From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate001.proxmox.com (gate001.proxmox.com [IPv6:2a0f:8001:1:32::40]) by lore.proxmox.com (Postfix) with ESMTPS id 28C8A1FF0C0 for ; Wed, 18 Nov 2026 16:20:37 +0100 (CET) Received: from gate001.proxmox.com (localhost.localdomain [127.0.0.1]) by gate001.proxmox.com (Proxmox) with ESMTP id 705CB21710; Wed, 18 Nov 2026 16:20:35 +0100 (CET) Date: Wed, 18 Nov 2026 16:20:29 +0100 From: Arthur Bied-Charreton To: Stefan Hanreich Subject: Re: [PATCH docs/manager/network/proxmox{,-ve-rs,-perl-rs} v2 00/11] Status reporting for WireGuard fabrics Message-ID: References: <20261117183500.750636-1-s.hanreich@proxmox.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20261117183500.750636-1-s.hanreich@proxmox.com> X-Bm-Milter-Handled: 55990f41-d878-4baa-be0a-ee34c49e34d2 X-Bm-Transport-Timestamp: 1795015231016 X-SPAM-LEVEL: Spam detection results: 0 AWL 0.545 Adjusted score from AWL reputation of From: address DMARC_MISSING 0.1 Missing DMARC policy KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment (newer systems) RCVD_IN_DNSWL_MED -2.3 Sender listed at https://www.dnswl.org/, medium trust SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record Message-ID-Hash: ZUOK5MTFWPQ4GRQGNHUSU5UK5TS3HW2X X-Message-ID-Hash: ZUOK5MTFWPQ4GRQGNHUSU5UK5TS3HW2X X-MailFrom: a.bied-charreton@proxmox.com X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header CC: pve-devel@lists.proxmox.com X-Mailman-Version: 3.3.10 Precedence: list List-Id: Proxmox VE development discussion List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: On Tue, Nov 17, 2026 at 07:34:47PM +0100, Stefan Hanreich wrote: > ## Introduction > > This patch series adds status reporting for WireGuard fabrics to pvestatd and > exposes additional information about WireGuard interfaces and neighbors (= > peers) via the pre-existing interface/neighbors endpoints for fabrics. > > [snip] > ## Open questions > > The return schema of the neighbor status endpoint had some non-optional strings. > This series makes 'uptime' optional, since it is not returned for the WireGuard > fabric, as well as 'neighbor', since the endpoint of a WireGuard peer is not > known before the first handshake if it has none configured. Initially, I > returned hard-coded strings for those fields for the WireGuard fabric, but > decided to omit them instead, since they could be potentially misleading and > don't really add anything. Still a bit unsure about this, since it is > technically a breaking API change. It should be easy to just add this in a new > iteration again - even if only 'n/a' or something similar is returned for the > non-optional fields. fwiw i think omitting them is fine, returning a placeholder would be more confusing IMO. [snip] setup: 3-node mesh WG fabric tested the following cases: - neighbors are 'idle' without traffic with persistent-keepalive disabled - traffic between node & neighbor changes neighbor status to 'active' - neighbor drops back to 'idle' after some inactivity - all neighbors become 'active' after setting persistent keepalive (25s) - cutting a neighbor off drops it to 'down', re-allowing traffic sets it back to 'active' - setting the wg0 link to down on a node reports it as 'down' and the whole fabric as 'not ok' on that node - deleting wg0 altogether shows it as 'missing', the whole fabric as 'not ok' on that node - adding another dummy link called wg0 shows it as 'error', the whole fabric as 'not ok' on that node - removing a neighbor (wg set wg0 peer ... remove) reports the fabric as 'not ok' on that node - an extra peer that is not in the config appears in the neighbors list, fabric is shown as 'not ok' on that node - removing the wg executable shows the fabric as 'not ok' on that node just one question: would it make sense to log `wg show` failures somehow? currently the fabric is just shown as 'not ok' but one does not see why anywhere. might be a bit spammy though since this is called regularly by pvestatd, but maybe it would be possible to only log on state changes? besides that i haven't run into any unexpected behavior while testing, therefore consider this series: Tested-by: Arthur Bied-Charreton