the vmbr0 used eno1 as bridge_port, but it is already
in use by bond0, making the example wrong (as in, it does not work) and
confusing
since the example is about ip addresses on bonds, remove the vmbr0
Signed-off-by: Dominik Csapak
---
pve-network.adoc | 9 -
1 file changed, 9
or in widget-toolkit that if they are
modified, both pve and pmg have to be looked at
I'll send patch today
great :)
- Mail original -
De: "Dominik Csapak"
À: "pve-devel"
Envoyé: Mercredi 8 Janvier 2020 09:59:15
Objet: Re: [pve-devel] [PATCH proxmox-widget-too
On 1/8/20 6:55 AM, Victor Hooi wrote:
Hi,
I have been using Proxmox to run MacOS VMs, as per Nichola's guide here:
https://www.nicksherlock.com/2019/10/installing-macos-catalina-10-15-on-proxmox-6/
(and earlier versions).
In one of the steps in the guide - you need to go into the OMVF BIOS
se
comments inline
On 1/8/20 4:31 AM, Alexandre Derumier wrote:
Signed-off-by: Alexandre Derumier
---
node/NetworkEdit.js | 9 +
1 file changed, 9 insertions(+)
diff --git a/node/NetworkEdit.js b/node/NetworkEdit.js
index ca0011f..eeb612c 100644
--- a/node/NetworkEdit.js
+++ b/node/Net
On 1/8/20 4:31 AM, Alexandre Derumier wrote:
Following pve-common serie
Alexandre Derumier (2):
networkedit: bond: add bond-primary field
networkedit: add mtu field
node/NetworkEdit.js | 25 +
1 file changed, 25 insertions(+)
looks mostly ok, (i'll write an e
get an api error that bwlimit expects
an integer and not '')
Signed-off-by: Dominik Csapak
---
www/manager6/form/BandwidthSelector.js | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/www/manager6/form/BandwidthSelector.js
b/www/manager6/form/BandwidthSelector.js
i
so that it gets checked before the special case of removing the vmstate
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index e7ed6e9..0c4486b 100644
--- a/PVE/API2/Qemu.pm
+++ b
we need that on another place, so refactor in its own sub
Signed-off-by: Dominik Csapak
---
changes from v1:
* improved comments
PVE/QemuConfig.pm | 18 +++---
PVE/QemuServer.pm | 24
2 files changed, 27 insertions(+), 15 deletions(-)
diff --git a/PVE
and reference it also in the snapshot chapter
Signed-off-by: Dominik Csapak
---
new in v2
qm.adoc | 26 ++
1 file changed, 26 insertions(+)
diff --git a/qm.adoc b/qm.adoc
index b13f0f4..f0d63af 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -1193,6 +1193,28 @@ It will be called
Signed-off-by: Dominik Csapak
---
new in v2
www/manager6/qemu/Options.js | 21 +
1 file changed, 21 insertions(+)
diff --git a/www/manager6/qemu/Options.js b/www/manager6/qemu/Options.js
index e6f279c8..da61a6a8 100644
--- a/www/manager6/qemu/Options.js
+++ b/www/manager6
only VM.PowerMgmt is not enough, since we allocate space on a storage,
so we need VM.Config.Disk on the vm and Datastore.AllocateSpace on the storage
Signed-off-by: Dominik Csapak
---
changes from v1:
* move priv check inside if condition in vm_susepnd and
note that the caller has to check the
::QemuServer::vm_suspend
the gui patch needs the docs patch as dev dependcy for the onlinehelp section
qemu-server:
Dominik Csapak (3):
refactor finding of vmstate storage
suspend to disk: check more permissions
move vmstate check to check_vm_modify_config_perm
PVE/API2/Qemu.pm | 24
just for the record, here the summary of the
offline discussion about this
for now we should add the vmstatestorage option to the gui
(to allow the user to overwrite the target storage)
and as soon as we have the api calls for
checking finer grained permissions on paths, we
can implmement a stor
by mocking the lspci call
the mocked lspci code is basically the same as the real one,
