Question about oai-outdoor-ota + mobile-endpoints: mobile endpoints remain pending

[Ghinwa]

Hello everyone,

I am currently testing the oai-outdoor-ota profile on POWDER and I am seeing the same issue again with the mobile endpoints side.

My setup is:

• Project: wireless-dt

• Outdoor profile: oai-outdoor-ota

• Reserved resources: cnode-mario, d430 x2

• Spectrum: 3410–3450 MHz

What I observe:

• The main outdoor experiment appears to be running.

• Grafana is accessible.

• The gNB panel shows activity/logs, so oai-gnb.service seems to be running.

• The core network panel also shows logs.

• However, the mobile-endpoints experiment stays in pending with the message:
  “Your experiment cannot be instantiated yet, trying again in a few minutes.”

• As a result, the UE / UE metrics panels remain empty.

I have attached:

• a screenshot of the mobile-endpoints experiment page showing the pending state,

• and a screenshot of the Grafana dashboard showing gNB/core activity while the UE panels remain empty.

•  the experiment status link if that is useful for debugging and it still running:

• https://www.powderwireless.net/instantiate.php?profile=49132079-309d-11ed-b318-e4434b2381fc&rerun_instance=91e005c9-e42b-4c8f-8c4f-f6e6ee9aa356

• URL

From my side, this makes it unclear whether the issue is:

1. a problem with mobile-endpoints instantiation / resource availability,

2. expected behavior due to current shuttle endpoint availability,

3. or something incorrect in how I launched the experiments.

Could you please clarify:

• Is a long pending state for the mobile-endpoints profile expected in this situation, or does it usually indicate a problem? since it's been few hours now waiting for it.

• Is there anything specific I should check regarding reservation matching or profile parameters?

Thank you very much.

Best regards,
Ghinwa

[David M Johnson]

On 3/28/26 14:36, Ghinwa wrote:
> Hello everyone,
>
> I am currently testing the oai-outdoor-ota profile on POWDER and I am
> seeing the same issue again with the mobile endpoints side.

> However, the mobile-endpoints experiment stays in *pending* with the

> message:
> “Your experiment cannot be instantiated yet, trying again in a few
> minutes.”

> 2. expected behavior due to current shuttle endpoint availability,

> Is a long pending state for the mobile-endpoints profile expected in
> this situation, or does it usually indicate a problem? since it's
> been few hours now waiting for it.

Thanks for the detailed report! In this case, most likely there were no
available shuttles, which is possible since they operate on a reduced
schedule on the weekend. Sorry we don't make this more clear. The
shuttles are in use today in an active experiment, so please try again
later this week, and make a reservation for a weekday.

> Ghinwa

David