Signed-off-by: Stefan Reiter
---
Much simpler than v1, simply chmods the temp file before copying.
Interestingly, on my machine anyway, scp seems to preserve file modes
without the -p flag too, but I added it in for consistency.
wget (and thus apl_download) creates a new file, and follows
Signed-off-by: Stefan Reiter
---
Based on recent conversation in pve-user list.
pvecm.adoc | 33 +++--
1 file changed, 31 insertions(+), 2 deletions(-)
diff --git a/pvecm.adoc b/pvecm.adoc
index 61453f7..773c2dd 100644
--- a/pvecm.adoc
+++ b/pvecm.adoc
@@ -149,7
Signed-off-by: Stefan Reiter
---
This one still fits nicely IMO, for the next version though, maybe 2.6+
or similar would be better.
www/manager6/Utils.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/Utils.js b/www/manager6/Utils.js
index 4be22f7f..e7244be5
returns the deleted element, deleting a
non-existant hash entry returns undef, which is fine after this point:
"deleting non-existent elements returns the undefined value in their
corresponding positions."
- https://perldoc.perl.org/functions/delete.html
Signed-off-by: Stefan Reiter
---
Broken symlinks will no longer be enumerated, thus not causing a
format validation in content GUI due to the missing filesize.
Signed-off-by: Stefan Reiter
---
PVE/Storage.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/PVE/Storage.pm b/PVE/Storage.pm
index 588e775
Signed-off-by: Stefan Reiter
---
PVE/API2/Storage/Status.pm | 25 +
1 file changed, 21 insertions(+), 4 deletions(-)
diff --git a/PVE/API2/Storage/Status.pm b/PVE/API2/Storage/Status.pm
index 9a5a952..8649a7d 100644
--- a/PVE/API2/Storage/Status.pm
+++ b/PVE/API2/Storage
Signed-off-by: Stefan Reiter
---
pve-installation.adoc | 2 +-
pve-intro.adoc | 6 +++---
pve-network.adoc| 28 ++--
pve-package-repos.adoc | 2 +-
qm-pci-passthrough.adoc | 12 ++--
5 files changed, 25 insertions(+), 25 deletions(-)
diff
On June 13, 2019 at 3:41 PM Thomas Lamprecht wrote:
>
>
> On 6/13/19 2:30 PM, Stefan Reiter wrote:
> > Broken symlinks will no longer be enumerated, thus not causing a
> > format validation in content GUI due to the missing filesize.
>
> so they just get hidden now?
as with tlbflush.
hv_evmcs is a VM control structure in L1 guest memory, allowing an L1 guest
to modify L2 VMCS and entering L2 without having the L0 host perform an
expensive VMCS update on trapping the nested vmenter.
Signed-off-by: Stefan Reiter
---
v1 -> v2:
* Added commit descript
d. In case it is not specified, behaviour stays the
same (default 10 min timeout).
Signed-off-by: Stefan Reiter
---
BTW, I did check "system_powerdown" and "quit", both are not blocking,
so no changes needed there.
PVE/QemuServer.pm | 9 ++---
1 file changed, 6 inser
> On June 21, 2019 10:53 AM Thomas Lamprecht wrote:
>
>
> Am 6/19/19 um 10:23 AM schrieb Stefan Reiter:
> > Kernels 4.18+ (4.17+ for evmcs) support new Hyper-V enlightenments for
> > Windows KVM guests. QEMU supports these since 3.0 and 3.1 respectively.
>
> On June 24, 2019 3:27 PM Thomas Lamprecht wrote:
>
>
> On 6/24/19 2:37 PM, Stefan Reiter wrote:
> > Migration tests ran OK, works fine in both directions (old <-> new), as
> > long as QEMU version stays the same and both systems have a kernel
> >
Thanks for applying, will include a changelog next time.
> On June 17, 2019 at 11:55 AM Thomas Lamprecht wrote:
>
>
> On 6/17/19 11:05 AM, Stefan Reiter wrote:
> > Symlinks with a non-existing target fail Perls '-f' test and were thus
> > not deleteable via the API (
they don't cause any penalties, enabling them regardless should be fine.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 7 +++
1 file changed, 7 insertions(+)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 341e0b0..88d07c2 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
Broken symlinks (and other files without a size) will now show up as 0
byte instead of causing a format validation error in the API.
Signed-off-by: Stefan Reiter
---
PVE/Storage.pm | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/PVE/Storage.pm b/PVE/Storage.pm
index
Symlinks with a non-existing target fail Perls '-f' test and were thus
not deleteable via the API (failing with '$path does not exist').
Signed-off-by: Stefan Reiter
---
Deleting intact symlinks deletes the symlink but leaves the file it
points to intact. This behaviour has not changed
Signed-off-by: Stefan Reiter
---
v2:
* Removed mentions of ring addresses to avoid confusion
(removed in Corosync 3 - though the rest of the file should be
updated for coro 3 too)
* Changed section title
* Added CAUTION: about hostnames, removed NOTE: about PVE tooling
* Reworded
Signed-off-by: Stefan Reiter
---
There were so many in this file it actually bothered me :)
src/PVE/Service/pve_firewall.pm | 42 -
1 file changed, 21 insertions(+), 21 deletions(-)
diff --git a/src/PVE/Service/pve_firewall.pm b/src/PVE/Service/pve_firewall.pm
Calling cfs_read_file with no corosync.conf (i.e. on a standalone node)
returns {} instead of undef. The previous patches assumes undef for this
scenario. To avoid confusing checks all over the place, simply leave the
config as undef if no file exists.
Signed-off-by: Stefan Reiter
---
src/PVE
If no corosync.conf exists (i.e. a standalone node), the output is left
the same.
Signed-off-by: Stefan Reiter
---
Is there a project standard regarding list output formatting?
I personally think it looks good and readable, but consistency with other
CLI tools would of course be preferrable
Lamprecht wrote:
On 7/8/19 6:26 PM, Stefan Reiter wrote:
Parts about multicast and RRP have been removed entirely. Instead, a new
section 'Corosync Redundancy' has been added explaining the concept of
links and link priorities.
note bad at all, still some notes inline.
Signed-off-by: Stefan Reiter
Parts about multicast and RRP have been removed entirely. Instead, a new
section 'Corosync Redundancy' has been added explaining the concept of
links and link priorities.
Signed-off-by: Stefan Reiter
---
v1 -> v2:
* Spelling mistakes
* Rewording to improve clarity
* Fixed redunda
Signed-off-by: Stefan Reiter
---
No changes for v2. As mentioned, I did not find any references to the changed
names.
pvecm.adoc | 30 +++---
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/pvecm.adoc b/pvecm.adoc
index 05756ca..1c0b9e7 100644
fix mentioned above.
Signed-off-by: Stefan Reiter
---
src/PVE/LXC/Create.pm | 23 +--
1 file changed, 17 insertions(+), 6 deletions(-)
diff --git a/src/PVE/LXC/Create.pm b/src/PVE/LXC/Create.pm
index 029c940..3f893e5 100644
--- a/src/PVE/LXC/Create.pm
+++ b/src/PVE/LXC/Create
Seems to be a regression introduced with
f360d7f16b094fa258cf82d2557d06f3284435e4 (related to #2028).
$conf->{'lxc'} would always be defined, hence we never replaced it with
the restored options.
Co-developed-by: Oguz Bektas
Signed-off-by: Stefan Reiter
---
Nevermind v1, perl arrays and has
Since we use only systemd, we can simply remove this file. Without
removing, the "systemd enable" command fails, complaining about unset
run-levels.
Signed-off-by: Stefan Reiter
---
The author of corosync-qdevice is aware of the issue:
https://bugs.launchpad.net/ubuntu/+source/corosy
Signed-off-by: Stefan Reiter
---
pvecm.adoc | 5 +
1 file changed, 5 insertions(+)
diff --git a/pvecm.adoc b/pvecm.adoc
index 7525bb5..d8f2341 100644
--- a/pvecm.adoc
+++ b/pvecm.adoc
@@ -309,6 +309,11 @@ cluster again, you have to
* then join it, as explained in the previous section
a remote ring1_addr cannot
contact a local ring0_addr, and vice versa.
Multicast is always allowed, for backwards compatibility (as was the
case before).
Signed-off-by: Stefan Reiter
maybe s/create/generate/ ? creating a rule is what the admin does when
configuring the firewall, generating is wh
Needed for corosync related firewall updates.
Signed-off-by: Stefan Reiter
---
src/PVE/Tools.pm | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm
index 4dd073f..77af21c 100644
--- a/src/PVE/Tools.pm
+++ b/src/PVE/Tools.pm
@@ -47,6 +47,8 @@ safe_print
Since corosync rules are now only created when a corosync.conf file is
present, a static corosync.conf has been added and will be loaded for
testing.
New test rules have been introduced to check corosync rules relating to
different rings/links.
Signed-off-by: Stefan Reiter
---
test
The sub 'for_all_corosync_addresses' iterates through all nodes in a
passed corosync config and calls a lambda function for every ringX_addr
on every node it finds.
Signed-off-by: Stefan Reiter
---
src/PVE/Firewall.pm | 20
1 file changed, 20 insertions(+)
diff --git a/src
te ring1_addr cannot
contact a local ring0_addr, and vice versa.
Multicast is always allowed, for backwards compatibility (as was the
case before).
Signed-off-by: Stefan Reiter
---
src/PVE/Firewall.pm | 69 +++--
1 file changed, 54 insertions(+), 15 deletions(
during test runs.
Source for corosync behaviour:
http://people.redhat.com/ccaulfie/docs/KnetCorosync.pdf
Signed-off-by: Stefan Reiter
---
src/PVE/Firewall.pm | 68 +++--
test/corosync.conf | 4 +++
2 files changed, 69 insertions(+), 3 deletions(-)
diff
:
Stefan Reiter (1):
Export getaddrinfo helpers
src/PVE/Tools.pm | 2 ++
1 file changed, 2 insertions(+)
pve-firewall:
Stefan Reiter (4):
Add function to iterate all ringX_addr for all nodes
Create corosync firewall rules independant of localnet
Update and add tests for corosync firewall
e again.
Signed-off-by: Stefan Reiter
---
v2:
* Coding style fixes
* Use IP-version check in iterator now
src/PVE/Firewall.pm | 77 +++--
1 file changed, 61 insertions(+), 16 deletions(-)
diff --git a/src/PVE/Firewall.pm b/src/PVE/Firewall.pm
index b08cea5..c497
Only applies to corosync 3.
Testing config is changed to allow simulation of multicast rules.
Signed-off-by: Stefan Reiter
---
Only for master, stable-5 (with corosync 2) is not compatible, since
this change assumes that the default transport is knet (or anything else
unicast, really).
src
codepaths.
Signed-off-by: Stefan Reiter
---
test/corosync.conf | 52 ++
test/fwtester.pl | 11 ++-
test/test-default-rules1/tests | 4 +++
3 files changed, 66 insertions(+), 1 deletion(-)
create mode 100644 test/corosync.conf
diff
address will be
best-effort resolved as hostnames. This has to happen in the exact same
way as corosync does internally, to ensure consistency with firewall
rules.
Signed-off-by: Stefan Reiter
---
v2:
* resolve_hostname_like_corosync now returns IP addresses if passed
(this simplifies
if
corosync.conf has changed, but I think that would be ok to send as a
different patch. The delay is hardly noticeable in my testing.
Maybe an inotify on corosync.conf in the firewall daemon?
pve-cluster:
Stefan Reiter (1):
Add functions to resolve hostnames and iterate corosync nodes
data/PVE
Signed-off-by: Stefan Reiter
---
Hope it's the correct style now.
I decided to this into its own commit, since it could technically be applied to
the docs for pve 5 as well (although with little visual effect).
pvecm.adoc | 30 +++---
1 file changed, 15 insertions
Parts about multicast and RRP have been removed entirely. Instead, a new
section 'Corosync Redundancy' has been added explaining the concept of
links and link priorities.
Signed-off-by: Stefan Reiter
---
pvecm.adoc | 372 +
1 file changed, 147
We are currently looking at a re-do of CPU models and flags in general,
including adding support for custom CPU models [0]. Feature flags like
aes would be automatically supported by this (i.e. you could create a
custom model derived from kvm64 and add aes to it).
Then again, I'm not sure
Whitespace removal and consolidating VZDump's job id format into a
local variable.
Signed-off-by: Stefan Reiter
---
PVE/API2/Backup.pm| 25 +
www/manager6/dc/Backup.js | 10 +-
2 files changed, 14 insertions(+), 21 deletions(-)
diff --git a/PVE/API2
ch need it fixed, so include that as well (use "mode" instead of
individual flags + "compress" is not a boolean).
Signed-off-by: Stefan Reiter
---
www/manager6/dc/Backup.js | 97 ---
1 file changed, 90 insertions(+), 7 deletions(-)
diff
perms are
needed now)
* Only contact online nodes
* Fix masking
* Fix pve-cluster-backup data model
* Split cleanup into seperate commit and include new variable for job_id format
manager: Stefan Reiter (3):
fix #1594: add "Run now" button to cluster backup page
fix #1594: allo
This was previously gated to CLI only, but it causes a vzdump job
started with the newly introduced "Run Now" button to fail if it
includes VMIDs on other nodes.
Signed-off-by: Stefan Reiter
---
PVE/API2/VZDump.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
On 8/14/19 12:03 PM, Thomas Lamprecht wrote:
Am 8/6/19 um 2:41 PM schrieb Stefan Reiter:
I don't see a reason to blanket-forbid excluding VMs in pool backups, so I felt
leaving the API unchanged was the better option in this case. The GUI is the
broken part, the API is working fine, albeit
clude" mode again, with the backend still
being on "pool" - to prevent this, we always delete a jobs "exclude" list
when it's switched to "pool".
Co-authored-by: Tim Marx
Signed-off-by: Stefan Reiter
---
PVE/API2/Backup.pm | 1 +
1 file changed, 1 insertion(+
On 8/14/19 4:20 PM, Dominik Csapak wrote:
this creates a reboot trigger file (inspired by pve-container)
and relies on the 'qm cleanup' call by the qmeventd to detect
and restart the vm afterwards
Signed-off-by: Dominik Csapak
---
PVE/API2/Qemu.pm | 45
-by: Stefan Reiter
---
Note that starting a VM with an img file attached breaks live migration to
machines without this patch - not sure how relevant, considering this is
a new addition anyway.
PVE/API2/Storage/Status.pm | 6 +++---
PVE/Storage.pm | 2 +-
PVE/Storage/Plugin.pm
By default, all file types are shown, but the user now has the option of
filtering only by supported types in the file selector dialog.
Signed-off-by: Stefan Reiter
---
www/manager6/storage/ContentView.js | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/www/manager6
Signed-off-by: Stefan Reiter
---
pvecm.adoc | 81 --
1 file changed, 60 insertions(+), 21 deletions(-)
diff --git a/pvecm.adoc b/pvecm.adoc
index e986a75..5379c3f 100644
--- a/pvecm.adoc
+++ b/pvecm.adoc
@@ -103,25 +103,33 @@ to the other
By default, all file types are shown, but the user now has the option of
filtering only by supported types in the file selector dialog.
Signed-off-by: Stefan Reiter
Acked-by: Dominik Csapak
---
I found it weird too, but also couldn't find a different way to go about it...
www/manager6
-off-by: Stefan Reiter
---
v1 -> v2:
* Refactored regex into variable
I left the [Ii][Ss]... since we don't know if the variable is used with //i
or not.
PVE/API2/Storage/Status.pm | 4 ++--
PVE/Storage.pm | 4 +++-
PVE/Storage/Plugin.pm | 4 ++--
3 files changed, 7 inserti
Signed-off-by: Stefan Reiter
---
v2:
* Changed some wording to remove "you"s and made sections clearer.
Big thanks to Aaron for the helpful review.
* Do not remove [[pvecm_create_cluster]] tag
pvecm.adoc | 77 --
1 file c
Signed-off-by: Stefan Reiter
---
Sorry for missing 2/2 on previous patch series, had some trouble with the
mailing list not accepting my mails.
v3:
* Changed "Add Node" section to allow better image placement
v2:
* Changed some wording to remove "you"s and made sections c
Signed-off-by: Stefan Reiter
---
pvecm.adoc | 6 ++
1 file changed, 6 insertions(+)
diff --git a/pvecm.adoc b/pvecm.adoc
index c41e691..7bccaa7 100644
--- a/pvecm.adoc
+++ b/pvecm.adoc
@@ -114,6 +114,8 @@ cluster name follows the same rules as node names.
Create via Web GUI
Signed-off-by: Stefan Reiter
---
images/screenshot/gui-cluster-join.png | Bin 0 -> 61710 bytes
1 file changed, 0 insertions(+), 0 deletions(-)
create mode 100644 images/screenshot/gui-cluster-join.png
diff --git a/images/screenshot/gui-cluster-join.png
b/images/screenshot/gui-clus
Signed-off-by: Stefan Reiter
---
.../screenshot/gui-cluster-join-information.png | Bin 0 -> 53366 bytes
1 file changed, 0 insertions(+), 0 deletions(-)
create mode 100644 images/screenshot/gui-cluster-join-information.png
diff --git a/images/screenshot/gui-cluster-join-information.png
Signed-off-by: Stefan Reiter
---
images/screenshot/gui-cluster-create.png | Bin 0 -> 18124 bytes
1 file changed, 0 insertions(+), 0 deletions(-)
create mode 100644 images/screenshot/gui-cluster-create.png
diff --git a/images/screenshot/gui-cluster-create.png
b/images/screenshot/gui-clus
On 9/3/19 9:53 AM, Thomas Lamprecht wrote:
On 22.08.19 14:39, Stefan Reiter wrote:
By default, all file types are shown, but the user now has the option of
filtering only by supported types in the file selector dialog.
Signed-off-by: Stefan Reiter
Acked-by: Dominik Csapak
---
I found
On 8/29/19 10:06 AM, Dominic Jäger wrote:
Signed-off-by: Dominic Jäger
---
I did not know that we have to run 'make update' when adding a new
language to i18n and have overseen a lot of errors as a consequence [0].
We might want to give (new) developers a hint at some place so that something
Looked through the patches, nice job catching all those. One thing inline.
On 9/2/19 11:19 AM, Fabian Ebner wrote:
Signed-off-by: Fabian Ebner
---
pvesm.adoc | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/pvesm.adoc b/pvesm.adoc
index 5300f50..c2be9fe
On 9/2/19 11:20 AM, Fabian Ebner wrote:
Signed-off-by: Fabian Ebner
---
vzdump.adoc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/vzdump.adoc b/vzdump.adoc
index fb1ac3d..e990982 100644
--- a/vzdump.adoc
+++ b/vzdump.adoc
@@ -78,7 +78,7 @@ consistency, the use of
Special care is taken not to overwrite any special flags, or ones
manually set on the VM by the user. We warn if a flag is overruled.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 26 +-
1 file changed, 25 insertions(+), 1 deletion(-)
diff --git a/PVE
Supports custom basemodels (model shown to QEMU, i.e. must be a default
model), vendors and (host-)phys-bits for VMs with large amounts of RAM
(see bug #2318).
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 32 +++-
1 file changed, 31 insertions(+), 1 deletion
MP command. This is how libvirt
queries supported flags for its "host-passthrough" CPU type.
query_supported_cpu_flags is thus rather slow and shouldn't be called
unnecessarily.
Signed-off-by: Stefan Reiter
---
Changes from RFC:
* Clearer regexes
* Use existing QMP infrastructure
* Add lock
e feedback if this approach works better than the cluster-cpu
one.
[0]: https://pve.proxmox.com/pipermail/pve-devel/2019-July/038268.html
[1]: e.g.:
cpu-model: custom-cpu-name
host-phys-bits 1
flags +aes;+avx;+avx2
basemodel kvm64
qemu: Stefan Reiter (1):
Trigger pve-api-updates on updat
A QEMU update can change supported CPU flags, thus we need to restart
API services (especially pvestatd) to refresh their cached values.
Signed-off-by: Stefan Reiter
---
debian/triggers | 1 +
1 file changed, 1 insertion(+)
create mode 100644 debian/triggers
diff --git a/debian/triggers b
Custom CPU types can be specified via the API, but to prevent arbitrary
ones we have to manually check if the given model exists (as default or
custom).
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 31 +--
1 file changed, 29 insertions(+), 2 deletions
pvestatd will read supported CPU flags once on startup (since these
never change during runtime, and QEMU updates trigger a service
restart), then broadcasts them as a key-value pair to the cluster.
Signed-off-by: Stefan Reiter
---
Changes from RFC:
* Cache joined value
* Add eval around QEMU
Signed-off-by: Stefan Reiter
---
This will definitely require some sort of versioning mechanism, otherwise CPU
definitions could be changed after starting a VM, thus breaking live-migration
by starting the migration-target with different parameters.
Hints, ideas, recommendations?
PVE/
as before.
Signed-off-by: Stefan Reiter
---
The list is only going to grow I suspect, especially with custom models
potentially being available in the future. Let's give the users some eye-candy
to more easily parse what they see :)
www/css/ext6-pve.css | 11 ++
www/manager6/form
Most of the code is adapted from KVComboBox, from which it inherited
before, with some generic code removed.
The templating code adds a header for every group transition it
encounters, thus providing support for arbitrary groups (e.g. custom
models in the future).
Signed-off-by: Stefan Reiter
t;AMD" or "intel" and see all related CPUs.
v4: Regular GridPanel with searching improvements
v3: Grouped grid-panel
v1/2: ComboBox with headers
manager: Stefan Reiter (2):
Make CPU Model Selector a searchable grid view
Allow CPUModelSelector to be searched by vendor as well
with value set in widget definition (would need to be
preferredValue), but we always call setValue() anyway (and if we don't,
value will be '', aka the default, which is correct too), so just remove
that from ProcessorEdit.js.
Signed-off-by: Stefan Reiter
---
www/manager6/form/CPUModelSelecto
Uses code from ExtJS 6.0.1 ComboBox.js to overwrite 'doLocalQuery',
adding a new property 'searchFields' which allows to specify fields from
the bound store which will all be searched (OR filter, i.e. the searched
value can appear in any column specified).
Signed-off-by: Stefan Reiter
---
form
Signed-off-by: Stefan Reiter
---
www/manager6/form/CPUModelSelector.js | 1 +
1 file changed, 1 insertion(+)
diff --git a/www/manager6/form/CPUModelSelector.js
b/www/manager6/form/CPUModelSelector.js
index dea6c44c..55532a0d 100644
--- a/www/manager6/form/CPUModelSelector.js
+++ b/www/manager6
Will implement all of this for v2, just a note inline.
On 9/6/19 1:42 PM, Fabian Grünbichler wrote:
On September 2, 2019 4:27 pm, Stefan Reiter wrote:
* query_understood_cpu_flags returns all flags that QEMU/KVM knows about
* query_supported_cpu_flags returns all flags that QEMU/KVM can use
Thanks again for the thorough review, I'll try to address everything
mentioned in v2. Some stuff inline for this patch in particular.
On 9/9/19 11:53 AM, Fabian Grünbichler wrote:
On September 2, 2019 4:27 pm, Stefan Reiter wrote:
Inherits from SectionConfig to provide base parsing
-by: Stefan Reiter
---
I'm personally not 100% convinced this grid approach makes it easier to parse
for a user, but I'd be fine with this being applied, if consensus is otherwise.
A search function like with our ISO selector for example would be nice too I
think, but I could not get that to work
On 9/16/19 9:46 AM, Fabian Ebner wrote:
Signed-off-by: Fabian Ebner
---
PVE/API2/Qemu.pm | 3 +++
1 file changed, 3 insertions(+)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index b30931d..ab5912c 100644
--- a/PVE/API2/Qemu.pm
+++ b/PVE/API2/Qemu.pm
@@ -3341,6 +3341,9 @@
Tried my hand at a proper code review, I like the patches in general.
On 9/13/19 3:16 PM, Aaron Lauterer wrote:
Signed-off-by: Aaron Lauterer
---
www/manager6/Makefile | 1 +
www/manager6/form/SpiceEnhancementSelector.js | 66 +++
2 files changed,
Akin to what you mentioned on the cover, there's a connection missing
between the "Graphic card" field and the SPICE ones - there is no clear
indication of *why* they are disabled by default in a new VM.
Maybe make it a section of some sort titled "SPICE"?
On 9/13/19 3:16 PM, Aaron Lauterer
On 9/13/19 3:16 PM, Aaron Lauterer wrote:
Signed-off-by: Aaron Lauterer
---
www/manager6/Utils.js| 18 ++
www/manager6/qemu/Options.js | 13 +
2 files changed, 31 insertions(+)
diff --git a/www/manager6/Utils.js b/www/manager6/Utils.js
index
Since we cannot easily proxy the vzdump command to other nodes, we
expect the client to call this on all nodes they want to run the backup
job on (and use standard proxyto=>'node').
Signed-off-by: Stefan Reiter
---
I wasn't too sure regarding permissions, so I left it at the rather str
Includes a "confirm" dialog to not accidentally run a potentially large backup
job.
Signed-off-by: Stefan Reiter
---
www/manager6/dc/Backup.js | 67 +++
1 file changed, 61 insertions(+), 6 deletions(-)
diff --git a/www/manager6/dc/Backup.js b/www/m
On 8/7/19 10:31 AM, Dominik Csapak wrote:
looks mostly ok (did not test it), but some comments inline
On 8/5/19 4:58 PM, Stefan Reiter wrote:
To allow masking to be correctly handled, we use a counter variable to
keep the mask visible until all tasks have been completed. Any errors
On 8/7/19 11:40 AM, Dominik Csapak wrote:
at this point, the user may have switched the panel already and it
might seem weird to get an error 'out of nowhere'...
you could check 'me.destroyed' (see extjs docs) if the current panel
still is valid and only show the errors if it is.
also, if
...and replace instead with command line argument.
Avoids a deprecation warning.
Signed-off-by: Stefan Reiter
---
PVE/API2/VZDump.pm | 3 ---
PVE/VZDump.pm | 2 +-
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/PVE/API2/VZDump.pm b/PVE/API2/VZDump.pm
index 8e409674..c2a6fab4
have a hook that does this).
[0]: https://kernel-team.pages.debian.net/kernel-handbook/ch-update-hooks.html
Signed-off-by: Stefan Reiter
---
Found a hook we can use :)
Ignore the previous patch to docs then, since the ESP refresh now happens
automatically. I'll send a new one later.
debian/pve
Mostly to avoid calling hooks directly in other scripts, future-proofing.
Signed-off-by: Stefan Reiter
---
bin/pve-efiboot-tool | 30 +++---
1 file changed, 23 insertions(+), 7 deletions(-)
diff --git a/bin/pve-efiboot-tool b/bin/pve-efiboot-tool
index a2ff4e1..d9bb829
Add a mention to 'pve-efiboot-tool refresh' in places where
'update-initramfs' is mentioned.
Signed-off-by: Stefan Reiter
---
local-zfs.adoc | 5 -
qm-pci-passthrough.adoc | 18 +++---
system-booting.adoc | 6 +++---
3 files changed, 18 insertions(+), 11 deletions
"ip address show" prints:
inet ADDR peer PEERADDR/32 scope global DEVICE
for pointtopoint interfaces. We're not interested in the netmask itself,
just make the Regex match.
Signed-off-by: Stefan Reiter
---
src/PVE/Network.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletio
This was previously gated to CLI only, but it causes a vzdump job
started with the newly introduced "Run Now" button to fail if it
includes VMIDs on other nodes.
Signed-off-by: Stefan Reiter
---
PVE/API2/VZDump.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
ine nodes
* Fix masking
* Fix pve-cluster-backup data model
* Split cleanup into seperate commit and include new variable for job_id format
manager: Stefan Reiter (3):
fix #1594: add "Run now" button to cluster backup page
fix #1594: allow skipping VMIDs on other nodes from API
backup
job.
Curiously, the "pve-cluster-backup" data model seems to have been broken
entirely. I'm not sure how it was working before, but the changes in
this patch need it fixed, so include that as well (use "mode" instead of
individual flags + "compress" is not a boolean).
Whitespace removal and consolidating VZDump's job id format into a
local variable.
Signed-off-by: Stefan Reiter
---
PVE/API2/Backup.pm| 25 +
www/manager6/dc/Backup.js | 10 +-
2 files changed, 14 insertions(+), 21 deletions(-)
diff --git a/PVE/API2
and fix "edit kernel command line" to include actual refresh command
Signed-off-by: Stefan Reiter
---
Technically a v2 to my previous pve-docs patch, but things have changed quite
a bit with the update-initramfs hooks.
qm-pci-passthrough.adoc | 3 ---
system-booting.ado
he GUI would then misrepresent this,
showing that all VMs will be backed up (when in reality some will be excluded)
or straight up break and show "exclude" mode again, with the backend still being
on "pool".
Signed-off-by: Stefan Reiter
---
www/manager6/dc/Backup.js | 2 ++
1 file chan
1 - 100 of 476 matches
Mail list logo
