Wired B210 setup with modified srsran

101 views
Skip to first unread message

Joshua Moore

unread,
May 31, 2023, 1:51:33 PM5/31/23
to Powder Users
I have 3 b210s (2 ues) wired with 30db attenuation between both recieve and transmit but when i turn on the modifications in srsran it stops working. Is there a certain configuration with values i should be using to get this working or is this realized only through channel emulation for running nexran xapp?
Message has been deleted

Joshua Moore

unread,
May 31, 2023, 5:01:15 PM5/31/23
to Powder Users
With the modified srsRAN i can attach 1 UE and exchange traffic, but when i attach another the throughput goes to 0 on the first and never reconnects. Is this due to the modifications?
Screenshot from 2023-05-31 14-30-52.png

Joshua Moore

unread,
May 31, 2023, 5:01:34 PM5/31/23
to Powder Users
Screenshot from 2023-05-31 15-57-30.png

Dustin Maas

unread,
Jun 7, 2023, 12:18:36 PM6/7/23
to Powder Users

This should work fine in a conducted RF setup. One thought.. are you using the same IMSI for both UEs?

--
You received this message because you are subscribed to the Google Groups "Powder Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to powder-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/powder-users/f0cf43df-8f4c-4e36-9938-a2dbe0b3c85an%40googlegroups.com.

Joshua Moore

unread,
Jun 7, 2023, 1:47:27 PM6/7/23
to powder...@googlegroups.com
That was the issue we were able to solve a few days ago. Nonetheless, thank you for your reply. We have it running but the throughput never changes. Is there a problem with the e release, we are trying other releases now 

On Jun 7, 2023, at 11:18 AM, Dustin Maas <dusti...@gmail.com> wrote:


You received this message because you are subscribed to a topic in the Google Groups "Powder Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/powder-users/TlaDCcPLzW8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to powder-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/powder-users/CAD91B33-DFEA-4C2A-AAF9-E5894A06B0F6%40gmail.com.

Dustin Maas

unread,
Jun 7, 2023, 3:01:38 PM6/7/23
to powder...@googlegroups.com

I wouldn’t suspect the release if everything else appears to be working properly. I think we’ll need logs in order to help here though.

Joshua Moore

unread,
Jun 7, 2023, 5:18:40 PM6/7/23
to powder...@googlegroups.com
I’m attaching our nexran logs. Are any other logs relevant? 

nexranlogs

Dustin Maas

unread,
Jun 8, 2023, 12:05:20 PM6/8/23
to powder...@googlegroups.com

These look like I’d expect. Please share logs for srsepc and srsenb, as well as the command line arguments you used when running srsenb.

Message has been deleted
Message has been deleted
Message has been deleted

Joshua Moore

unread,
Jun 8, 2023, 2:14:30 PM6/8/23
to Powder Users
None of my messages with attachments are going through

Joshua Moore

unread,
Jun 8, 2023, 2:17:45 PM6/8/23
to Powder Users

Dustin Maas

unread,
Jun 8, 2023, 3:24:01 PM6/8/23
to Powder Users

These look like logs for an instance where a single UE is attached via ZMQ. I thought we were talking about a 2-UE scenario with a conducted RF setup.

--

You received this message because you are subscribed to the Google Groups "Powder Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to powder-users...@googlegroups.com.

Joshua Moore

unread,
Jun 8, 2023, 3:29:08 PM6/8/23
to powder...@googlegroups.com
Sorry for the confusion. Yes you are correct. I was trying to make this work before trying the usrps

On Jun 8, 2023, at 2:24 PM, Dustin Maas <dusti...@gmail.com> wrote:


You received this message because you are subscribed to a topic in the Google Groups "Powder Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/powder-users/TlaDCcPLzW8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to powder-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/powder-users/DCAB6C68-51E0-4F5C-A2D6-FE97E5FE1EB3%40gmail.com.

