Matrix & Directions return 502 – "DNS lookup failure for heigitsv03.heigit.org"

openrouteservice matrix
1

Hi ORS/HeiGIT team,

Since ~July 23, 2026, both the Matrix and the Directions API fail. The error is generated by your own infrastructure (Apache at heigitlnxproxy01.central.heigit-internal.de), not by my client:

502 Proxy Error — “The proxy server received an invalid response from an upstream server.
Reason: DNS lookup failure for: heigitsv03.heigit.org”

It looks like your front proxy cannot resolve the internal backend host heigitsv03.heigit.org.

What I tested (server-side, plain cURL from my own host, valid API key, well within quota):

  1. Matrix on the new domain
    POST https://api.heigit.org/openrouteservice/v2/matrix/driving-car
    → HTTP 502, “DNS lookup failure for heigitsv03.heigit.org”

  2. Matrix on the old (deprecated) domain
    POST https://api.openrouteservice.org/v2/matrix/driving-car
    → connection times out after 25s, 0 bytes received

  3. Directions on the new domain
    POST https://api.heigit.org/openrouteservice/v2/directions/driving-car/geojson
    → HTTP 502, same “DNS lookup failure for heigitsv03.heigit.org”

So it affects Matrix AND Directions, on both the old and new domains — all pointing to the same unresolvable internal backend. Since the call is server-side with a valid key and low usage, this is not a CORS/key/quota issue on my side.

Minimal request to reproduce:
{“locations”:[[7.4474,46.9480],[7.4390,46.9510]],“metrics”:[“duration”]}

The timing suggests it started after the server maintenance on July 21, 2026. Could you please check/restore the resolution of heigitsv03.heigit.org behind your proxy?

Thanks a lot!

2

Hey,

thanks a lot for your information :+1:

We are aware of the problem and we are looking into it.
We cannot give any promises on a timeline for resolving the issue.

Best regards
the openrouteservice team

3

Hey,

this was indeed a DNS issue in our supply chain.
Since Saturday, around 15:30 CEST, the issue was resolved.

We are very sorry for any inconveniences caused.
We are working to make sure these issues do not happen again.

Best regards
the openrouteservice team

4

Hello,

I’m getting a 403 {“error”: “Quota exceeded”} response on the Directions V2 / driving-hgv endpoint, but my dashboard shows my full quota available (Directions V2: 2000/2000, all other endpoints also showing full).

Details:

  • Org ID: 5b3ce3597851110001cf6248
  • Endpoint: POST /v2/directions/driving-hgv
  • Response has no X-Ratelimit-* or quota headers at all, just Server: nginx, X-Generator: tyk.io, and the error body
  • This has been consistent across multiple tests over the past ~24 hours, including after waiting well past a minute between requests (ruling out the per-minute limit)
  • Dashboard consistently shows full quota remaining the entire time

Is there a sync delay between the dashboard and the enforcement layer, or something else going on with my account/key? Happy to provide the full key via DM if needed. Any help would be appreciated — this is blocking a project I’m actively working on.

Thanks!

5

Since this morning (2026-07-25, ~07:30 UTC) every request with my API key returns HTTP 403 {"error": "Access to this API has been disallowed"} — on both api.openrouteservice.org and api.heigit.org, for all endpoints (directions v2 and geocoding/pelias), with both api_key query param and Authorization header.

What I’ve ruled out:

  • Quota: dashboard shows full remaining quotas (Directions 2000/2000, Geocoding 3000/3000). Daily usage has never exceeded ~50 requests.
  • Stale key: I regenerated the key today; the new key gets the same 403 on both hosts, still ~45 minutes after regeneration.
  • Endpoints: tested api.heigit.org/openrouteservice/v2/directions/foot-walking and api.heigit.org/pelias/v1/search per the migration announcement.

Earlier the same morning (~05:40 UTC) the gateway returned plain nginx 502 Bad Gateway for a while, then switched to these 403s — this looks like my account’s keys lost their gateway policy binding during that episode. Could you check the key/policy sync for my account? (Account email = the one attached to this forum account; happy to DM the key prefix.)

6

Both posts are related to the same issue, so I moved them here.

7

For the folks getting 403s — did you also hit the DNS timeouts, or are those separate? Might help figure out if it’s all one issue.