[pve-devel] [PATCH pve-docs 1/1] sdn: add section about IS-IS fabric

[Gabriel Goller <g.goller@proxmox.com>]

Add short section about IS-IS fabrics, nearly the same as the OpenFabric
section. Also add a warning on the IS-IS Controller about the
deprecation and a link to the IS-IS Fabrics. Also mention the
differences between them, which is that the fabrics don't redistribute
connected routes.

Signed-off-by: Gabriel Goller <g.goller@proxmox.com>
---
 pvesdn.adoc | 111 +++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 105 insertions(+), 6 deletions(-)

diff --git a/pvesdn.adoc b/pvesdn.adoc
index 87596916d4cb..4cf0355ef9da 100644
--- a/pvesdn.adoc
+++ b/pvesdn.adoc
@@ -509,6 +509,10 @@ bgp-multipath-as-path-relax:: Allow ECMP if your peers have different ASN.
 ISIS Controller
 ~~~~~~~~~~~~~~~
 
+WARNING: Consider using the xref:pvesdn_isis[IS-IS] Fabric, the ISIS Controller
+will be deprecated soon. Note that the IS-IS Fabric does not redistribute
+connected routes into the fabric.
+
 The ISIS controller is not used directly by a zone.
 You can use it to configure FRR to export EVPN routes to an ISIS domain.
 
@@ -603,12 +607,12 @@ behavior and proper source address selection throughout the fabric.
 Notes on IPv6
 ^^^^^^^^^^^^^
 
-IPv6 is currently only usable on OpenFabric fabrics. These IPv6 Fabrics need
-global IPv6 forwarding enabled on all nodes contained in the fabric. Without
-IPv6 forwarding, non-full-mesh fabrics won't work because the transit nodes
-don't forward packets to the outer nodes. Currently there isn't an easy way to
-enable IPv6 forwarding per-interface like with IPv4, so it has to be enabled
-globally. This can be accomplished by appending this line:
+IPv6 is currently only usable on OpenFabric and IS-IS fabrics. These IPv6
+Fabrics need global IPv6 forwarding enabled on all nodes contained in the
+fabric. Without IPv6 forwarding, non-full-mesh fabrics won't work because the
+transit nodes don't forward packets to the outer nodes. Currently there isn't
+an easy way to enable IPv6 forwarding per-interface like with IPv4, so it has
+to be enabled globally. This can be accomplished by appending this line:
 
 ----
 post-up sysctl -w net.ipv6.conf.all.forwarding=1
@@ -711,6 +715,101 @@ WARNING: When you remove an interface with an entry in `/etc/network/interfaces`
 that has `manual` set, then the IP will not get removed on applying the SDN
 configuration.
 
+
+[[pvesdn_isis]]
+IS-IS
+~~~~~
+
+NOTE: Unlike the IS-IS Controller, the IS-IS Fabric does not distribute all the
+connected routes into the fabric. So it does not set
+`redistribute ipv4/6 connected level-1`.
+
+IS-IS is a routing protocol specifically designed for data center fabrics.
+
+[thumbnail="screenshot/gui-datacenter-create-fabric-isis.png"]
+
+Configuration options:
+
+[[pvesdn_isis_fabric]]
+On the Fabric
+^^^^^^^^^^^^^
+
+Name:: This is the name (domain) of the IS-IS fabric and can be at most 8 characters long.
+
+IPv4 Prefix:: IPv4 CIDR network range (e.g., 192.0.2.0/24) used to verify that
+all router-IDs in the fabric are contained within this prefix.
+
+IPv6 Prefix:: IPv6 CIDR network range (e.g., 2001:db8::/64) used to verify that
+all router-IDs in the fabric are contained within this prefix.
+
+WARNING: For IPv6 fabrics to work, global forwarding needs to be enabled on all
+nodes. Check xref:pvesdn_config_fabrics_ipv6[Notes on IPv6] for how to do it and additional info.
+
+Hello Interval:: Controls how frequently (in seconds) hello packets are sent to
+discover and maintain connections with neighboring nodes. Lower values detect
+failures faster but increase network traffic. This option is global on the
+fabric, meaning every interface on every node in this fabric will inherit this
+hello-interval property. The default value is 3 seconds.
+
+CSNP Interval:: Sets how frequently (in seconds) the node synchronizes its
+routing database with neighbors. Lower values keep the network topology information
+more quickly in sync but increase network traffic. This option is global on the
+fabric, meaning every interface on every node in this fabric will inherit this
+property. The default value is 10 seconds.
+
+[[pvesdn_isis_node]]
+On the Node
+^^^^^^^^^^^
+
+[thumbnail="screenshot/gui-datacenter-create-node-isis.png"]
+
+Options that are available on every node that is part of a fabric:
+
+Node:: Select the node which will be added to the fabric. Only nodes that
+currently are in the cluster will be shown.
+
+IPv4:: A unique IPv4 address used to generate the IS-IS
+Network Entity Title (NET). Each node in the same fabric must have a different
+Router-ID, while a single node must use the same NET address across all fabrics
+(If this is not given {pve} will automatically choose one and ensure that the
+configuration is valid).
+
+IPv6:: A unique IPv6 address used to generate the IS-IS
+Network Entity Title (NET). Each node in the same fabric must have a different
+Router-ID, while a single node must use the same NET address across all fabrics.
+If a IPv4 and IPv6 address is configured, the IPv4 one will be used to derive
+the NET.
+
+WARNING: When using IPv6 addresses, the last 3 segments are used to generate
+the NET. Ensure these segments differ between nodes.
+
+Interfaces:: Specify the interfaces used to establish peering connections with
+other IS-IS nodes. Preferably select interfaces without pre-assigned IP
+addresses, then configure addresses in the IPv4/IPv6 column if needed. A dummy
+"loopback" interface with the router-id is automatically created.
+
+On The Interface
+^^^^^^^^^^^^^^^^
+
+The following optional parameters can be configured per interface when enabling
+the additional columns:
+
+IP::: A IPv4 that should get automatically configured on this interface. Must
+include the netmask (e.g. /31). If no address is set, the interface is
+configured as a `point-to-point` interface.
+
+IPv6::: A IPv6 that should get automatically configured on this interface. Must
+include the netmask (e.g. /127). If no address is set, the interface is
+configured as a `point-to-point` interface.
+
+Hello Multiplier::: Defines how many missed hello packets constitute a failed
+connection. Higher values make the connection more resilient to packet loss but
+slow down failure detection. The default value is 10.
+
+WARNING: When you remove an interface with an entry in `/etc/network/interfaces`
+that has `manual` set, then the IP will not get removed on applying the SDN
+configuration.
+
 [[pvesdn_ospf]]
 OSPF
 ~~~~
-- 
2.47.2



_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel