xist.
Signed-off-by: Dominik Csapak
---
I looked in the backend, we save the 'account' object verbatim from the
api response, and the RFC [0] says it should have a contact field, but
it's not there in my tests with the staging + normal api, so not sure
if this is the right fix.
0: htt
On Tue, 30 Sep 2025 10:48:52 +0200, Shan Shaji wrote:
> The app only uses TLS/SSL through iOS platform APIs, which are already
> exported by Apple, and is therefore classified under 5A992/5D992 (Mass
> Market [1]). Since march 29, 2021, mass market software no longer
> requires a self-classificatio
On 9/24/25 12:36 PM, Shan Shaji wrote:
After upgrading to Flutter v3.35, the dart analyze command reported
several issues. This patch series resolves them by applying `dart fix`
to automatically correct issues where possible. For cases where
automated fixes were not available, I manually repla
On 9/24/25 12:37 PM, Shan Shaji wrote:
Starting form language version 3.9, the flutter constrain upper bound is
now respected in the root package. Setting the flutter constraint can be
usefull to ensure everyone will use the same SDK.
Signed-off-by: Shan Shaji
---
pubspec.lock | 2 +-
pub
On 9/24/25 3:05 PM, Shan Shaji wrote:
The app was not honoring the user installed certificate and was still
throwing `HandShakeException` when using `IOClient`. Inorder to fix the
issue used the `cupertino_http` package. This patch series only includes
the changes specific to iOS. Android rela
I'm not a lawyer, but are you sure we are compliant with this?
the https client will use encryption, so it's not like we don't use
encryption at all. (Especially with the cupertino_http package)
I guess this is exempt, but having some reference for that (aside
some medium/reddit post) would be
On 9/4/25 11:25 AM, Shan Shaji wrote:
On the options page for VMs and CTs it was easy to change the
configs by mistake. To avoid that, added a lock/unlock button
on top of the screen. The toggle buttons will only be enabled
if the button is clicked.
Suggested-by: Thomas Lamprecht
Signed-off-
ide + class config would be merged
there, but it seems that the chart initializes the axes already before
the initComponent.
Using 'setSegmenter' later on is also not possible really, because for
some reason this only changes the top-most point in the axis labels, but
not the intermedi
On 9/4/25 12:09 PM, Shan Shaji wrote:
The app was not honoring the user installed certificate and was still
throwing `HandShakeException` when using `IOClient`. Inorder to fix the
issue used the `cronet_http` package. This patch series only includes
the changes specific to android.
changes si
Hi,
just to let you know, we did not forget this series.
I'll look at it this week, thanks!
Best regards
Dominik
On 8/27/25 9:35 AM, Johannes Altmanninger via pve-devel wrote:
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.prox
On 9/17/25 4:08 PM, Fiona Ebner wrote:
Reported-by: Daniel Koć
Signed-off-by: Fiona Ebner
---
Reported here, but the reporter's mail address was hidden/mangled:
https://lore.proxmox.com/pve-devel/20250731200200.0a3e99...@firstgate.proxmox.com/
src/panel/Certificates.js | 2 +-
1 file chan
after applying the other series, this does not seem to apply anymore,
can you send a rebased one please?
thanks
On 9/8/25 4:11 PM, Shan Shaji wrote:
The app was not honoring the user installed certificate and was still
throwing `HandShakeException` when using `IOClient`. Inorder to fix the
iss
On 8/29/25 1:49 PM, Shan Shaji wrote:
This patch series includes changes from the following patch [0].
Addition of HELP and LEGAL section was done on top of it.
[0] -
https://lore.proxmox.com/pve-devel/20250801100051.83035-1-s.sh...@proxmox.com/
pve_flutter_frontend:
Shan Shaji (3):
ref
On 8/18/25 12:01 PM, Shan Shaji wrote:
On iOS the noVNC console option was not available this patch series
includes the changes to enable the console view. Also seperate the onTap
implementation to a seperate function as the current implementation is
repeating.
Additionaly the console view wa
performance on initial update)
Signed-off-by: Dominik Csapak
---
this should probably also be backported to stable-bookworm
www/manager6/tree/ResourceTree.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/tree/ResourceTree.js
b/www/manager6/tree/ResourceTree.js
index
check ('boolean') failed - got 'true'`
To fix this, also apply this conversion in the case when the default_key
is not given.
Signed-off-by: Dominik Csapak
---
src/PVE/JSONSchema.pm | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/PVE/JSONSchema.pm b/src/PVE/JSON
is, remote the 'text' property from the to checked ones, and
add all the properties that are used to compile the text one.
This reduces the time of updateTree in my test-setup (~1 guests)
when nothing changed from ~100ms to ~15ms and reduces scroll stutter
during such an update.
On 9/5/25 3:49 PM, Aaron Lauterer wrote:
On 2025-09-05 14:06, Dominik Csapak wrote:
the idea was that we get any of the 'new' versions on lookup, but that
lead to iterating through possibly all keys. Since that was called for
each resource in e.g. /cluster/resources api call, t
ons are more likely to
only come in small numbers. (Still have to look into if we can avoid
that behavior there too)
This improves the time spend in 'registerNode' (in my ~1 guests test
setup) from 4,081.6 ms to about 2.7ms.
Signed-off-by: Dominik Csapak
---
www/manager6/tree/Res
mem_usage'
in my test setup (~1 non-running vms) from 4,408.2 ms to 12.4 ms.
(Measured with the `Performance` tab in Chromium)
Signed-off-by: Dominik Csapak
---
www/manager6/Utils.js | 12
www/manager6/data/ResourceStore.js | 20
2 file
.
The overall perforamnce improvement i see on my local test cluster with
~1 vms is from
API ~22s + GUI ~8.5s
to
API 500ms + GUI 500ms
which is a ~30x speedup (since it's mostly quadratic code that is
removed, this is more noticable the more resources there are ofc)
Dominik Csap
ently only valid key here which
makes this lookup much cheaper.
In my test setup with ~1 guests, it reduces the time for a call
to /cluster/resources from ~22s to ~400ms
Signed-off-by: Dominik Csapak
---
NOTE: this is only a workaround and Aaron is working to making this
whole part unnecessary, b
On 8/26/25 12:11 PM, Thomas Lamprecht wrote:
On 12/08/2025 11:59, Dominik Csapak wrote:
by default, pci devices will be bound to 'vfio-pci' driver and reset.
For most devices this is necessary, but there are a few exceptions,
e.g.:
* some mellanox nics have support for
On 8/14/25 16:34, Fiona Ebner wrote:
As reported in the community forum [0], the 'Edit' button would stay
disabled. Use a proxmoxButton to fix it, which installs a monitor for
selectionchange.
[0]: https://forum.proxmox.com/threads/169258/post-792029
Fixes: fb289b0d ("ha: affinity rules: mak
hi,
thanks for sending the patch again without dart format changes (mostly,
see my comment on the last hunk)
while the changes themselves look ok to me, i would have liked it more
for the different changes to be different patches
e.g. adding the appbar is ok, but does not have anything to do wi
(I'd prefer starting out with front end filters)
* failure mode resolution (I'd wait until someone requests that)
* token handling (probably not necessary since we do check the
permissions upfront for the correct token.)
Signed-off-by: Dominik Csapak
---
changes from v1:
* rebased o
changes look good to me
something I noticed (that's not your fault at all) which irks me a bit
is that we don't really have any dependency management between
proxmox_login_manager and pve_flutter_frontend
as in, we don't ever bump the version of proxmox_login_manager
(and the dart_api_client of
On 8/13/25 10:29, Shan Shaji wrote:
The noVNC console was disabled in iOS and was only available in
android.
To fix the issue enabled the noVNC console view on iOS devices. The
changes also includes a refactor of the function responsible for
displaying the webview. Additionally, an `AppBar` h
otherwise the refresh logic will 'jump' back to the beginning of the
line if the user has scrolled horizontally.
Signed-off-by: Dominik Csapak
---
should be cherry-pickable for stable-bookworm too if desired
src/panel/JournalView.js | 5 +++--
src/panel/LogView.js | 2 +-
2 fil
ur side, introduce the 'keep-driver'
option for 'hostpciX', which will not rebind/reset the device.