only difference is the source of the devices and
there is no verbose flag
Signed-off-by: Dominik Csapak
---
changes from v1:
* improved lspci sub (thanks @thomas)
.../q35-linux-hostpci-multifunction.conf
if the device was explicitly
given, but did not exists (we showed the commandline, now we fail)
this also slightly changes the commandline for qemu (adding always
the domain), which is not a problem since we cannot live migrate
or snapshot such vms, but we have to adapt the tests
Signed-off
On 12/7/19 12:25 PM, Thomas Lamprecht wrote:
On 12/6/19 2:29 PM, Dominik Csapak wrote:
by mocking the lspci call
the mocked lspci code is basically the same as the real one,
only difference is the source of the devices and
there is no verbose flag
Signed-off-by: Dominik Csapak
---
.../q35
Reviewed-By: Dominik Csapak
Tested-By: Dominik Csapak
On 12/6/19 2:04 PM, Tim Marx wrote:
Signed-off-by: Tim Marx
---
PVE/API2/Disks/LVMThin.pm | 21 +++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git a/PVE/API2/Disks/LVMThin.pm b/PVE/API2/Disks/LVMThin.pm
device was explicitly
given, but did not exists (we showed the commandline, now we fail)
this also slightly changes the commandline for qemu (adding always
the domain), which is not a problem since we cannot live migrate
or snapshot such vms, but we have to adapt the tests
Signed-off-by
by mocking the lspci call
the mocked lspci code is basically the same as the real one,
only difference is the source of the devices and
there is no verbose flag
Signed-off-by: Dominik Csapak
---
.../q35-linux-hostpci-multifunction.conf | 16 +++
.../q35-linux-hostpci
UX nit: we need to give the user the option to set a vmstatestorage on
the GUI now - otherwise this might reliably pick a storage that the user
cannot access, while there are other storages where they might be
allowed to allocate a vmstate volume but they don't get that choice.
-> make vmstat
so that it gets checked before the special case of removing the vmstate
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index 9b25106..d05c91f 100644
--- a/PVE/API2/Qemu.pm
+++ b
only VM.PowerMgmt is not enough, since we allocate space on a storage,
so we need VM.Config.Disk on the vm and Datastore.AllocateSpace on the storage
Signed-off-by: Dominik Csapak
---
i am not sure if we need the last hunk (the check inside vm_suspend) at all,
since we now always give a
we need that on another place, so refactor in its own sub
Signed-off-by: Dominik Csapak
---
PVE/QemuConfig.pm | 15 ++-
PVE/QemuServer.pm | 23 +++
2 files changed, 25 insertions(+), 13 deletions(-)
diff --git a/PVE/QemuConfig.pm b/PVE/QemuConfig.pm
index
applied
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
applied both patches and followed up with a fix for the unit conversion
i set backendUnit to 'KiB' and removed the multiplication with 1024
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
this, there are some weird glitches on resizing with the
scrollbar
Signed-off-by: Dominik Csapak
---
www/manager6/Utils.js | 23 -
www/manager6/panel/GuestSummary.js | 53 ++
2 files changed, 46 insertions(+), 30 deletions(-)
diff --git a/www
number of columns than
their preference, we add a browser setting to set the number of columns
manually to either 'auto', 1,2 or 3
Dominik Csapak (5):
gui: refactor {lxc,qemu}/Summary to panel/GuestSummary
gui: GuestSummary: choose columns based on container width
gui: {dc,nod
to make sure the panels are always readable
Signed-off-by: Dominik Csapak
---
www/manager6/node/Summary.js | 1 +
www/manager6/panel/GuestSummary.js | 1 +
2 files changed, 2 insertions(+)