Dustin Maas

unread,
Jun 9, 2023, 5:20:33 PM6/9/23
to Powder Users

These logs confirm what I noticed today, which is that I see the slicer methods that get run when a UE attaches printing logs to stdout, but I’m not seeing any of the logging I expect to see when a UE gets bound to a slice, slice share gets updated, etc. I’ll try to reproduce here with the same O-RAN release.

-Dustin

On 8 Jun 2023, at 11:30, Nisha Adhikari wrote:

The commands to run the srsenb are at the top of enblog.txt. Thanks


On Thursday, 8 June 2023 at 11:05:20 UTC-5 Dustin Maas wrote:

These look like I’d expect. Please share logs for srsepc and srsenb, as well as the command line arguments you used when running srsenb.

--

You received this message because you are subscribed to the Google Groups "Powder Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to powder-users...@googlegroups.com.

Dustin Maas

unread,
Jun 12, 2023, 11:43:13 AM6/12/23
to Powder Users, David M. Johnson

I wasn’t able to reproduce this, but it may be a RIC or RMR bug.. please try the following "fix":

kubectl -n ricplt rollout restart \
    deployments/deployment-ricplt-e2term-alpha \
    deployments/deployment-ricplt-e2mgr \
    deployments/deployment-ricplt-submgr \
    deployments/deployment-ricplt-rtmgr \
    deployments/deployment-ricplt-appmgr \
    statefulsets/statefulset-ricplt-dbaas-server

Joshua Moore

unread,
Jun 12, 2023, 4:41:16 PM6/12/23
to Powder Users
So we tried that and now it is printing slice report in the enb but still not changing the throughput.


[slicer] configuring slices...
[slicer] updating proportional sf allocation...
[slicer] slice: fast, proportional sf allocation: 5
[slicer] slice: slow, proportional sf allocation: 256
[slicer] getting slice status...
<2023-06-12T20:33:13.768007 [E2AP   ] [I] [    0] Received RICcontrolRequest

E2SM-NexRAN-ControlOutcome>2023-06-12T20:33:13.768009 [E2SM   ] [D] [    0] nexran: handle_control


    2023-06-12T20:33:13.768010 [E2SM   ] [D] [    0] control header:

<2023-06-12T20:33:13.768037 [E2SM   ] [D] [    0] control message:

controlOutcomeFormat12023-06-12T20:33:13.768153 [E2SM   ] [D] [    0] configured slice fast share 5

>2023-06-12T20:33:13.768170 [E2SM   ] [D] [    0] building slice status report:


2023-06-12T20:33:13.768180 [E2SM   ] [D] [    0] nexran slice status report:

        <sliceStatusReport>
            <sliceStatusList>
                <SliceStatus>
                    <sliceName>fast</sliceName>
                    <schedPolicy>
                        <proportionalAllocationPolicy>
                            <share>5</share>
                        </proportionalAllocationPolicy>
                    </schedPolicy>
                    <ueList>
                    </ueList>
                </SliceStatus>
                <SliceStatus>
                    <sliceName>slow</sliceName>
                    <schedPolicy>
                        <proportionalAllocationPolicy>
                            <share>256</share>
                        </proportionalAllocationPolicy>
                    </schedPolicy>
                    <ueList>
                        <UeStatus>
                            <imsi>31 30 31 30 31 32 33 34 35 36 37 38 38</imsi>
                            <connected><false/></connected>
                        </UeStatus>
                    </ueList>
                </SliceStatus>
            </sliceStatusList>
        </sliceStatusReport>
    </controlOutcomeFormat1>
</E2SM-NexRAN-ControlOutcome>
<E2AP-PDU>

David M. Johnson

unread,
Jun 12, 2023, 4:51:39 PM6/12/23
to powder...@googlegroups.com
On 6/12/23 14:41, Joshua Moore wrote:
> So we tried that and now it is printing slice report in the enb but
> still not changing the throughput.

This is good; now your E2 control messages are going through. (This
seems likely to be a bug in the RIC submgr or in RMR, where after the
RIC has been up for quite some time (hours), it will stop passing
Control messages to the RAN nodes. I have seen this a number of times,
but not attempted to debug e.g. via the rtmgr debuginfo API endpoint, so
can't say more beyond that.)

