VRM API stuck on ‘Connecting to VRM API’

Been wrestling with this one myself over the past few weeks, so thought I'd start a proper thread rather than necropost on older discussions.

My setup is a Victron Multiplus-II with a Cerbo GX, and I've been trying to pull solar forecast data through the VRM API into Node-RED to do some smarter battery management decisions — particularly around pre-conditioning my Fogstar Drift cells before a high-demand morning.

The issue I keep hitting is the API connection just hanging indefinitely at the handshake stage. It's not a credentials problem — the token is valid and I can hit the endpoint fine via Postman. Something in the Node-RED HTTP request node seems to be fumbling the auth header formatting when polling VRM specifically.

A few things worth checking if you're seeing the same:

• Token format — VRM wants Bearer <token>, and some node configs silently strip the prefix
• SSL verification — disabling it temporarily can isolate whether it's a certificate chain issue
• Rate limiting — VRM API will quietly drop connections if you're hammering it; add a delay node between polls
• Portal ID — double-check it's pulling the correct installation ID, not defaulting to a cached value

I'm on Node-RED 3.1.x running on a local Pi 4, not the Cerbo's built-in instance, which adds another network hop that might be relevant.

Has anyone in a UK setup managed to get reliable solar forecast polling working through VRM? Particularly interested whether anyone's tied it into Octopus Agile tariff logic for smarter charge scheduling — that's the end goal here.

What are your experiences?

@RetiredSquaddie — classic one this. Nine times out of ten when I've seen "Connecting to VRM API" stuck like that on the Cerbo, it's one of three things:

1. MQTT broker conflict — if you're running anything else on the same network polling the Cerbo (Home Assistant, Node-RED, etc.), it can starve the VRM connection
2. Stale VRM token — remove the installation from VRM portal entirely, factory reset the Cerbo's VRM registration (not a full reset), re-pair
3. DNS resolution failing silently — the Cerbo can have a valid IP and still not resolve ccgxlogging.victronenergy.com — worth checking your router's DNS settings

On my shepherd's hut setup with a Cerbo GX and MPPT, I had exactly this after adding an EV charger on the same VLAN. Separating the MQTT traffic sorted it immediately.

What firmware version is your Cerbo running?

Nine times out of ten it's either a dodgy DNS config on the Cerbo or someone's router blocking outbound on port 443 — check your firewall logs before you go blaming Victron's servers like everyone always does.

Has anyone checked whether the VRM portal itself is having issues at that moment? I got caught out once on the narrowboat assuming it was my setup, spent ages faffing with the Cerbo GX settings, only to find Victron's own status page was showing an incident 😅

Also worth asking — @RetiredSquaddie, is your Cerbo on the latest firmware? I had a really stubborn "stuck connecting" issue on mine that turned out to be a known bug in an older firmware version. Updating via the remote console sorted it immediately.

And just to rule out the obvious — can the Cerbo actually reach the internet at all? I check mine by pinging something from the diagnostics menu. Sometimes on the boat the 4G router drops out quietly without any obvious indication.

Good shout from @LindaClark90 — always worth ruling out a VRM-side outage first before you go poking around your own kit.

What actually sorted mine on the boat was checking the system date/time on the Cerbo GX. Had a battery go flat over winter, Cerbo lost its time sync, and SSL handshakes were silently failing because the certificate timestamps didn't make sense anymore. Looked exactly like a stuck "Connecting" loop.

Go to Settings → General → Date & Time and make sure it's set to sync via the internet rather than manually. Once GPS or NTP sorts it out, VRM tends to reconnect within a minute or two.

Worth a look before you start questioning your router or DNS — took me an embarrassingly long afternoon to find that one!

@RetiredSquaddie had this exact grief last winter with my Cerbo. After ruling out DNS and port 443 (cheers @48VQueen, solid first checks), and confirming VRM wasn't having a moment (@LindaClark90 is right, check vrm.victronenergy.com/status first), my culprit was the system date/time being wrong on the Cerbo.

TLS handshakes fail silently if your timestamp is off — the certificate validation chucks it. Go to Settings → General → Date/Time and make sure NTP sync is actually working, not just enabled. Mine showed "NTP enabled" but was querying a blocked NTP server.

Also worth checking: Settings → VRM online portal → Connection error — it'll often give you a more specific error code than the generic connecting spinner suggests.