From: Hannes Laimer <h.laimer@proxmox.com>
To: Proxmox VE development discussion <pve-devel@lists.proxmox.com>,
	Stefan Hanreich <s.hanreich@proxmox.com>
Subject: Re: [pve-devel] [PATCH proxmox-ve-rs 2/6] frr: add deserialization types for openfabric and ospf
Date: Mon, 3 Nov 2025 15:50:32 +0100	[thread overview]
Message-ID: <e3f18921-5137-4863-b4c1-ed296512d888@proxmox.com> (raw)
In-Reply-To: <20251030154851.540408-5-s.hanreich@proxmox.com>
I think some tests would make sense here
On 10/30/25 16:48, Stefan Hanreich wrote:
> From: Gabriel Goller <g.goller@proxmox.com>
> 
> These are used to deserialize `vtysh` command outputs. The output is in
> json if the `json` parameter is appended to the command. Currently the
> following commands are parsed:
> 
>   * show openfabric neighbor
>   * show ip ospf neighbor
>   * show ip route <protocol>
> 
> Signed-off-by: Gabriel Goller <g.goller@proxmox.com>
> Signed-off-by: Stefan Hanreich <s.hanreich@proxmox.com>
> ---
>   proxmox-frr/Cargo.toml           |   1 +
>   proxmox-frr/debian/control       |   4 ++
>   proxmox-frr/src/de/mod.rs        |  56 +++++++++++++++++
>   proxmox-frr/src/de/openfabric.rs | 101 +++++++++++++++++++++++++++++++
>   proxmox-frr/src/de/ospf.rs       |  64 ++++++++++++++++++++
>   proxmox-frr/src/lib.rs           |   1 +
>   6 files changed, 227 insertions(+)
>   create mode 100644 proxmox-frr/src/de/mod.rs
>   create mode 100644 proxmox-frr/src/de/openfabric.rs
>   create mode 100644 proxmox-frr/src/de/ospf.rs
> 
> diff --git a/proxmox-frr/Cargo.toml b/proxmox-frr/Cargo.toml
> index 8b01fa4..8ada547 100644
> --- a/proxmox-frr/Cargo.toml
> +++ b/proxmox-frr/Cargo.toml
> @@ -13,6 +13,7 @@ rust-version.workspace = true
>   thiserror = { workspace = true }
>   anyhow = "1"
>   tracing = "0.1"
> +serde = { workspace = true, features = [ "derive" ] }
>   
>   proxmox-network-types = { workspace = true }
>   proxmox-sdn-types = { workspace = true }
> diff --git a/proxmox-frr/debian/control b/proxmox-frr/debian/control
> index 9996619..bce0c70 100644
> --- a/proxmox-frr/debian/control
> +++ b/proxmox-frr/debian/control
> @@ -9,6 +9,8 @@ Build-Depends-Arch: cargo:native <!nocheck>,
>    librust-anyhow-1+default-dev <!nocheck>,
>    librust-proxmox-network-types-0.1+default-dev (>= 0.1.1-~~) <!nocheck>,
>    librust-proxmox-sdn-types-0.1+default-dev <!nocheck>,
> + librust-serde-1+default-dev <!nocheck>,
> + librust-serde-1+derive-dev <!nocheck>,
>    librust-thiserror-2+default-dev <!nocheck>,
>    librust-tracing-0.1+default-dev <!nocheck>
>   Maintainer: Proxmox Support Team <support@proxmox.com>
> @@ -27,6 +29,8 @@ Depends:
>    librust-anyhow-1+default-dev,
>    librust-proxmox-network-types-0.1+default-dev (>= 0.1.1-~~),
>    librust-proxmox-sdn-types-0.1+default-dev,
> + librust-serde-1+default-dev,
> + librust-serde-1+derive-dev,
>    librust-thiserror-2+default-dev,
>    librust-tracing-0.1+default-dev
>   Provides:
> diff --git a/proxmox-frr/src/de/mod.rs b/proxmox-frr/src/de/mod.rs
> new file mode 100644
> index 0000000..a6674d9
> --- /dev/null
> +++ b/proxmox-frr/src/de/mod.rs
> @@ -0,0 +1,56 @@
> +use std::{collections::HashMap, net::IpAddr};
> +
> +use proxmox_network_types::ip_address::Cidr;
> +use serde::{Deserialize, Serialize};
> +
> +pub mod openfabric;
> +pub mod ospf;
> +
> +/// A nexthop of a route
> +#[derive(Debug, Serialize, Deserialize, Clone)]
> +pub struct NextHop {
> +    #[serde(rename = "interfaceIndex")]
> +    pub interface_index: i32,
> +    #[serde(rename = "interfaceName")]
> +    /// Name of the outgoing interface
> +    pub interface_name: String,
> +    /// If the nexthop is active
> +    pub active: bool,
> +    /// If the route has the onlink flag. Onlink means that we pretend that the nexthop is
> +    /// directly attached to this link, even if it does not match any interface prefix.
> +    #[serde(rename = "onLink")]
> +    pub on_link: bool,
> +    /// Remap-Source, this rewrites the source address to the following address, if this
> +    /// nexthop is used.
> +    #[serde(rename = "rmapSource")]
> +    pub remap_source: Option<IpAddr>,
> +    /// Weight of the nexthop
> +    pub weight: i32,
> +    /// If this nexthop entry is a duplicate of another (the first one has this unset)
> +    pub duplicate: Option<bool>,
> +}
> +
> +/// route
> +#[derive(Debug, Serialize, Deserialize, Clone)]
> +pub struct Route {
> +    /// Array of all the nexthops associated with this route. When you have e.g. two
> +    /// connections between two nodes, there is going to be one route, but two nexthops.
> +    pub nexthops: Vec<NextHop>,
> +    /// Metric of the route
> +    pub metric: i32,
> +    /// Protocol from which the route originates
> +    pub protocol: String,
> +    #[serde(rename = "vrfName")]
> +    pub vrf_name: String,
> +    /// If the route is installed in the kernel routing table
> +    pub installed: Option<bool>,
> +}
> +
> +/// Struct to parse zebra routes by FRR.
> +///
> +/// To get the routes from FRR, instead of asking the daemon of every protocol for their
> +/// routes we simply ask zebra which routes have been inserted and filter them by protocol.
> +/// The following command is used to accomplish this: `show ip route <protocol> json`.
> +/// This struct can be used the deserialize the output of that command.
> +#[derive(Debug, Serialize, Deserialize, Default)]
> +pub struct Routes(pub HashMap<Cidr, Vec<Route>>);
> diff --git a/proxmox-frr/src/de/openfabric.rs b/proxmox-frr/src/de/openfabric.rs
> new file mode 100644
> index 0000000..837159b
> --- /dev/null
> +++ b/proxmox-frr/src/de/openfabric.rs
> @@ -0,0 +1,101 @@
> +use serde::{Deserialize, Serialize};
> +
> +/// State of the adjacency of a OpenFabric neighbor
> +#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
> +pub enum AdjacencyState {
> +    Initializing,
> +    Up,
> +    Down,
> +    Unknown,
> +}
> +
> +/// Neighbor Interface
> +///
> +/// Interface used to communicate with a specific neighbor
> +#[derive(Debug, Serialize, Deserialize)]
> +pub struct NeighborInterface {
> +    /// The name of the interface
> +    pub name: String,
> +    /// The state of the adjacency, this is "Up" when everything is well
> +    pub state: Option<AdjacencyState>,
> +    /// Time since the last adj-flap (basically the uptime)
> +    #[serde(rename = "last-ago")]
> +    pub last_ago: String,
> +}
> +
> +/// Adjacency information
> +///
> +/// Circuits are Layer-2 Broadcast domains (Either point-to-point or LAN).
> +#[derive(Debug, Serialize, Deserialize)]
> +pub struct Circuit {
> +    /// The hostname of the adjacency peer
> +    pub adj: Option<String>,
> +    /// The interface of the neighbor
> +    pub interface: Option<NeighborInterface>,
> +}
> +
> +/// An openfabric area the same as SDN fabric.
> +#[derive(Debug, Serialize, Deserialize)]
> +pub struct Area {
> +    /// The are name, this is the same as the fabric_id, so the name of the fabric.
> +    pub area: String,
> +    /// Circuits are Layer-2 Broadcast domains (Either point-to-point or LAN).
> +    pub circuits: Vec<Circuit>,
> +}
> +
> +/// The parsed neighbors.
> +///
> +/// This models the output of:
> +/// `vtysh -c 'show openfabric neighbor json'`.