Anyway, now that your Control messages are coming through, can you
please check again that you have work-conserving mode disabled (the
`--slicer.workshare=0` argument to `srsenb`, as shown in
https://www.powderwireless.net/show/PowderProfiles/O-RAN)? Then you
will observe the throughput reduction even if you have only a single UE
connected.

David
> __
>
> I wasn’t able to reproduce this, but it may be a RIC or RMR bug..
> please try the following "fix":
>
> |kubectl -n ricplt rollout restart \
> deployments/deployment-ricplt-e2term-alpha \
> deployments/deployment-ricplt-e2mgr \
> deployments/deployment-ricplt-submgr \
> deployments/deployment-ricplt-rtmgr \
> deployments/deployment-ricplt-appmgr \
> statefulsets/statefulset-ricplt-dbaas-server |
>
> On 9 Jun 2023, at 15:20, Dustin Maas wrote:
>
> These logs confirm what I noticed today, which is that I see the
> slicer methods that get run when a UE attaches printing logs to
> stdout, but I’m not seeing any of the logging I expect to see
> when a UE gets bound to a slice, slice share gets updated, etc.
> I’ll try to reproduce here with the same O-RAN release.
>
> -Dustin
>
> On 8 Jun 2023, at 11:30, Nisha Adhikari wrote:
>
> The commands to run the srsenb are at the top of enblog.txt.
> Thanks
>
> On Thursday, 8 June 2023 at 11:05:20 UTC-5 Dustin Maas wrote:
>
> These look like I’d expect. Please share logs for
> |srsepc| and |srsenb|, as well as the command line
> arguments you used when running |srsenb|.
>
> --
> You received this message because you are subscribed to the
> Google Groups "Powder Users" group.
> To unsubscribe from this group and stop receiving emails
> from it, send an email to powder-users...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/powder-users/1bb30f31-edd8-47a3-9dcc-612184ea9787n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/1bb30f31-edd8-47a3-9dcc-612184ea9787n%40googlegroups.com?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "Powder Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to powder-users...@googlegroups.com
> <mailto:powder-users...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/powder-users/5799a9c1-24e6-4e09-b63f-5eeeefa47699n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/5799a9c1-24e6-4e09-b63f-5eeeefa47699n%40googlegroups.com?utm_medium=email&utm_source=footer>.

Joshua Moore

unread,
Jun 12, 2023, 6:12:16 PM6/12/23
to Powder Users
Thanks for the reply. We have that option enabled already. Is there anything else that it might be in your opinion that we can check?

Joshua Moore

unread,
Jun 12, 2023, 6:15:57 PM6/12/23
to Powder Users
The nexran logs print this </E2SM-KPM-IndicationMessage>
{"ts":1686607952866,"crit":"DEBUG","id":"nexran","mdc":{},"msg":"kpm indication report style 4 "}
{"ts":1686607952866,"crit":"DEBUG","id":"nexran","mdc":{},"msg":"nexran Indication handler"}
{"ts":1686607952866,"crit":"INFO","id":"nexran","mdc":{},"msg":"KpmIndication: KpmReport(period=5120 ms) available_dl_prbs=0 available_ul_prbs=0 "}
{"ts":1686607952866,"crit":"DEBUG","id":"nexran","mdc":{},"msg":"no slices in KPM report; not autoequalizing"}
{"ts":1686607952866,"crit":"DEBUG","id":"nexran","mdc":{},"msg":"current shares: fast[share=1024] slow[share=5] "}
{"ts":1686607957985,"crit":"DEBUG","id":"nexran","mdc":{},"msg":"RMR message (type 12050, source enB_macro_001_001_0019b0)"}
<E2AP-PDU>


which might mean something?
Message has been deleted

Joshua Moore

unread,
Jun 15, 2023, 10:08:23 PM6/15/23
to Powder Users
We were finally able to run nexran locally thanks for all your help. We had to install on a fresh Ubuntu install to correct the issue for some reason. 
Reply all
Reply to author
Forward
0 new messages