diff --git a/www/manager6/node/Summary.js b/www/manager6/node/Summary.js
index f106d312..ccda0cad
instead of viewport size, similar to the guest summary
Signed-off-by: Dominik Csapak
---
www/manager6/dc/Summary.js | 16 +++-
www/manager6/node/Summary.js | 20 +---
2 files changed, 16 insertions(+), 20 deletions(-)
diff --git a/www/manager6/dc/Summary.js b/www
introduce a new browser setting, with which the users can choose between
the automatic mode of choosing columns, or always using 1/2/3 columns
Signed-off-by: Dominik Csapak
---
www/manager6/Utils.js | 11 ++-
www/manager6/dc/Summary.js | 7 +++
www/manager6
they are both similar enough to have it only once
Signed-off-by: Dominik Csapak
---
www/manager6/Makefile | 3 +-
www/manager6/lxc/Config.js| 2 +-
www/manager6/lxc/Summary.js | 139 --
.../Summary.js => pa
so that the user knows what got loaded as vmstate
Signed-off-by: Dominik Csapak
---
PVE/QemuServer.pm | 1 +
1 file changed, 1 insertion(+)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 75e4f67..5cf04e6 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
@@ -4058,6 +4058,7 @@ sub
if a user removed the vmstate from the config for whatever reason,
a vmstart did not remove the 'suspended' lock
so always delete it and delete the vmstate only if it really was there
Signed-off-by: Dominik Csapak
---
PVE/QemuServer.pm | 8 +---
1 file changed, 5 insertions(+), 3
es with it. The user still has to have the right
priviliges and the vm cannot be 'protected' for this to work
also, the lock will only be skipped, if the deletion of the vmstate
is the only thing that is done
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 15 ++-
1 fil
this series improves suspend to disk handling for some edge/error cases,
and adds the possibility to remove the 'suspended' state without
having to resume the vm
qemu-server:
Dominik Csapak (4):
logging vmstate into Resume/Start Task
delete suspended lock when we have no vmstate
we did not actually delete the state if we deleted the 'vmstate' config,
leaving stray vmstates on the disks
actually implement the removal, requiring 'VM.Config.Disk' and
'VM.PowerMgmt' privs
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 9 +
so that a user can remove the 'suspended' state without resuming it
Signed-off-by: Dominik Csapak
---
www/manager6/qemu/HardwareView.js | 5 +
1 file changed, 5 insertions(+)
diff --git a/www/manager6/qemu/HardwareView.js
b/www/manager6/qemu/HardwareView.js
index 937842f9..e834a
has to manually edit the user.cfg or set the email via
pveum)
i noticed this while testing and using the 'insert emoji' menu in the
context menu
Signed-off-by: Dominik Csapak
---
proxinstall | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/proxinstall b/proxinstall
in some languages (e.g. German) the default (100) is too short for some labels,
resulting in cut-off text. We often use 120 in such cases, so we should
do here as well
Signed-off-by: Dominik Csapak
---
www/manager6/dc/OptionView.js | 1 +
1 file changed, 1 insertion(+)
diff --git a/www
otherwise, an api error leads to an empty UserView, and the user
wonders why no users show up
Signed-off-by: Dominik Csapak
---
www/manager6/dc/UserView.js | 2 ++
1 file changed, 2 insertions(+)
diff --git a/www/manager6/dc/UserView.js b/www/manager6/dc/UserView.js
index 8918fb2b..922d072b
just a high level comment, since i did not found anything in the commit:
i think we should warn/notify the user if he boots in bios mode
and installs/tries to install on 4kn disks, since the user will
not be able to boot in that case (since we do not install a bootloader?)
i'd be very confused i
soon (to avoid noise)
* drop the requirement of a PVE issued CA
* limit the length of the renewal also by the expiry date of the ca
pve-manager:
Dominik Csapak (1):
renew pve-ssl.pem when it nearly expires
PVE/CertHelpers.pm | 6 ++
bin/pveupdate | 32
Signed-off-by: Dominik Csapak
---
changes from v1:
* limit by expiry date of ca also
data/PVE/Cluster/Setup.pm | 20 +++-
1 file changed, 19 insertions(+), 1 deletion(-)
diff --git a/data/PVE/Cluster/Setup.pm b/data/PVE/Cluster/Setup.pm
index 81e3ef8..7b3bc31 100644
--- a/data/PVE
but only if the cert is issued by the ca in /etc/pve/pve-root-ca.pem
(by checking the issuer and openssl verify)
this way we can reduce the lifetime of the certs without having
to worry that they ran out
Signed-off-by: Dominik Csapak
---
changes from v1:
* only make checks if the cert expires
+eval {
+# get CA and check issuer
+my $capath = "/etc/pve/pve-root-ca.pem";
+my $cainfo = PVE::Certificate::get_certificate_info($capath);
+if ($cainfo->{subject} !~ m|/CN=Proxmox Virtual Environment/.*/O=PVE
Cluster Manager CA|) {
+ die "Root CA is not issued by Proxmox
On 11/20/19 8:10 PM, Thomas Lamprecht wrote:
On 10/31/19 1:36 PM, Dominik Csapak wrote:
this series lets users add tags to guest configs
these do not have any concrete meaning but are intended to be used
by management software such as config management systems
this is the basic implementation
on multiSelect we have to check the values that is in the values
array, but we get the 'displaystring' in the validator so we
have to get the 'real' value (aka the underlying array of values)
before checking if they are in the store
Signed-off-by: Dominik Csapak
---
f
t set an empty field wrongfully to invalid)
Signed-off-by: Dominik Csapak
---
form/ComboGrid.js | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/form/ComboGrid.js b/form/ComboGrid.js
index 3cc64f9..0c5cf1b 100644
--- a/form/ComboGrid.js
+++ b/form/ComboGrid.js
@@ -52,7
We are in the QemuServer package not in LXC, so use the correct
package for the Config, namely QemuConfig
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 1890448..54c8c88 100644
On 11/13/19 2:44 PM, Marco Gabriel - inett GmbH wrote:
Hi,
Hi,
long time ago, I requested adding the container hostname or vm name in bugzilla
request #438. As I wanted to practice developing features for pve and how to
use the git repos, I implemented the feature. I should have asked befo
we maybe want to backport this for pve5 also.. (should apply i guess)
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
the code returns undef in case there is no 'tos', and the code
calling this api call handles a non-existing tos already, but
fails in that case becasue of the failing return value verification
Signed-off-by: Dominik Csapak
---
PVE/API2/ACMEAccount.pm | 1 +
1 file changed, 1 insertio
On 11/12/19 2:55 PM, Thomas Lamprecht wrote:
On 11/12/19 2:23 PM, Dominik Csapak wrote:
but prepend '' to ids where no domain is given, to keep the ability
to use the shorthand syntax (e.g. 00:01.0 instead of :00:01.0)
Signed-off-by: Dominik Csapak
---
src/PVE/SysFSToo
t exist in the store...)
can we ignore this? if not, i would send a v2 where i try to detect
the format we get from the backend and set the value in the selector
accordingly (but i am not sure if its worth the effort, since partially
upgrading a cluster is not recommended)
pve-common:
Dominik Csap
but prepend '' to ids where no domain is given, to keep the ability
to use the shorthand syntax (e.g. 00:01.0 instead of :00:01.0)
Signed-off-by: Dominik Csapak
---
src/PVE/SysFSTools.pm | 20 +---
1 file changed, 13 insertions(+), 7 deletions(-)
diff --git
relax the regex for hostpci to allow different pci domains than
Signed-off-by: Dominik Csapak
---
PVE/QemuServer.pm | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 1890448..709dcdb 100644
--- a/PVE/QemuServer.pm
+++ b
but remove the default domain '' before sending to the backend,
and add it if no domain is given in the config
Signed-off-by: Dominik Csapak
---
www/manager6/form/PCISelector.js | 2 +-
www/manager6/qemu/PCIEdit.js | 11 +--
2 files changed, 10 insertions(+), 3
formatted numbers
Signed-off-by: Dominik Csapak
---
src/PVE/Tools.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm
index 02c2886..89de4ec 100644
--- a/src/PVE/Tools.pm
+++ b/src/PVE/Tools.pm
@@ -1008,7 +1008,7 @@ sub df {
sha-512 crypted passwords are longer than 64 byte, and also
it does not make sense to limit passwords to such a short length.
