[pve-devel] [PATCH installer v3 8/8] fix #4829: tui: bootdisk: expose new `arc_max` ZFS option for PVE installations

2023-10-31 Thread Christoph Heiss
To set the maximum value for arc_max accordingly, simply pass down `RuntimeInfo` directly instead of the disks array to the views. Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * fix ZFS_ARC_MIN_SIZE to be MiB rather than bytes Changes v2 -> v3: * renamed `ZFS_ARC_MIN_SIZE` -> `ZFS_AR

[pve-devel] [PATCH installer v3 5/8] fix #4829: proxinstall: expose new `arc_max` ZFS option for PVE installations

2023-10-31 Thread Christoph Heiss
Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * no changes Changes v2 -> v3: * no changes proxinstall | 15 +++ 1 file changed, 15 insertions(+) diff --git a/proxinstall b/proxinstall index 64c8bab..f1a3c02 100755 --- a/proxinstall +++ b/proxinstall @@ -1162,6 +1162,21 @

[pve-devel] [PATCH installer v3 2/8] tui: views: add optional suffix label for `NumericEditView`s

2023-10-31 Thread Christoph Heiss
Most of the churn here is due to changing the inner view from an `EditView` to a `LinearLayout`. Also prompted the introduction of two small helpers .inner() and .inner_mut() to simplify things everywhere else in the view. Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * no changes Chang

[pve-devel] [PATCH installer v3 1/8] run env: add comment for query_total_memory()

2023-10-31 Thread Christoph Heiss
This is mainly to explicitly document the unit of its return value. No functional changes. Signed-off-by: Christoph Heiss --- Changes v2 -> v3: * new patch Proxmox/Install/RunEnv.pm | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Proxmox/Install/RunEnv.pm b/Proxmox/Install/RunEnv.pm in

[pve-devel] [PATCH installer v3 7/8] fix #4829: tui: setup: add new ZFS `arc_max` option

2023-10-31 Thread Christoph Heiss
Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * updated comment for ZfsBootdiskOptions::defaults_from() accordingly Changes v2 -> v2: * documented the unit of the return value of default_zfs_arc_max() proxmox-tui-installer/src/options.rs| 57 +++-- proxmox-tui

[pve-devel] [PATCH installer v3 6/8] fix #4829: test: add tests for new zfs_arc_max calculations

2023-10-31 Thread Christoph Heiss
Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * no changes Changes v2 -> v3: * no changes Makefile| 3 ++ debian/control | 1 + test/Makefile | 10 ++ test/zfs-arc-max.pl | 81 + 4 files changed, 95 insertions(+

[pve-devel] [PATCH installer v3 4/8] fix #4829: install: add new ZFS `arc_max` setup option

2023-10-31 Thread Christoph Heiss
Signed-off-by: Christoph Heiss --- Changes v1 -> v2: * no changes Changes v2 -> v3: * better documented calculation and renamed some variables to reflect its unit (thanks Thomas!) * moved modprobe config setup into separate sub Proxmox/Install.pm| 14 Proxmox/Inst

[pve-devel] [PATCH installer v3 3/8] tui: bootdisk: simplify product handling by passing the config directly

2023-10-31 Thread Christoph Heiss
No functional changes. Signed-off-by: Christoph Heiss --- Changes v2 -> v3: * new patch proxmox-tui-installer/src/views/bootdisk.rs | 50 - 1 file changed, 29 insertions(+), 21 deletions(-) diff --git a/proxmox-tui-installer/src/views/bootdisk.rs b/proxmox-tui-installer/

[pve-devel] [PATCH installer v3 0/8] fix #4829: set up lower default limit for ZFS ARC in installer

2023-10-31 Thread Christoph Heiss
Fixes #4829. Introduces a new ZFS install option `arc_max` (aptly named after the corresponding module option `zfs_arc_max`). For PVE installations, this can be adjusted when creating a ZFS RAID under "Advanced Options". The default value is choosen as 10% of system memory, clamped to between 64 M

[pve-devel] [PATCH manager 3/3] pvereport: add sdn config directory

2023-10-31 Thread Aaron Lauterer
the /etc/pve/sdn directory contains the config files, not just what they translate to in interface configs (/etc/network/interfaces.d/snd). The current way will also include dotifiles that may contain the current/running state. Which can be useful to troubleshoot. Signed-off-by: Aaron Lauterer -

Re: [pve-devel] [PATCH pve-network] Fix #4917: evpn: forbid vlan-aware bridge

2023-10-31 Thread Wolfgang Bumiller
On Fri, Oct 27, 2023 at 01:53:28PM +0200, Alexandre Derumier wrote: > Do it on vnet update instead throwing a warning at config generation. > > Signed-off-by: Alexandre Derumier > --- > src/PVE/Network/SDN/Zones/EvpnPlugin.pm | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --gi

Re: [pve-devel] [PATCH qemu-server] fix #4957: add vendor and product information passthrough for SCSI-Disks

2023-10-31 Thread Wolfgang Bumiller
On Mon, Oct 30, 2023 at 05:30:15PM +0100, Thomas Lamprecht wrote: > I mean, the properties are relatively straight forward, but some commit > message would be still nice to have – e.g., how did you check if the values > propagate into the guest, can this > > On 25/10/2023 14:37, Hannes Duerr wrote

[pve-devel] applied: [PATCH pve-ceph] fix compatibility with CPUs not supporting SSE 4.1 instructions

2023-10-31 Thread Thomas Lamprecht
On 18/09/2023 17:46, Stefan Hanreich wrote: > One of our users ran into issues with running Ceph on older CPU > architectures [1]. This is apparently due to a bug in gcc-12 that > leads to SSE 4.1 instructions always being executed rather than > dynamically dispatching functions using those instruc

[pve-devel] [PATCH acme v3 0/5] fix #4497: add support for external account bindings

2023-10-31 Thread Folke Gleumes
Changes since v2: * reverted the new_account abi to be non breaking Changes since v1: * fixed nit's * expanded meta endpoint by all return values defined in the rfc * expanded new_account signature by field for eab credentials * allow for eab even if not required This patch series adds funct

[pve-devel] [PATCH manager v3 4/5] fix #4497: cli/acme: detect eab and ask for credentials

2023-10-31 Thread Folke Gleumes
Since external account binding is advertised the same way as the ToS, it can be detected when creating an account and asked for if needed. Signed-off-by: Folke Gleumes --- No changes in v3 PVE/CLI/pvenode.pm | 26 -- 1 file changed, 24 insertions(+), 2 deletions(-) diff

[pve-devel] [PATCH manager v3 3/5] api/acme: deprecate tos endpoint in favor of meta

2023-10-31 Thread Folke Gleumes
The ToS endpoint ignored data that is needed to detect if EAB needs to be used. Instead of adding a new endpoint that does the same request, the tos endpoint is deprecated and replaced by the meta endpoint, that returns all information returned by the directory. Signed-off-by: Folke Gleumes --- N

[pve-devel] [PATCH manager v3 5/5] ui/acme: switch to new meta endpoint

2023-10-31 Thread Folke Gleumes
Besides the switch from tos to meta endpoint, this fixes a visual bug, where the 'Accept TOS' button would show up, even if no ToS was needed. Signed-off-by: Folke Gleumes --- Changes since v2: fixed tabs/spaces www/manager6/node/ACME.js | 12 1 file changed, 8 insertions(+), 4 del

[pve-devel] [PATCH acme v3 1/5] fix #4497: add support for external account bindings

2023-10-31 Thread Folke Gleumes
implementation acording to rfc8555 section 7.3.4 Signed-off-by: Folke Gleumes --- Changes since v2: Transport eab credentials in the info hash, but don't reuse it as payload. Instead, needed values are extracted and, if needed, transformed into a new hash. While this limits how the info hash can

[pve-devel] [PATCH manager v3 2/5] fix #4497: acme: add support for external account bindings

2023-10-31 Thread Folke Gleumes
Signed-off-by: Folke Gleumes --- No changes in v3 PVE/API2/ACMEAccount.pm | 27 ++- 1 file changed, 26 insertions(+), 1 deletion(-) diff --git a/PVE/API2/ACMEAccount.pm b/PVE/API2/ACMEAccount.pm index b790843a..ec4eba24 100644 --- a/PVE/API2/ACMEAccount.pm +++ b/PVE/API2

Re: [pve-devel] [PATCH v2 guest-common 1/1] Add foreach_passthrough_device

2023-10-31 Thread Thomas Lamprecht
On 30/10/2023 14:12, Wolfgang Bumiller wrote: > On Tue, Oct 24, 2023 at 02:55:54PM +0200, Filip Schauer wrote: >> + >> +$func->($key, "dev/bus/usb/$bus_id/$device_id", @param); > > So this will do... something :-) > But unfortunately it won't deal with the *interesting* bits. > > So while