Restores AI Assistant support for OpenAI-compatible endpoints on private IPs, loopback
and Docker networks (LiteLLM, LM Studio, LocalAI, vLLM, OmniRoute, self-hosted proxies)
and surfaces the server's error under the Load button (#325). Aligns the Secure Gateway
wizard's Alpine template selection and pct create with the host's real architecture on
x86_64 and arm64 (#324). Consolidates changes landing on develop: atomic notification
delivery, custom SSH port for Borg remote targets (#236), optional GitHub API token for
app version tracking (#306), and richer replication failure notifications.
Restores AI Assistant support for OpenAI-compatible endpoints on private IPs, loopback
and Docker networks (LiteLLM, LM Studio, LocalAI, vLLM, OmniRoute, self-hosted proxies)
and surfaces the server's error under the Load button (#325). Aligns the Secure Gateway
wizard's Alpine template selection and pct create with the host's real architecture on
x86_64 and arm64 (#324). Consolidates changes landing on develop: atomic notification
delivery, custom SSH port for Borg remote targets (#236), optional GitHub API token for
app version tracking (#306), and richer replication failure notifications.
feat(host-backup): custom SSH port for Borg remote targets
Add Borg destinations no longer assume port 22. The Monitor form and
the shell TUI both take an optional port (default 22, range 1-65535)
and embed it in the persisted ssh://user@host:port/path URL. BORG_RSH,
the sshpass key installer and the capacity probe all honour it.
Reported by @songochain in #236 — NAS-style Borg hosts on non-standard
SSH ports are now first-class targets.
Existing borg-targets.txt entries without a port keep working; the
port is only serialised into the URL when it differs from 22.
Resolves the conflict on AppImage/config/verified_ai_models.json in favour
of develop's curated version with _exclude — main had the bot's first
naive refresh which surfaced allam-2-7b as the Groq recommended model.
develop also absorbs the bot-maintained files main had accumulated since
1.2.5 (helpers_cache, app_tracking_hints, project-growth, FUNDING).
Stable release consolidating the v1.2.4 beta cycle (1.2.4.1-beta and 1.2.4.2-beta) into 1.2.5.
Highlights:
- Apps dashboard: single launcher for every LXC-registered app and user-defined Custom Web Link, with category badges, search, sort and one-click deep-links back to the guest modal.
- LXC Apps & Updates end-to-end: App tab inside every guest modal, upstream version tracking, and Easy Updates that cover OS packages, registered apps, Docker Engine and per-image updates on the same 24-hour cycle.
- Application detection catalog with 380+ tracked workloads generated live from community-scripts across seven detector methods.
- Monitor now speaks 8 languages: English, Spanish, German, French, Italian, Portuguese, Slovak and Swedish (i18n scaffolding by @vaso73).
- NVIDIA multi-GPU passthrough by exact BDF so one card can be assigned to a VM while another stays operational on the host or LXC.
- Navigation reorder, Memory & Swap real memory-pressure signal, native Pushover channel, Actions API, plus wide-reaching improvements across health, hardware, network, backup and post-install.
Full release notes: see CHANGELOG.md and https://github.com/MacRimi/ProxMenux/releases
Stable release consolidating the v1.2.4 beta cycle (1.2.4.1-beta and 1.2.4.2-beta) into 1.2.5.
Highlights:
- Apps dashboard: single launcher for every LXC-registered app and user-defined Custom Web Link, with category badges, search, sort and one-click deep-links back to the guest modal.
- LXC Apps & Updates end-to-end: App tab inside every guest modal, upstream version tracking, and Easy Updates that cover OS packages, registered apps, Docker Engine and per-image updates on the same 24-hour cycle.
- Application detection catalog with 380+ tracked workloads generated live from community-scripts across seven detector methods.
- Monitor now speaks 8 languages: English, Spanish, German, French, Italian, Portuguese, Slovak and Swedish (i18n scaffolding by @vaso73).
- NVIDIA multi-GPU passthrough by exact BDF so one card can be assigned to a VM while another stays operational on the host or LXC.
- Navigation reorder, Memory & Swap real memory-pressure signal, native Pushover channel, Actions API, plus wide-reaching improvements across health, hardware, network, backup and post-install.
Full release notes: see CHANGELOG.md and https://github.com/MacRimi/ProxMenux/releases
ProxMenuxBot
github-actions[bot]
github-actions[bot]
MacRimiandGitHub
MacRimi
proxmenux-bot