increase to 1024 (should be enough for a while and limits
possible attack surface for the api/guest-agent)
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu/Agent.pm | 2
i only partially quoted, since most is already clear
yes librados and ceph config is available, but that does not mean the
cluster is designed so that all nodes can reach the monitor nodes...
e.g.:
5 nodes with node0-node2 ceph nodes, node3 a 'compute' node, and
node4 is a node in the same clu
On 11/6/19 3:49 PM, Thomas Lamprecht wrote:
On 11/6/19 8:36 AM, Dominik Csapak wrote:
On 11/5/19 6:33 PM, Thomas Lamprecht wrote:
On 11/5/19 1:51 PM, Dominik Csapak wrote:
+my $entry = {
+time => $time,
+ops_r => $pgmap->{read_op_per_sec},
+ops_w => $pgmap->{w
On 11/5/19 6:33 PM, Thomas Lamprecht wrote:
On 11/5/19 1:51 PM, Dominik Csapak wrote:
add a helper to cache the ceph performance data inside pmxcfs
with broadcast_node_kv, and also a helper to read it out
merge the data from all nodes that sent performance data
the '$perf_cache'
this calls the api once when opening the dashboard, and fills the charts
with that cached data, then it continues like previously with its
interval
Signed-off-by: Dominik Csapak
---
www/manager6/ceph/Status.js | 18 ++
1 file changed, 18 insertions(+)
diff --git a/www/manager6
t only its values,
and the reader (pvedaemon) uses it to cache all nodes data
Signed-off-by: Dominik Csapak
---
merging the data on read seems like a good idea, since we have the data
and it should make little sense to throw any way, but i noticed some
weird glitches when the pvestat update calls are
Signed-off-by: Dominik Csapak
---
altough the rados call have a default timeout of 5 seconds,
i am not sure if this is too much of a load
we could only run that call e.g. every 3rd pvestatd loop, to minimize
the load. combined with the merging of the data on the reader side,
i guess we would
this gets the cached ceph performance data back from pmxcfs
Signed-off-by: Dominik Csapak
---
PVE/API2/Cluster/Ceph.pm | 24
1 file changed, 24 insertions(+)
diff --git a/PVE/API2/Cluster/Ceph.pm b/PVE/API2/Cluster/Ceph.pm
index fa79d077..31373ef2 100644
--- a/PVE/API2
with this patch, we are able to directly fill the charts on the ceph
dashboard with cached performance data, instead of always showing empty
ones each time we load the dashboard
Dominik Csapak (4):
ceph: add perf data cache helpers
ceph: cache perf data in pvestatd
ceph: add 'iostat
On 11/5/19 1:21 PM, Thomas Lamprecht wrote:
On 11/5/19 1:04 PM, Alexandre DERUMIER wrote:
Hi,
be carefull of timeouts, because it could hang pvestatd.
(That's why it was implemented in udp for graphite && influxdb)
That's true, as almost always there's a reason why seemingly simple
things ar
Hi,
thanks for the patch :)
looks good for the most part, but afaics you did not
sign our CLA yet, please see our Developer Documentation[0] for that.
also it would be nice to extend our documentation for this
in the pve-docs git repository[1], in the file
'pve-external-metric-server.adoc'[2], b
this is useful as meta information for e.g., provisioning or config
management systems
adding the info also to the 'status' api call to make it easier to show
it in the gui
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 1 +
PVE/QemuServer.pm | 14 +-
2 files c
this is useful as meta information for e.g., provisioning or config
management systems
adding the info also to the 'status' api call to make it easier to show
it in the gui
Signed-off-by: Dominik Csapak
---
thanks @oguz for including this already in the 'fastplug' opti
which is just the pencil symbol and opens the tageditor
Signed-off-by: Dominik Csapak
---
www/manager6/Makefile | 1 +
www/manager6/button/TagEdit.js | 34 ++
2 files changed, 35 insertions(+)
create mode 100644 www/manager6/button/TagEdit.js
diff
helpers to
* generate a color from a string consistently
* generate a html tag for a tag
* related css classes
Signed-off-by: Dominik Csapak
---
Utils.js | 35 ++-
css/ext6-pmx.css | 13 +
2 files changed, 47 insertions(+), 1 deletion(-)
diff
sets the correct line-height and margin, to not look so out of place
(it was slighty off-center vertically and had too much space around)
the margin has to be set on the item, because extjs sets the
position manually and does not look at the css classes...