> +#[derive(Debug, Serialize, Deserialize, Default)]
> +pub struct Neighbors {
> +    /// Every sdn fabric is also an openfabric 'area'
> +    pub areas: Vec<Area>,
> +}
> +
> +/// The NetworkType of a OpenFabric interface
> +#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
> +pub enum NetworkType {
> +    #[serde(rename(deserialize = "p2p", serialize = "Point-To-Point"))]
> +    PointToPoint,
> +    #[serde(rename(deserialize = "lan", serialize = "Broadcast"))]
> +    Lan,
> +    #[serde(rename(deserialize = "loopback", serialize = "Loopback"))]
> +    Loopback,
> +    #[serde(rename = "Unknown")]
> +    Unknown,
> +}
> +
> +/// The State of a OpenFabric interface
> +#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
> +pub enum CircuitState {
> +    Init,
> +    Config,
> +    Up,
> +    Unknown,
> +}
> +
> +#[derive(Debug, Serialize, Deserialize)]
> +#[serde(rename_all = "kebab-case")]
> +pub struct Interface {
> +    pub name: String,
> +    pub state: CircuitState,
> +    #[serde(rename = "type")]
> +    pub ty: NetworkType,
> +}
> +
> +#[derive(Debug, Serialize, Deserialize)]
> +pub struct InterfaceCircuits {
> +    pub interface: Interface,
> +}
> +
> +#[derive(Debug, Serialize, Deserialize, Default)]
> +pub struct InterfaceArea {
> +    pub area: String,
> +    pub circuits: Vec<InterfaceCircuits>,
> +}
> +
> +#[derive(Debug, Serialize, Deserialize, Default)]
> +pub struct Interfaces {
> +    pub areas: Vec<InterfaceArea>,
> +}
> diff --git a/proxmox-frr/src/de/ospf.rs b/proxmox-frr/src/de/ospf.rs
> new file mode 100644
> index 0000000..a75609d
> --- /dev/null
> +++ b/proxmox-frr/src/de/ospf.rs
> @@ -0,0 +1,64 @@
> +use std::collections::HashMap;
> +
> +use serde::{Deserialize, Serialize};
> +
> +/// Information about the Neighbor (Peer) of the Adjacency.
> +#[derive(Debug, Serialize, Deserialize)]
> +#[serde(rename_all = "camelCase")]
> +pub struct Neighbor {
> +    /// The full state of the neighbor. This is "{converged}/{role}".
> +    #[serde(rename = "nbrState")]
> +    pub neighbor_state: String,
> +    /// The uptime of the interface
> +    #[serde(rename = "upTime")]
> +    pub up_time: String,
> +    /// The interface name of this adjacency. This is always a combination of interface
> +    /// name and address. e.g. "ens21:5.5.5.3".
> +    #[serde(rename = "ifaceName")]
> +    pub interface_name: String,
> +}
> +
> +/// The parsed OSPF neighbors
> +#[derive(Debug, Deserialize, Default)]
> +pub struct Neighbors {
> +    /// The OSPF neighbors. This is nearly always a ip-address - neighbor mapping.
> +    pub neighbors: HashMap<String, Vec<Neighbor>>,
> +}
> +
> +/// All possible OSPF network-types that can be returned from frr
> +#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
> +pub enum NetworkType {
> +    #[serde(rename = "Null")]
> +    Null,
> +    #[serde(rename(deserialize = "POINTOPOINT", serialize = "Point-To-Point"))]
> +    PointToPoint,
> +    #[serde(rename(deserialize = "BROADCAST", serialize = "Broadcast"))]
> +    Broadcast,
> +    #[serde(rename = "NBMA")]
> +    Nbma,
> +    #[serde(rename(deserialize = "POINTOMULTIPOINT", serialize = "Point-To-Multipoint"))]
> +    PointToMultipoint,
> +    #[serde(rename(deserialize = "VIRTUALLINK", serialize = "Virtual Link"))]
> +    VirtualLink,
> +    #[serde(rename(deserialize = "LOOPBACK", serialize = "Loopback"))]
> +    Loopback,
> +}
> +
> +#[derive(Debug, Deserialize)]
> +#[serde(rename_all = "camelCase")]
> +pub struct Interface {
> +    /// The interface state
> +    pub if_up: bool,
> +    /// The network type (e.g. point-to-point, broadcast, etc.)
> +    ///
> +    /// Note there is also a "state" property, but that models the state of the interface (ism),
> +    /// which can also be "point-to-point", but it can also be e.g. "Down" or e.g. "DROther"!
> +    /// So networkType is the configured network type and state is the state of interface, which
> +    /// sometimes is the same as the networkType.
> +    pub network_type: NetworkType,
> +}
> +
> +#[derive(Debug, Deserialize, Default)]
> +pub struct Interfaces {
> +    pub interfaces: HashMap<String, Interface>,
> +}
> diff --git a/proxmox-frr/src/lib.rs b/proxmox-frr/src/lib.rs
> index 35b62cb..2e6ab62 100644
> --- a/proxmox-frr/src/lib.rs
> +++ b/proxmox-frr/src/lib.rs
> @@ -1 +1,2 @@
> +pub mod de;
>   pub mod ser;
_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
next prev parent reply	other threads:[~2025-11-03 14:50 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-30 15:48 [pve-devel] [PATCH common/manager/network/proxmox{-ve-rs, -perl-rs} 00/35] Improve status reporting for SDN / networking Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-common 1/2] iproute2: add helper for detecting bridge members Stefan Hanreich
2025-11-03 14:14   ` Hannes Laimer
2025-11-03 16:31     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-common 2/2] iproute2: add helper for querying vlan information Stefan Hanreich
2025-11-03 14:16   ` Hannes Laimer
2025-11-03 16:31     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 1/6] frr: make room for deserialization structs Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 2/6] frr: add deserialization types for openfabric and ospf Stefan Hanreich
2025-11-03 14:50   ` Hannes Laimer [this message]
2025-11-03 16:32     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 3/6] ve-config: add helper function to iterate over all nodes in all fabrics Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 4/6] ve-config: add optional tag property to vnet Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 5/6] frr: fix some route deserialization types Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-ve-rs 6/6] frr: add deserialization types for EVPN Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 01/10] pve-rs: firewall: cargo: fmt Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 02/10] pve-rs: firewall: add missing documentation comments Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 03/10] pve-rs: cargo: bump proxmox-apt and proxmox-ve-config versions Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 04/10] pve-rs: fabrics: update proxmox-frr import path Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 05/10] pve-rs: fabrics: fix clippy lint warnings Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 06/10] pve-rs: fabrics: add function to get status of fabric Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 07/10] pve-rs: fabrics: add function to get l2vpn and l3vpn routes for evpn Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 08/10] pve-rs: fabrics: add function to get routes learned by a fabric Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 09/10] pve-rs: fabrics: add function to get the interfaces used for " Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH proxmox-perl-rs 10/10] pve-rs: fabrics: add function to get the neighbors " Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 1/9] refactor: rework api module structure for the /nodes/{node}/sdn subdir Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 2/9] fabrics: add fabrics status to SDN::status function Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 3/9] sdn: status: add zone type to sdn resource Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 4/9] api: nodes: fabrics: add endpoint for querying route status Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 5/9] api: nodes: fabrics: add endpoint for querying neighbor information Stefan Hanreich
2025-11-03 14:22   ` Hannes Laimer
2025-11-03 16:34     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 6/9] api: nodes: fabrics: add endpoint for querying interface status Stefan Hanreich
2025-11-03 14:24   ` Hannes Laimer
2025-11-03 16:32     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 7/9] api: nodes: zones: add bridge status Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 8/9] api: nodes: zones: add ip vrf endpoint for evpn zones Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-network 9/9] api: nodes: vnets: add mac-vrf endpoint for evpn vnets Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 1/8] api: nodes: use new status module for sdn subdirectory Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 2/8] refactor: ui: sdn browser: parametrize zone content panel Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 3/8] pvestatd: add network resource to status reporting Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 4/8] pvestatd: sdn: adapt to changes in " Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 5/8] ui: resource tree: add network resource Stefan Hanreich
2025-11-03 14:19   ` Hannes Laimer
2025-11-03 16:38     ` Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 6/8] ui: sdn browser: Add ip-vrf panel for evpn zones Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 7/8] ui: sdn browser: add mac vrf panel Stefan Hanreich
2025-10-30 15:48 ` [pve-devel] [PATCH pve-manager 8/8] ui: sdn browser: add zone bridge view Stefan Hanreich
2025-11-03 14:54 ` [pve-devel] [PATCH common/manager/network/proxmox{-ve-rs, -perl-rs} 00/35] Improve status reporting for SDN / networking Hannes Laimer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox
  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):
  git send-email \
    --in-reply-to=e3f18921-5137-4863-b4c1-ed296512d888@proxmox.com \
    --to=h.laimer@proxmox.com \
    --cc=pve-devel@lists.proxmox.com \
    --cc=s.hanreich@proxmox.com \
    /path/to/YOUR_REPLY
  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
  Be sure your reply has a Subject: header at the top and a blank line
  before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox