
el the query without resorting to
> the queryDelay hack.
>
> Reported-By: Mira Limbeck
> Signed-off-by: Stefan Hanreich
> ---
> www/manager6/form/IPRefSelector.js | 13 +++--
> 1 file changed, 3 insertions(+), 10 deletions(-)
>
> diff --git a/www/manager6/form/
}
> },
> },
lgtm, consider this:
Reviewed-by: Mira Limbeck
Tested-by: Mira Limbeck
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
On 1/15/24 17:28, Stefan Hanreich wrote:
> Fallback to v.ref when we do not use an alias or ipset in order to be
> able to use normal CIDRs as source / destination address again
>
> Signed-off-by: Stefan Hanreich
> ---
> www/manager6/form/IPRefSelector.js | 4 +++-
> 1 file changed, 3
On 12/19/23 18:09, Aaron Lauterer wrote:
> The commit message doesn't explain the actual issue that it is trying to
> solve.
> AFAICT we do not need the ceph.conf symlinked right away for normal PVE
> operations. If it is not present in /etc/ceph/ceph.conf, the RBD and
> CephFS connections will
The original patch removed the first branch of `if` condition, leaving the
`else` in which matched the `if` above.
The correct condition for this `else` is any bond mode that is not `lacp`
(4).
Signed-off-by: Mira Limbeck
---
...bond-remove-bond-min-links-0-warning.patch | 19
Looks like a python-cryptography maintainer opened a pull request on
Github with the exact same change:
https://github.com/proxmox/pve-common/pull/8
___
pve-devel mailing list
pve-devel@lists.proxmox.com
:
Version: Unknown (2)
```
Signed-off-by: Mira Limbeck
---
I wasn't able to create a test setup where I could test this yet, will
try again on monday.
Stoiko tested it on his setup with Let's Encrypt Staging and it worked
fine. Although he didn't extract the CSR to verify it.
A customer reported
On 12/16/22 11:31, Thomas Lamprecht wrote:
On 15/12/2022 17:57, Mira Limbeck wrote:
Since some users keep their passwords in the VM/CT configs as comments
and those are most of the time unnecessary when looking through the
report, filter those.
I'd rather have this optional, opt-out is fine
the filtering, a new sub 'file2text' is introduced that
can filter the file contents if required.
This sub replaces the 'cat ...' commands.
Signed-off-by: Mira Limbeck
---
I did not add print to STDERR in file2text for now since it got quite
chatty.
If this is wanted, I'll send a v2 adding it. But since
If the config doesn't contain the cloud-init disk anymore after the
rollback, we have to clean it up since otherwise no further disk can be
attached unless the one still existing on the storage is deleted.
Signed-off-by: Mira Limbeck
Reviewed-by: Stefan Hanreich
Tested-by: Stefan Hanreich
On 11/11/22 16:18, Stefan Hanreich wrote:
Test Setup:
I created a new VM, without any Cloud-Init drive, and immediately
created a snapshot. Then I setup a Cloud-Init drive according to the
PVE documentation via CLI. I created another snapshot of this state
with a Cloud-Init drive.
What I
On 11/11/22 09:36, DERUMIER, Alexandre wrote:
Le mercredi 09 novembre 2022 à 15:19 +0100, Mira Limbeck a écrit :
Why not add the bridge in the pve-bridge script as well? This way
there
would be no need for patch 2 for qemu-server since we always add the
MAC
address to the FDB whenever the tap
On 8/24/22 18:26, Alexandre Derumier wrote:
This disabling mac learning && unicast flood for the tap interface
for vmstart, we don't add mac directly to fdb.
We set it latter if it's a migration or a fresh start.
for nic hotplug, we directly add mac to fdb
Signed-off-by: Alexandre Derumier
On 8/24/22 18:26, Alexandre Derumier wrote:
at the end of a live migration, we need to remove old mac entries
on source host (vm is not yet stopped), before resume vm on target host
Signed-off-by: Alexandre Derumier
---
PVE/QemuMigrate.pm| 1 +
PVE/QemuServer.pm
If the config doesn't contain the cloud-init disk anymore after the
rollback, we have to clean it up since otherwise no further disk can be
attached unless the one still existing on the storage is deleted.
Signed-off-by: Mira Limbeck
---
v2:
- chose the add_unused_volume way as @fiona
On 9/30/22 10:21, Fiona Ebner wrote:
Am 29.09.22 um 15:36 schrieb Mira Limbeck:
If the config doesn't contain the cloud-init disk anymore after the
rollback, we have to clean it up since otherwise no further disk can be
attached unless the one still existing on the storage is deleted.
Signed
If the config doesn't contain the cloud-init disk anymore after the
rollback, we have to clean it up since otherwise no further disk can be
attached unless the one still existing on the storage is deleted.
Signed-off-by: Mira Limbeck
---
PVE/QemuConfig.pm | 34
When a disk exists but is not referenced in the config, it will be
reused instead of dying during the attempt to allocate the disk.
Signed-off-by: Mira Limbeck
---
This patch is not required to fix the rollback code, but might be nice
to have in addition since there will still be some users
MAC filter was enabled as long as the VM firewall config existed even
with an invalid config.
With this change the check now matches the one for CTs.
Signed-off-by: Mira Limbeck
---
src/PVE/Firewall.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/PVE/Firewall.pm b
85 ("coroutine-sleep: allow qemu_co_sleep_wake that wakes nothing")
29a6ea24eb ("coroutine-sleep: replace QemuCoSleepState pointer with struct in the
API")
[0]: https://forum.proxmox.com/threads/112130/
Signed-off-by: Fiona Ebner
---
Tested-by: Mira Limbeck
Found a str
Signed-off-by: Mira Limbeck
---
local-zfs.adoc | 6 ++
1 file changed, 6 insertions(+)
diff --git a/local-zfs.adoc b/local-zfs.adoc
index 2dc25fd..1f4a4ed 100644
--- a/local-zfs.adoc
+++ b/local-zfs.adoc
@@ -547,6 +547,12 @@ improve performance when sufficient memory exists in a
system
On 5/16/22 13:04, DERUMIER, Alexandre wrote:
Le vendredi 06 mai 2022 à 12:11 +0200, Fabian Ebner a écrit :
This will break possibly existing workflows like
1. add second cloud-init
2. remove first cloud-init
to change the cloud-init storage.
Also, currently, in the gui, we can't add 2
On 5/16/22 10:32, DERUMIER, Alexandre wrote:
---
Are there any scenarios where having multiple cloud-init drives is
useful?
I don't remember exactly how cloud-init daemon is mounting drives,
but I'm pretty sure that with multiple cloud-init drives,
only 1 will be mounted and read.
So,I'm 100%
On 6/17/22 10:42, Wolfgang Bumiller wrote:
On Wed, Jun 15, 2022 at 04:09:50PM +0200, Mira Limbeck wrote:
Azure AD seems to have a variable authorization code size, depending on
the browser state according to one report in bug #4074 [0].
Sometimes the size is greater than our current limit
on
current experience, a size limit of 2048 might be enough for every
current OpenID Connect provider.
[0] https://bugzilla.proxmox.com/show_bug.cgi?id=4074
[1] https://datatracker.ietf.org/doc/html/rfc6749#section-4.1.2
Signed-off-by: Mira Limbeck
---
The PBS implementation doesn't seem to be using
Signed-off-by: Mira Limbeck
---
upstream pull request:
https://github.com/algesten/ureq/pull/495
.../patches/add-https-proxy-support.patch | 231 ++
src/ureq/debian/patches/series| 1 +
2 files changed, 232 insertions(+)
create mode 100644 src/ureq/debian
is the case when using `send`. See
https://docs.rs/ureq/2.4.0/ureq/index.html#content-length-and-transfer-encoding
See https://forum.proxmox.com/threads/openid-401-with-azure-ad.105892/
for the issue.
Signed-off-by: Mira Limbeck
---
v3:
- fixed link in inline comment
v2:
- added inline
On 4/1/22 11:32, Mira Limbeck wrote:
Azure doesn't accept `Transfer-Encoding: chunked` on their token endpoint,
but with the switch to ureq we always send requests with this set.
Fix by switching to `Content-Length` in the header instead. ureq only
sets `Transfer-Encoding: chunked` when
is the case when using `send`. See
https://docs.rs/ureq/2.4.0/ureq/index.html#content-length-and-transfer-encoding
See https://forum.proxmox.com/threads/openid-401-with-azure-ad.105892/
for the issue.
Signed-off-by: Mira Limbeck
---
v2:
- added inline comment
- removed the unnecessary `let
On 4/1/22 06:21, Dietmar Maurer wrote:
let response = if let Method::POST = request.method {
-req.send(&*request.body)
+let bytes = request.body.as_slice();
+req.send_bytes(bytes)
Does this have the side effect of changing the transfer encoding? If so, it is
worth
is the case when using `send`. See
https://github.com/algesten/ureq/blob/main/README.md#content-length-and-transfer-encoding
See https://forum.proxmox.com/threads/openid-401-with-azure-ad.105892/
for the issue.
Signed-off-by: Mira Limbeck
---
probably also fixes https://bugzilla.proxmox.com
ureq has support for a HTTP proxy, but no support for HTTPS proxy yet.
ureq doesn't query `all_proxy` and `ALL_PROXY` environment variables by
itself, the way curl does. So set the proxy in code if any of the above
environment variables are set.
Signed-off-by: Mira Limbeck
---
v2:
- changed
ureq has support for a HTTP proxy, but no support for HTTPS proxy yet.
ureq doesn't query `all_proxy` and `ALL_PROXY` environment variables by
itself the way curl does. So set the proxy in code if any of the above
environment variables are set.
Signed-off-by: Mira Limbeck
---
src
Perl's automatic conversion can lead to integers being converted to
strings, for example by matching it in a regex.
To make sure we always return an integer in the API call, add an
explicit cast to integer.
Signed-off-by: Mira Limbeck
Reviewed-by: Fabian Ebner
---
v3:
- fixed style nits
`qemu-img info --output=json` returns the size and used values as integers in
the JSON format, but the regex match converts them to strings.
As we know they only contain digits, we can simply cast them back to integers
after the regex.
The API requires them to be integers.
Signed-off-by: Mira
`qemu-img info --output=json` returns the size and used values as integers in
the JSON format, but the regex match converts them to strings.
As we know they only contain digits, we can simply cast them back to integers
after the regex.
The API requires them to be integers.
Signed-off-by: Mira
Perl's automatic conversion can lead to integers being converted to
strings, for example by matching it in a regex.
To make sure we always return an integer in the API calls, add
explicit casts to integer.
Signed-off-by: Mira Limbeck
---
v2: new
PVE/API2/Storage/Content.pm | 6 --
1 file
On 2/17/22 15:10, Fabian Ebner wrote:
Am 17.02.22 um 14:33 schrieb Mira Limbeck:
On 2/17/22 14:24, Fabian Ebner wrote:
Am 17.02.22 um 13:55 schrieb Mira Limbeck:
'qemu-img info' with output format 'json' returns the size and used
values as
integers, but the regex match converts them
On 2/17/22 14:24, Fabian Ebner wrote:
Am 17.02.22 um 13:55 schrieb Mira Limbeck:
'qemu-img info' with output format 'json' returns the size and used values as
integers, but the regex match converts them to strings.
As we know they only contain digits, we can simply cast them back to integers
'qemu-img info' with output format 'json' returns the size and used values as
integers, but the regex match converts them to strings.
As we know they only contain digits, we can simply cast them back to integers
after the regex.
The API requires them to be integers.
Signed-off-by: Mira Limbeck
pool_exist($pool);
- $rpcenv->check_perm_modify($authuser, "/pool/$pool");
}
if ($rpcenv->check($authuser, "/vms/$vmid", ['VM.Allocate'], 1)) {
Works when no Permissions.Modify and no Pool.Allocate permission is set.
With the patch adding vendor-data support to cloud-init, a use of
uninitialized value was introduced. This can be fixed by setting it to
an empty string if no vendor-data is defined.
vendor-data can only be set via --cicustom and is optional.
Signed-off-by: Mira Limbeck
---
PVE/QemuServer
Looks good.
Reviewed-by: Mira Limbeck
On 10/30/21 4:49 PM, Constantin Herold wrote:
Signed-off-by: Constantin Herold
---
PVE/QemuServer.pm | 8
PVE/QemuServer/Cloudinit.pm | 25 -
2 files changed, 28 insertions(+), 5 deletions(-)
diff --git
With the release of PBS, and switching /etc/apt/auth.conf to use
/debian/pve as path instead of just /debian, the sources.list for
pve-enterprise must contain the /pve at the end. Otherwise there is no
match and updating via the pve-enterprise repository doesn't work.
Signed-off-by: Mira Limbeck
I tested your patch with a Ubuntu 20.04 cloudimg. Works fine
The patch looks good except for three small things. See comments inline.
On 8/23/21 4:34 PM, Constantin Herold wrote:
v2 rebase with master
Any updates on this ? Patch has been available since june
reference:
Thank you for the patch. Sorry that I missed that on friday.
Which distributions and cloud-init versions did you test this on?
The patch looks good to me. One addition I'd make, as we have a hard
limit of 4MiB currently for the generated ISO, inline.
On 8/12/21 9:17 PM, Constantin Herold
With live migration with local cloud-init disk now possible via the CLI
also remove the limitation in the GUI.
Signed-off-by: Mira Limbeck
---
www/manager6/window/Migrate.js | 10 +-
1 file changed, 1 insertion(+), 9 deletions(-)
diff --git a/www/manager6/window/Migrate.js b/www
The content of the ISO should be the same on both nodes, so offline
migrate the ISO, but don't regenerate it on VM start on the target node.
This way even with snippets the content will not change during live
migration.
Signed-off-by: Mira Limbeck
---
PVE/QemuMigrate.pm | 3 ++-
PVE
Automatically updating packages on boot is not always a good idea, so
remove that setting.
The `package_upgrade` setting can still be set via snippets if required.
Signed-off-by: Mira Limbeck
---
PVE/QemuServer/Cloudinit.pm | 2 --
1 file changed, 2 deletions(-)
diff --git a/PVE/QemuServer
data is implemented via snippets only.
[0] https://cloudinit.readthedocs.io/en/latest/topics/vendordata.html
Signed-off-by: Mira Limbeck
---
PVE/QemuServer.pm | 8
PVE/QemuServer/Cloudinit.pm | 16 ++--
2 files changed, 22 insertions(+), 2 deletions(-)
diff
These 2 files can be helpful for issues with multipath. The multipath -v3
output is too large most of the time and not required for analyzing and
solving the issues.
Signed-off-by: Mira Limbeck
---
PVE/Report.pm | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/PVE
Why do you add the macaddress here? I couldn't find anything in this nor
in the previous patch series explaining why this is done.
On 3/28/21 5:12 PM, Alexandre Derumier wrote:
---
PVE/API2/Qemu.pm| 73 +
PVE/CLI/qm.pm | 1 +
Thank you for the patch series and the GUI patches.
Some comments inline.
On 3/28/21 5:11 PM, Alexandre Derumier wrote:
Instead using vm pending options for pending cloudinit generated config,
write current generated cloudinit config in a new [special:cloudinit] SECTION.
Currently, some
A fix was also provided in bugzilla by user wsapplegate:
https://bugzilla.proxmox.com/show_bug.cgi?id=3314
Tested on Ubuntu 20.04, CentOS 8 and Debian 10.
Signed-off-by: Mira Limbeck
---
anything else required regarding attribution?
PVE/QemuServer/Cloudinit.pm | 2 +-
1 file changed, 1
20.4), CentOS 8 (ci 19.4), Debian 10 (ci
20.2).
Signed-off-by: Mira Limbeck
---
PVE/QemuServer.pm | 3 ++-
PVE/QemuServer/Cloudinit.pm | 4 ++--
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 1c0b5c2..7dd4bcf 100644
--- a/PVE
Now that SLAAC is supported, we can revert commit 76fdf552.
SLAAC requires cloud-init 19.4 or newer.
Signed-off-by: Mira Limbeck
---
www/manager6/qemu/IPConfigEdit.js | 8
1 file changed, 8 insertions(+)
diff --git a/www/manager6/qemu/IPConfigEdit.js
b/www/manager6/qemu
The docs mention -- as valid syntax for firewall rules, but the
code that parses the .fw files only accepts -. To make it
consistent with the docs and the API, also accept --.
In addition allow 'proto' as option, not only '-p'.
Signed-off-by: Mira Limbeck
---
src/PVE/Firewall.pm | 14
Thank you for the patch.
I tried to test this patch with both a CentOS 8 and Ubuntu 20.04 VM and
it never worked right.
I added another line for 'once|yes|on|true' where ssh_deletekeys is set
to 'true' as that is missing in this patch. Then I tried it with CentOS
8 and cloud-init 19.4.
In
exhibits the same behavior).
Signed-off-by: Mira Limbeck
---
v3:
- split the functionality from Socket and into their own files
- fixed MNL_SOCKET_BUFFER_SIZE to check configured page size at runtime
(added lazy_static dependency)
- changed is_ipv6 function to check the attribute keys instead
Signed-off-by: Mira Limbeck
---
v3:
- unchanged
v2:
- new addition, can be ignored as it only adds batch and print function
bindings which are not used in the code currently
src/mnl.rs | 31 +--
1 file changed, 29 insertions(+), 2 deletions(-)
diff --git a/src
Expectation support requires net.netfilter.nf_conntrack_helper to be set
to 1. In addition the helper modules have to be loaded as well. In the
tests nf_conntrack_ftp was used as helper.
Signed-off-by: Mira Limbeck
---
v3:
- split expect functions into their own files
- made required functions
Signed-off-by: Mira Limbeck
---
v3:
- updated package versions
- changed build dir and added temporary build dir for debcargo
- removed build-profile 'nodoc' (required creating debian/SOURCE file)
.cargo/config| 5
Makefile | 68
On 2/5/21 9:11 AM, Fabian Ebner wrote:
Am 03.02.21 um 15:25 schrieb Mira Limbeck:
Requires the pve-conntrack-tool. On migration the conntrack information
from the source node is dumped and sent to the target node where it is
then inserted.
This helps with open connections during migration when
Forgot to change this, it is actually the minimum of
sysconf(_SC_PAGESIZE) and 8192. Will fix this in v3.
On 2/3/21 3:25 PM, Mira Limbeck wrote:
+pub const MNL_SOCKET_BUFFER_SIZE: libc::c_int = 8192;
___
pve-devel mailing list
pve-devel
On 2/4/21 9:07 AM, Thomas Lamprecht wrote:
On 03.02.21 15:25, Mira Limbeck wrote:
Dumping conntrack information and importing conntrack information works
for IPv4 and IPv6. No filtering is supported for now. pve-conntrack-tool
will always return both IPv4 and IPv6 conntracks together
On 2/4/21 8:08 AM, Thomas Lamprecht wrote:
On 03.02.21 15:25, Mira Limbeck wrote:
Signed-off-by: Mira Limbeck
---
v2:
- unchanged
Some nits/comments inline.
You probably adapted this from proxmox-backup so my comments may hold true
there too.
pmg-log-tracker actually
.cargo/config
exhibits the same behavior).
Expectation support, which is necessary for FTP and other protocols, is
not yet implemented.
Signed-off-by: Mira Limbeck
---
v2:
- changed Conntracks to Socket
- reworked a lot of the code for less code duplication
- reduced usage of 'unsafe'
- added/changed things
specific are renamed to
contain 'conntrack' in their names.
Signed-off-by: Mira Limbeck
---
v2:
- mostly the same changes as for patch 1
src/main.rs| 249 +++--
src/netfilter_conntrack.rs | 44 +++
2 files changed, 285 insertions(+), 8
Internally we still have to use a C callback, but all it does is forward
to the closure we pass to it.
Signed-off-by: Mira Limbeck
---
v2:
- new addition
src/main.rs | 208 +++-
1 file changed, 109 insertions(+), 99 deletions(-)
diff --git
Signed-off-by: Mira Limbeck
---
v2:
- unchanged
.cargo/config| 5
Makefile | 63
debian/changelog | 5
debian/copyright | 16 +++
debian/debcargo.toml | 17
debian/docs | 1 +
6
API call.
Signed-off-by: Mira Limbeck
---
v2:
- added the migrate-conntracks option so that it only copies conntrack
information when requested
PVE/API2/Qemu.pm | 5 +
PVE/QemuMigrate.pm | 5 +
2 files changed, 10 insertions(+)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
Signed-off-by: Mira Limbeck
---
v2:
- new addition, can be ignored as it only adds batch and print function
bindings which are not used in the code currently
src/mnl.rs | 31 +--
1 file changed, 29 insertions(+), 2 deletions(-)
diff --git a/src/mnl.rs b/src
Thank you for the patch.
It doesn't apply on the latest qemu-server master. Looks like your
Cloudinit.pm file already contained changes which are not part of the patch.
Was it just the previous patch you sent?
Some additional comments inline.
On 1/14/21 6:11 PM, Alexandre Derumier wrote:
is possible as there are already active jobs
using the disks.
To fix it we have to call qemu_drive_mirror_monitor directly in the case
of cloudinit when completion is requested and there are jobs defined.
Signed-off-by: Mira Limbeck
---
PVE/QemuServer.pm | 6 ++
1 file changed, 6 insertions
We could add vendor data and put the ssh keys there:
https://cloudinit.readthedocs.io/en/latest/topics/vendordata.html
On 1/13/21 10:01 AM, Alexandre Derumier wrote:
Currently, we always regenerate sshkeys on any config change.
It should be done only before the first vm start, but currently
Tested Stefan's prebuilt qemu package with this patch applied and my VM
that has the issue.
Additionally tested the case of a full backup target
(https://forum.proxmox.com/threads/vm-hard-freezes-on-backup.81752/) and
it no longer hangs.
Looks good in my tests, so:
Tested-by: Mira Limbeck
On 12/15/20 2:57 PM, Thomas Lamprecht wrote:
On 02.12.20 13:50, Mira Limbeck wrote:
We only added the format extension when it was not 'raw'. But on file level
storages we always require it. To fix this, always add the format
extension if the storage provides the 'path' property
We only added the format extension when it was not 'raw'. But on file level
storages we always require it. To fix this, always add the format
extension if the storage provides the 'path' property.
This is the same logic we use in create_disks for cloudinit disks.
Signed-off-by: Mira Limbeck
Only fixes the clone_disk case, not the restore from backup one. Will
send a v2 with both fixes.
On 12/1/20 3:53 PM, Mira Limbeck wrote:
We only added the format extension when it was not 'raw'. But on file level
storages we always require it. To fix this, always add the format
extension
We only added the format extension when it was not 'raw'. But on file level
storages we always require it. To fix this, always add the format
extension if the storage provides the 'path' property.
This is the same logic we use in create_disks for cloudinit disks.
Signed-off-by: Mira Limbeck
Signed-off-by: Mira Limbeck
---
The configs in debian/ as well as the Makefile are copied from
pmg-log-tracker and modified to fit pve-conntrack-tool. I've kept the
vcs_git and vcs_browser configs, but replaced pmg-log-tracker with
pve-conntrack-tool.
Should that be removed as it doesn't exist
Requires the pve-conntrack-tool. On migration the conntrack information
from the source node is dumped and sent to the target node where it is
then inserted.
This helps with open connections during migration when the firewall is active.
Signed-off-by: Mira Limbeck
---
PVE/QemuMigrate.pm | 3
-tools'
conntrack also exhibits the same behavior). They are still dumped.
Expectation support, which is necessary for FTP and other protocols, is
not yet implemented.
Signed-off-by: Mira Limbeck
---
Cargo.toml | 15 ++
src/main.rs| 473
On 10/6/20 10:56 AM, Mira Limbeck wrote:
On 10/5/20 5:35 PM, Thomas Lamprecht wrote:
On 28.09.20 10:36, Mira Limbeck wrote:
After migration or a rollback the cloudinit disk might not be
allocated, so
volume_size_info() fails. As we override the value anyway for cloudinit
and efi disks simply
On 10/5/20 5:35 PM, Thomas Lamprecht wrote:
On 28.09.20 10:36, Mira Limbeck wrote:
After migration or a rollback the cloudinit disk might not be allocated, so
volume_size_info() fails. As we override the value anyway for cloudinit
and efi disks simply move the volume_size_info() call
All volumes contained in $vollist are activated. In this case a snapshot
of the volume. For cloudinit disks no snapshots are created so don't add
it to the list of volumes to activate as it otherwise fails with no
logical volume found.
Signed-off-by: Mira Limbeck
---
v2: unchanged
PVE/API2
After migration or a rollback the cloudinit disk might not be allocated, so
volume_size_info() fails. As we override the value anyway for cloudinit
and efi disks simply move the volume_size_info() call into the 'else'
branch.
Signed-off-by: Mira Limbeck
---
v2: changed subject
PVE
1 - 100 of 111 matches
Mail list logo