Signed-off-by: Dominik Csapak
---
sending as RFC, since i'm not sure if we want to go this (generic)
approach, or if we e.g. want to make special configs/cases for driver we
just to note, this would be an ideal candidate for regression tests.
also pmg has a slightly different method to check this, that we should
unify with this one.
I'll look into both more deeply in the coming days/weeks when I have
time, but I wanted to create a fix for this first.
e
slashes/whitespace/semicolons and other things browser vendors might use
in their user agent strings.
Signed-off-by: Dominik Csapak
---
PVE/Service/pveproxy.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/PVE/Service/pveproxy.pm b/PVE/Service/pveproxy.pm
index 6ee4a141
On 5/20/25 14:37, Shan Shaji wrote:
All apps that are submitted to Apple app store must provide a launch
screen with an Xcode storyboard. Previously the launch screen were
showing only a white background. This commit fixes it by adding the
proxmox logo and setting the storyboard view background c
On 8/4/25 13:07, Shan Shaji wrote:
In iOS, there was a blue background color visible beneath the bottom
bar, which caused the app to not utilize the full screen height.
Additionally, the colors of the app bar and the status bar
differed when the user began scrolling.
This issue was due to the
On 7/16/25 13:02, Shan Shaji wrote:
For iOS, the bundle identifier and display name were incorrect. The
issue was fixed by updating the bundle identifier to
`com.proxmox.proxmox-ve-companion` and the display name to "Proxmox VE".
The bundle identifer was changed to `com.proxmox.proxmox-ve-com
On 7/17/25 12:43, Shan Shaji wrote:
When running the app on `iOS`, by default flutter runs the
`pod install` command if any of the following conditions [0] are true.
Since there were no existing pod related files, this command generates
the `cocoapods` related files to run the app on iOS.
Add
Hi,
please correct me if I'm wrong, but wouldn't this change drastically
change the graphs on the web ui ? (did not test it though)
when e.g. a storage is offline, it will have intermediary empty values?
so instead of having
[
{ time: 1, value: 1},
{ time: 2 },
{ time: 3, value: 2}
]
where we
useful when we want to deliver pre-compressed files e.g. for wasm
Signed-off-by: Dominik Csapak
---
src/PVE/APIServer/AnyEvent.pm | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/PVE/APIServer/AnyEvent.pm b/src/PVE/APIServer/AnyEvent.pm
index a5f6681..59bc184 100644
--- a/src/PVE
so it's more like the 'used' color. It's a blueish shade from the
default extjs color scheme.
Signed-off-by: Dominik Csapak
---
www/manager6/node/Summary.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/node/Summary.js b/www/manager6/n
so it's more subtle versus the current gray line.
I used the normal green color from 'total' but increased the lightness
value by 10 points (in HSL color space) to make the color brighter.
Signed-off-by: Dominik Csapak
---
www/manager6/node/Summary.js | 2 +-
1 file changed, 1
instead of an area graph. Give it the same color as the 'hostmem' one
from the qemu guests (a neutral color that draws not too much
attraction).
While at it, order it last, since it was just ordered second so the
overlapping colors would not clash.
Signed-off-by: Dominik Csapak
---
re
mply check if the field is the one we're looking for
Signed-off-by: Dominik Csapak
---
this is necessary for the next patch to have correct tooltips
src/panel/RRDChart.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/panel/RRDChart.js b/src/panel/RRDChart.js
index c
n by default' logic.
While at it, use title case for the legend.
Originally-by: Thomas Lamprecht
Signed-off-by: Dominik Csapak
---
www/manager6/data/model/RRDModels.js | 28
www/manager6/node/Summary.js | 12 +++-
www/manager6/panel/GuestSummar
so one can see how much is available as an explicit graph.
The color is the same as the total, as most other colors produce weird
looking result since the graphs are transparent and overlay each other.
Signed-off-by: Dominik Csapak
---
not a huge fan of this, since it always runs reverse to the
the stacked graphs had more information, namely how much memory was
available and how much was used without the ZFS ARC. So put this
information into the tooltip where the users most likely want to see it.
Signed-off-by: Dominik Csapak
---
not sure if this is wanted/needed, since we did not have
vey what it actually does
Signed-off-by: Dominik Csapak
---
www/manager6/sdn/fabrics/NodeEdit.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/sdn/fabrics/NodeEdit.js
b/www/manager6/sdn/fabrics/NodeEdit.js
index 161917cc..d27515ff 100644
--- a/www/manager6/sdn
tion can be
trusted.
Signed-off-by: Dominik Csapak
---
changes from v1:
* drop hex dependency
Cargo.toml| 1 +
proxmox-http/Cargo.toml | 5 +++
proxmox-http/src/lib.rs | 5 +++
proxmox-http/src/tls.rs | 84 +++
proxmox-openid/Cargo.toml | 2
om v1:
* rebase on master (drops one patch)
* drop hex dependency
proxmox:
Dominik Csapak (2):
http: factor out openssl verification callback
client: use proxmox-http's openssl verification callback
Cargo.toml | 1 +
proxmox-client/Cargo.toml| 2 +-
proxmox-clien
has higher priority than openssls
validation.
Signed-off-by: Dominik Csapak
---
Cargo.toml| 2 +-
pbs-client/src/http_client.rs | 151 ++
2 files changed, 62 insertions(+), 91 deletions(-)
diff --git a/Cargo.toml b/Cargo.toml
index ea4133283
precedence over an explicit
fingerprint.
Signed-off-by: Dominik Csapak
---
proxmox-client/Cargo.toml| 2 +-
proxmox-client/src/client.rs | 48 ++--
2 files changed, 9 insertions(+), 41 deletions(-)
diff --git a/proxmox-client/Cargo.toml b/proxmox-client/Cargo.toml
no functional change intended, since the callback there should implement
the same behavior.
With this, we can drop the dependency on itertools.
Signed-off-by: Dominik Csapak
---
Cargo.toml | 3 +--
src/main.rs | 67 +
2 files changed, 28
now also to pve-devel, did a typo when writing the list address...
On 7/22/25 13:43, Dominik Csapak wrote:
this is necessary for some esxi versions to be added as PVE storage on trixie
otherwise it fails with:
thread 'main' panicked at
/usr/share/cargo/registry/hyper-util-0.1.12/
:
kvm: -device vfio-pci,host=:21:00.4,id=hostpci0,bus=pci.0,addr=0x10: vfio
:21:00.4: error getting device info: Input/output error
Signed-off-by: Dominik Csapak
---
...me-PM-on-devices-that-don-t-support-.patch | 57 +++
1 file changed, 57 insertions(+)
create mode
:
kvm: -device vfio-pci,host=:21:00.4,id=hostpci0,bus=pci.0,addr=0x10: vfio
:21:00.4: error getting device info: Input/output error
Signed-off-by: Dominik Csapak
---
this can probably be cherry picked for trixie
...me-PM-on-devices-that-don-t-support-.patch | 57 +++
1
IMHO this and the last patch should be combined, since they don't really
make sense without each other..
one comment inline:
On 7/15/25 16:32, Aaron Lauterer wrote:
First by using the new memhost field if guest is of type qemu and the field is
numerical.
Second by checking in the render functio
Reviewed-by: Dominik Csapak
On 7/15/25 16:32, Aaron Lauterer wrote:
They were missing and just showed the actual field names.
Signed-off-by: Aaron Lauterer
---
www/manager6/node/Summary.js | 1 +
www/manager6/panel/GuestSummary.js | 2 ++
2 files changed, 3 insertions(+)
diff
mhmm not too sure about this. IMHO this info belongs in the
documentation instead of in a tooltip, but I get why we
want to maybe have an explanation for these easily reachable.
First, I'd use the panel tool icons here instead of font awesomes,
like e.g. the undo zoom and close button (from the w
two comments inline:
On 7/15/25 16:32, Aaron Lauterer wrote:
We switch the memory graph to a stacked area graph, similar to what we
have now on the node summary page.
Since the order is important, we need to define the colors manually, as
the default color scheme would switch the colors as we u
Since this is getting rather crowded, we should probably think about
how we could redesign this panel to save a bit of space
(e.g. show the status and ha status on one line, moving the node
info somewhere else, ...)
FWICT this is not that big of a problem (yet)
Reviewed-by: Dominik Csapak
On
As with the patch before, we might want to refactor the colors somewhere
Otherwise:
Reviewed-by: Dominik Csapak
On 7/15/25 16:32, Aaron Lauterer wrote:
From: Folke Gleumes
Pressures are indicatios that processes needed to wait for their
resources. While 'some' means, that s
We should probably refactor the colors to Proxmox.Utils if we'll use
them more often in the future, but aside from that:
Reviewed-by: Dominik Csapak
On 7/15/25 16:32, Aaron Lauterer wrote:
To display the used memory and the ZFS arc as a separate data point,
keeping the old line overla
comments inline:
On 7/15/25 16:32, Aaron Lauterer wrote:
Memory columns will be used in an area graph, which cannot handle
gaps directly. Therefore we set the default value to 'null'. This makes
it easier to handle the tooltip when there is no data.
We calculate memused to subtract the arcsize.
Reviewed-by: Dominik Csapak
On 7/15/25 16:31, Aaron Lauterer wrote:
This makes targeting the undo button more stable in situations where it
might not be the 0 indexed item in the tools.
Signed-off-by: Aaron Lauterer
---
src/panel/RRDChart.js | 3 ++-
1 file changed, 2 insertions(+), 1
One nit inline, but aside from that:
Reviewed-by: Dominik Csapak
On 7/15/25 16:31, Aaron Lauterer wrote:
this way we can keep the current behavior, but also make it possible to
finely control a series if needed. For example, if we want a stacked
graph, or just a line without fill
On 7/15/25 14:51, Thomas Lamprecht wrote:
On Tue, 15 Jul 2025 14:08:51 +0200, Dominik Csapak wrote:
If we omit the 'efitype' property, the legacy 2m images get used, so
explicitly use the 4m variant.
Applied, to master and stable-8, thanks!
We probably should change the deaful
If we omit the 'efitype' property, the legacy 2m images get used, so
explicitly use the 4m variant.
Signed-off-by: Dominik Csapak
---
www/manager6/window/GuestImport.js | 1 +
1 file changed, 1 insertion(+)
diff --git a/www/manager6/window/GuestImport.js
b/www/manager6/window/Gues
wrong with
'com.proxmox.pveFlutterFrontend' ?
are there any guidelines or rules in the app store that would prevent us from
using that?
I'm not saying we can't change this, but it would be good to record the reason
at least...
Aside from that the patch looks good to me
On 7/15/25 11:41, Stefan Hanreich wrote:
On 7/15/25 11:30, Dominik Csapak wrote:
On 7/15/25 11:21, Stefan Hanreich wrote:
Was formulating my response to v1, but you were too fast with sending a
v2 for me. I think we need to also consider the case where we pinned and
replaced the names in /e/n
.proxmox.com/?p=pve-common.git;a=blob;f=src/PVE/INotify.pm;h=dbad08c3fc5494403a061b9d77d587314477a9f5;hb=HEAD#l897
On 7/15/25 11:07, Dominik Csapak wrote:
in api listing for all interfaces, and in the GET call for an individual
interface.
For intefaces that are already using an 'altname
superseded by v2:
https://lore.proxmox.com/all/20250715090749.1608768-1-d.csa...@proxmox.com/T/#t
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
idget-toolkit:
Dominik Csapak (1):
network: optionally show alternative interface names
src/node/NetworkEdit.js | 19 +++
src/node/NetworkView.js | 20
2 files changed, 39 insertions(+)
pve-manager:
Dominik Csapak (1):
api/ui: show/return altern
might be confused if the bridge ports
and interface names don't correlate.
enables the additional alternative names column in the network view on
the gui.
Signed-off-by: Dominik Csapak
---
changes from v1:
* also show alternative names for ifaces with altname stanza
(include legacy iface in
add a new (hidden by default) column for the interface names, and show
them when editing an existing interface with such alternative names.
Signed-off-by: Dominik Csapak
---
no changes from v1
src/node/NetworkEdit.js | 19 +++
src/node/NetworkView.js | 20
2
e:
Am 15.07.25 um 10:36 schrieb Dominik Csapak:
On 7/15/25 10:17, Thomas Lamprecht wrote:
calculated for *each*, so e.g. if i have
nicX, nicY and nicZ all point to the same device, we'd have to make the
altname list generation
more complicated
IMHO it would be ok to always
On 7/15/25 10:17, Thomas Lamprecht wrote:
Am 15.07.25 um 09:41 schrieb Dominik Csapak:
On 7/14/25 16:31, Thomas Lamprecht wrote:
Fwiw, it's not really wrong, one is the name in /e/n/i and the others
are interface names per `ip link`, but it naturally can be a bit
confusing as is.
Depends
On 7/14/25 16:31, Thomas Lamprecht wrote:
Am 14.07.25 um 15:53 schrieb Dominik Csapak:
On 7/14/25 15:49, Christoph Heiss wrote:
Tested the series, came across two thing:
Given e.g. the following interface:
11: ens8: mtu 1500 qdisc [..]
link/ether 01:02:03:04:05:06 brd
On 7/15/25 08:52, Thomas Lamprecht wrote:
Am 15.07.25 um 05:55 schrieb nansen.su:
[snip]>> +
+sub type {
+return 'opentelemetry';
+}
+
+sub properties {
+return {
+'otel-protocol' => {
I'm fine with the plugin specific otel prefix, but @Dominik: these here might be
a good fit
'sort keys' here, i'll wait until
tomorrow for a v2 if there are other remarks
On Mon Jul 14, 2025 at 12:42 PM CEST, Dominik Csapak wrote:
This series goes on top of Stefan Hahnreich's last series[0], but it seems
there is still some things missing from INotify (to e.g.
By enabling the import button for qcow2/vmdk/raw files, and showing a
window with a VMID selector and the disk edit panel.
Change the edit panel so that when we give an explicit volume id
directly, we don't let the user select one. Instead it show it in a
displayfield.
Signed-off-by: Do
by adding a new 'import' button in the disk tab, which adds the same
input panel as the one we have when doing an 'Import Hard Disk' for an
existing VM.
partially fixes #2424
Signed-off-by: Dominik Csapak
---
www/manager6/panel/MultiDiskEdit.js | 36 +++
#2424
Signed-off-by: Dominik Csapak
---
www/manager6/qemu/HDEdit.js | 67 +++
www/manager6/qemu/HardwareView.js | 6 +++
2 files changed, 65 insertions(+), 8 deletions(-)
diff --git a/www/manager6/qemu/HDEdit.js b/www/manager6/qemu/HDEdit.js
index c3a5d0af..8cc99
-d.csa...@proxmox.com/
Dominik Csapak (3):
ui: qemu hd edit: allow importing a disk from the import storage
ui: vm create wizard: allow importing disks
ui: import storage content: allow importing of vm disk images
www/manager6/panel/MultiDiskEdit.js | 36 -
www/manager6/qemu/HDEdit.js
add a new (hidden by default) column for the interface names, and show
them when editing an existing interface with such alternative names.
Signed-off-by: Dominik Csapak
---
src/node/NetworkEdit.js | 19 +++
src/node/NetworkView.js | 20
2 files changed, 39
in api listing for all interfaces, and in the GET call for an individual
interface.
If we don't show them, the user might be confused if the bridge ports
and interface names don't correlate.
enables the additional alternative names column in the network view on
the gui.
Signed-off-b
hat is different from
the shutdown one.
I'm a bit unsure myself, but *if* we want to have a different icon here than for
the guest shutdown, consider this patch:
Reviewed-by: Dominik Csapak
On 7/2/25 11:52, Aaron Lauterer wrote:
this makes it optically different to the guest shutdown and hopefull
ready return the
altnames.
Note that the pve-manager patch depends on the widget-toolkit patch
0:
https://lore.proxmox.com/pve-devel/20250709194526.560709-2-s.hanre...@proxmox.com/
proxmox-widget-toolkit:
Dominik Csapak (1):
network: optionally show alternative interface names
src
so users know they can override the concurrency settings.
Signed-off-by: Dominik Csapak
---
qmrestore.adoc | 5 +
1 file changed, 5 insertions(+)
diff --git a/qmrestore.adoc b/qmrestore.adoc
index 37f185a..7f0b4a3 100644
--- a/qmrestore.adoc
+++ b/qmrestore.adoc
@@ -27,6 +27,11 @@ Restore
for details see the proxmox-backup-qemu patch.
docs patch is there to document the environment variables.
proxmox-backup-qemu:
Dominik Csapak (1):
restore: make chunk loading more parallel
src/restore.rs | 80 +++---
1 file changed, 63 insertions
e-de...@lists.proxmox.com/
Signed-off-by: Dominik Csapak
Based-on-patch-by: Adam Kalisz
---
src/restore.rs | 80 +++---
1 file changed, 63 insertions(+), 17 deletions(-)
diff --git a/src/restore.rs b/src/restore.rs
index 5a5a398..6cafd78 100644
--- a
ff-by: Dominik Csapak
---
While testing upload errors to see if we can get a better error message
than 'Error 0 occurred', i noticed that in chrome we fail here to show
the valid error message.
Note that this does not improve the 'Error 0' message, only cases where
the browser
On 7/10/25 14:48, Dominik Csapak wrote:
[snip]
Just for the record i also benchmarked a slower system here:
6x16 TiB spinners in raid-10 with nvme special devices
over a 2.5 g link:
current approach is ~61 MiB/s restore speed
with my patch it's ~160MiB/s restore speed with not much increa
On 7/8/25 17:08, Adam Kalisz wrote:
On Tue, 2025-07-08 at 12:58 +0200, Dominik Csapak wrote:
On 7/8/25 12:04, Adam Kalisz wrote:
[snip]
Especially during a restore the speed is quite important if you
need to hit Restore Time Objectives under SLAs. That's why we were
targeting 1 GBp
7;async-channel' + fixed number
of tasks ?)
Thanks
Dominik
Thank you
Adam
On Tue, 2025-07-08 at 10:49 +0200, Dominik Csapak wrote:
by using async futures to load chunks and stream::buffer_unordered to
buffer up to 16 of them, depending on write/load speed, use tokio's
task
sp
the blocking threads and uses the (actually always
used) underlying async implementation for reading remote chunks.
0:
https://lore.proxmox.com/pve-devel/mailman.719.1751052794.395.pve-de...@lists.proxmox.com/
Signed-off-by: Dominik Csapak
Based-on-patch-by: Adam Kalisz
---
resend to both pve and p
Sorry, i sent the patch to the wrong list, this was originally meant for
pve-devel instead. Should i resend it there?
On 7/8/25 10:49, Dominik Csapak wrote:
by using async futures to load chunks and stream::buffer_unordered to
buffer up to 16 of them, depending on write/load speed, use tokio&#
fixed, by starting background
tokio tasks in the future itself, so they will run to completion even when
we're still writing, that should give better improvements.
(I'll also update my benchmark data)
Thanks
Dominik
Best regards
Adam
On Mon, 2025-07-07 at 10:14 +0200, Dominik
ease the blocking threads and uses the (actually always
used) underlying async implementation for reading remote chunks.
0:
https://lore.proxmox.com/pve-devel/mailman.719.1751052794.395.pve-de...@lists.proxmox.com/
Signed-off-by: Dominik Csapak
Based-on-patch-by: Adam Kalisz
---
@Adam could y
On 7/4/25 16:40, Shan Shaji wrote:
On Thu Jul 3, 2025 at 4:00 PM CEST, Dominik Csapak wrote:
+ backgroundColor: Theme.of(context).colorScheme.surfaceContainer,
this change seems to be unrelated?
This was intended. The change was made to match the backgroundColor of
`ProxmoxLoginSelector
On 7/4/25 09:16, Shan Shaji wrote:
Thanks for the review. I will update the changes according to the
reviews. I have some doubts that i have added as inline comments.
[snip]
diff --git a/lib/constants/pve_links.dart b/lib/constants/pve_links.dart
new file mode 100644
index 000..196b58c
superseded by v2:
https://lore.proxmox.com/pve-devel/20250704061852.251189-1-d.csa...@proxmox.com/
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
1 - 100 of 1747 matches
Mail list logo