Signed-off-by: Dominik Csapak
---
www
s api call of guests (for gui)
* show the tags in the gui
* make the tags editable in the gui
qemu-server:
Dominik Csapak (1):
add 'tags' option
PVE/API2/Qemu.pm | 1 +
PVE/QemuServer.pm | 14 +-
2 files changed, 14 insertions(+), 1 deletion(-)
pve-container:
Dominik C
api call)
Signed-off-by: Dominik Csapak
---
www/css/ext6-pve.css | 12
www/manager6/Makefile| 1 +
www/manager6/form/TagSelector.js | 117 +++
3 files changed, 130 insertions(+)
create mode 100644 www/manager6/form/TagSelector.js
diff
this way they are always visible when a user has selected a vm
Signed-off-by: Dominik Csapak
---
www/manager6/lxc/Config.js | 23 ---
www/manager6/qemu/Config.js | 24 +---
2 files changed, 41 insertions(+), 6 deletions(-)
diff --git a/www/manager6/lxc
ObjectGrid (an ancestor of PendingObjectGrid) does already have
a 'reload' function which does exactly the same, so get rid
of the local one here
Signed-off-by: Dominik Csapak
---
www/manager6/qemu/HardwareView.js | 36 ---
1 file changed, 14 inserti
reload when a new disk was added, and set the button status
on a reload
Signed-off-by: Dominik Csapak
---
www/manager6/lxc/Resources.js | 5 +
1 file changed, 5 insertions(+)
diff --git a/www/manager6/lxc/Resources.js b/www/manager6/lxc/Resources.js
index f6358122..1114c051 100644
--- a
Signed-off-by: Dominik Csapak
---
www/manager6/qemu/HDEdit.js | 53 ++---
1 file changed, 8 insertions(+), 45 deletions(-)
diff --git a/www/manager6/qemu/HDEdit.js b/www/manager6/qemu/HDEdit.js
index 46397d37..c810b4b0 100644
--- a/www/manager6/qemu/HDEdit.js
d the mp as
'pending', but without the size
Signed-off-by: Dominik Csapak
---
www/manager6/lxc/MPEdit.js | 20 ++--
1 file changed, 14 insertions(+), 6 deletions(-)
diff --git a/www/manager6/lxc/MPEdit.js b/www/manager6/lxc/MPEdit.js
index c36d26bd..780bef0f 100644
--
this helper conditionally sets the given value to the given property
on the given object, optionally a different value
this is useful for our MP/HD Edit panels, where we set the options
of the drive/mp this way for every gui option we have
Signed-off-by: Dominik Csapak
---
www/manager6
applied, thanks :)
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
On 10/30/19 9:30 AM, Thomas Lamprecht wrote:
On 10/30/19 9:23 AM, Dominik Csapak wrote:
On 10/30/19 8:51 AM, Thomas Lamprecht wrote:
On 10/3/19 1:50 PM, Dominik Csapak wrote:
so that the user can edit the tags in the gui
IMO the options is not really "correct", it's not a
On 10/30/19 8:51 AM, Thomas Lamprecht wrote:
On 10/3/19 1:50 PM, Dominik Csapak wrote:
so that the user can edit the tags in the gui
IMO the options is not really "correct", it's not a guest option
at all. Rather this fits to the "Notes" field of the Summary panel,
ma
looks mostly good to me, after a bit of search
i found that the reason for the delayed update
is that we/you queried the wrong store
the rstore in the context is the 'real store' which
triggered the update of the 'diffstore' that showed the
changes
the sequence was:
* api call
* update of rstor
will involve
much more work (probably a new api call)
for pmg, we do not show this button (atm)
proxmox-widget-toolkit:
Dominik Csapak (1):
NetworkView: add apply config button
node/NetworkView.js | 37 +++--
1 file changed, 35 insertions(+), 2 deletions(-)
pve
ant it for pmg
Signed-off-by: Dominik Csapak
---
i am not entirely happy with the changed gettext for the reboot,
but did not find a better one (suggestions welcome)
node/NetworkView.js | 37 +++--
1 file changed, 35 insertions(+), 2 deletions(-)
diff --
Signed-off-by: Dominik Csapak
---
www/manager6/node/Config.js | 1 +
1 file changed, 1 insertion(+)
diff --git a/www/manager6/node/Config.js b/www/manager6/node/Config.js
index 054ced64..91a999e1 100644
--- a/www/manager6/node/Config.js
+++ b/www/manager6/node/Config.js
@@ -169,6 +169,7
s Catalina) would
reject the certificate with the error: 'REVOKED' since
they have now stricter rules for certificates
since other os/browsers will probably also make the rules stricter,
it makes sense to shorten the time
pve-manager:
Dominik Csapak (1):
renew pve-ssl.pem when it
but only if the ca is ours, and the cert is issued by our ca
(by checking the issuer and openssl verify)
this way we can reduce the lifetime of the certs without having
to worry that they ran out
Signed-off-by: Dominik Csapak
---
PVE/CertHelpers.pm | 6 ++
bin/pveupdate | 33
instead of 10 years, to avoid issues with browsers/os that reject
certificates which have a longer lifetime
(e.g. macOs Catalina only accepts max 825 days if issued after july 2019)
Signed-off-by: Dominik Csapak
---
data/PVE/Cluster.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
any number of version levels,
for q35, i440fx and virt (to be more future proof)
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 328a0d1..b635760 100644
--- a/PVE/QemuServer.pm
uld do the deletion code and the pending add code
(e.g. delete the pending cloud init drive and creating it again)
to avoid that situation, we need to remove the option from the pending hash
in the 'delete loop'
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 1 +
1 file changed, 1 inse
otherwise, having multiple ipconfigX entries, can lead to different
instance-ids on different startups, which is not desired
Signed-off-by: Dominik Csapak
---
2 issues i have with this:
* we have a cyclic dependency between PVE::QemuServer and
PVE::QemuServer::Cloudinit, and this patch
tested a bit, much more testing wouldn't hurt for sure ;)
tested it with a few vms with varying amounts of iothreads, in snapshot
suspend and stop mode, seems all to work (incl restore)
Tested-By: Dominik Csapak
___
pve-devel mailing list
pve-
and use it also for efidisk creation and importdisk
this way we correctly handle zfs-over-iscsi options for those cases
also write tests for it
Signed-off-by: Dominik Csapak
---
changes from v1:
* changed error messages to something more sensible and include the volid
(also adapt the test for
Add tests for the qemu_img_convert parameters to the resulting
'qemu-img convert' call
we mock the 'run_command' and extract the 'cmd' parameter to
compare with what we expect
Signed-off-by: Dominik Csapak
---
changes from v1:
* changed /usr/bin/env perl to /usr/bi
On 10/17/19 9:26 AM, Thomas Lamprecht wrote:
On 10/16/19 11:27 AM, Dominik Csapak wrote:
storage config is motly copied from the config2command tests
some general info about what and *how* you test would be great here.
looks OK in general, and thanks for more test, some comments inline
and use it also for efidisk creation and importdisk
this way we correctly handle zfs-over-iscsi options for those cases
also write tests for it
Signed-off-by: Dominik Csapak
---
makes more sense when viewed with -w, but i think comparing the
complete source of qemu_img_convert before and after
301 - 400 of 3174 matches
Mail list logo
