direct.ett.healthit.gov mail server not accepting connections on port 25 (since Sept 26)

[REDA ELAMRI]

Hi ETT team,

ETT's Direct mail server for direct.ett.healthit.gov (MX ec2-35-169-0-104.compute-1.amazonaws.com, 35.169.0.104) is not accepting SMTP connections on port 25. We have seen it on three separate runs:

- Sept 26, ~8:40 PM and ~9:10 PM CDT: Receive SMTP Disposition-Notification (MT Test 39), card log shows "Couldn't connect to host, port: direct.ett.healthit.gov, 25".
- Sept 28, 11:42 AM CDT: MT Test 39 again, same error.
- Sept 28, 2:02 PM CDT: SMTP MT Test 23(a) (Message to Bad Address). ETT handed the message to our HISP fine, but our server could not deliver it to badad...@direct.ett.healthit.gov: "Couldn't connect to host, port: 35.169.0.104, 25; timeout 60000". The card now waits for an Accept/Reject we can't give, since the bad-address rejection never happened.

The ett.healthit.gov side works: today our MT 21(a) and 21(b) tests with processedd...@ett.healthit.gov (34.235.25.130) passed. So it looks limited to the direct.ett.healthit.gov mail service.

Could you check or restart it? We are running our accreditation tests against production this month and these are the tests we can't complete.

Thanks,
Reda Elamri
iShare Medical

[Sai Valluripalli]

Can you check now and confirm.

[Reda]

Hi Sai,

Confirmed, thank you: https://direct.ett.healthit.gov:25/ has been accepting connections again since the morning of Sept 29, and we completed our Delivery Notification tests (MT 39/40/41 and the sender set) against it that day.

One remaining item on the XDR side, which I posted separately on Sept 26 (thread "XDR 'Send using Edge Protocol - XDR' (h)(2) Test 10 - SUT's XDR reaches ETT but card stays 'Pending'", https://www.google.com/url?q=https://groups.google.com/g/edge-test-tool/c/gimbfL8N0fk&source=gmail&ust=1790812251950000&sa=E) and which has not had a reply yet:

The edge-ttp XDR simulator (Test By Criteria > H2 > (i)(C) Send using Edge Protocol - XDR, Tests 10/11/12/16) accepts our ITI-41 ProvideAndRegisterDocumentSet-b with HTTP 200 but never captures it or advances the card. The card stays Pending, the LOGS show only "ran tc 10" with no request, and no validation report is emailed. Latest attempts, all with this result:

- Sept 29, 21:42Z and 21:51Z (UTC), from our production HISP, source IP 66.6.125.16, Direct To Address xdr-s...@direct.isharemedicalrecords.com, endpoint edge-ttp__10
- Sept 26, 03:48Z / 04:07Z / 05:27Z, from our test HISP, source IP 136.35.205.142, Direct To Address ett-...@isharedev.xyz

The identical message sent to a local NIST XDS Toolkit instance (xdstools 7.10.0, edge-ttp__10 Document Recipient sim) is recorded, returns RegistryResponse Success, and fires the transaction notification callback, so it looks like the same hosted-simulator condition as the Jul 2026 "b1 XDR Sending Cases - Stuck at Pending" thread.

Could you please reset/restart the edge-ttp XDR simulator (or re-prime the notification wiring for Tests 10/11/12/16) and let us know when it is back? We will re-run immediately and confirm. We are working against an Oct 21 package freeze, so this is our last open ETT item. Happy to send the full outbound SOAP envelope or any logs.

Thank you,
Reda Elamri
iShare Medical

--
You received this message because you are subscribed to the Google Groups "Edge Test Tool (ETT)" group.
To unsubscribe from this group and stop receiving emails from it, send an email to edge-test-too...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/edge-test-tool/bdeb41d6-1d12-4e04-95e6-023c6dc2f3a2n%40googlegroups.com.

[Prince Boothe]

Hi ETT team,

The ETT SMTP submission server james.healthit.gov (35.169.165.4) is not accepting SMTP connections on port 25.

We've hit this consistently while running our §170.315(b)(1) Direct send tests (SMTP edge — Tests 8, 14, 18, submitting as vendor...@james.healthit.gov):

• Sept 29 and Sept 30, 2026 (repeated attempts, UTC): a plain TCP connect to james.healthit.gov:25 returns "Connection refused" immediately. The host itself is reachable — it responds with a TCP RST — but nothing is listening on port 25, so we never reach the SMTP greeting to run HELO / STARTTLS / AUTH or submit a message.
• Measured from an external host with confirmed outbound port 25 (verified by successfully connecting to other public MX on port 25 from the same host), so this is not an egress block on our side.

This looks isolated to james. The Direct mail server direct.ett.healthit.gov:25 (35.169.0.104) — which had the same problem last week — has been accepting connections again since the morning of Sept 29 (confirmed on this list). james.healthit.gov:25 appears to need the same restart.

Could you please check or restart the SMTP service on james.healthit.gov:25 and let us know when it's back? We'll re‑run immediately and confirm. We're running our (b)(1) accreditation send tests this month, and this is the endpoint we submit to.

Happy to provide connection logs or exact timestamps if helpful.

Thank you,
Prince Boothe
MemryHealth LLC

[Sai Valluripalli]

Prince,

Apache James is now available. Can you try now?

Thank you,

Sai

[Prince Boothe]

Thanks Sai — confirmed, james.healthit.gov:25 is accepting connections again and presenting the expected certificate (fingerprint matches on our end). Appreciate the quick restart. We'll proceed with our b.1 send tests.