Groups keyboard shortcuts have been updated

Dismiss

See shortcuts

Error while setting up srslte-shvlan-oran

203 views

Skip to first unread message

Hariharan Venkatraman

unread,

Nov 12, 2024, 4:55:10 PM11/12/24

to Powder Users

Hi, this is Hariharan Venkatraman, graduate student from university at buffalo. I got this error in the image when tried to setup the srslte-shvlan-oran profile (only after setting up of my ORAN profile with shared VLAN. I am using one B210 ue under controlled RF environment with the same named shared VLAN in this profile. Could anyone tell me why am I facing this and what is the alternative?

image_720.png

Kirk Webb

unread,

Nov 12, 2024, 5:15:52 PM11/12/24

to powder...@googlegroups.com

Assuming you have not terminated your experiment yet, please provide
the link (URL) to it on the POWDER Portal. The mapper had trouble with
what you specifically asked for.

-Kirk

On Tue, Nov 12, 2024 at 2:55 PM Hariharan Venkatraman
<hariharanve...@gmail.com> wrote:
>
> Hi, this is Hariharan Venkatraman, graduate student from university at buffalo. I got this error in the image when tried to setup the srslte-shvlan-oran profile (only after setting up of my ORAN profile with shared VLAN. I am using one B210 ue under controlled RF environment with the same named shared VLAN in this profile. Could anyone tell me why am I facing this and what is the alternative?
>
>

> --
> 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 visit https://groups.google.com/d/msgid/powder-users/23c7dd44-a845-4d18-8e5c-5412a56942dcn%40googlegroups.com.

Hariharan Venkatraman

unread,

Nov 12, 2024, 5:17:37 PM11/12/24

to Powder Users

Powder - Experiment Status

Kirk Webb

unread,

Nov 12, 2024, 10:52:38 PM11/12/24

to powder...@googlegroups.com

We're still looking into this issue. It appears that there may be some
state missing from the database. Stay tuned.

-Kirk

On Tue, Nov 12, 2024 at 3:18 PM Hariharan Venkatraman

<hariharanve...@gmail.com> wrote:
>
> Powder - Experiment Status
>
> On Tuesday, November 12, 2024 at 5:15:52 PM UTC-5 kw...@cs.utah.edu wrote:
>>
>> Assuming you have not terminated your experiment yet, please provide
>> the link (URL) to it on the POWDER Portal. The mapper had trouble with
>> what you specifically asked for.
>>
>> -Kirk
>>
>>
>> On Tue, Nov 12, 2024 at 2:55 PM Hariharan Venkatraman
>> <hariharanve...@gmail.com> wrote:
>> >
>> > Hi, this is Hariharan Venkatraman, graduate student from university at buffalo. I got this error in the image when tried to setup the srslte-shvlan-oran profile (only after setting up of my ORAN profile with shared VLAN. I am using one B210 ue under controlled RF environment with the same named shared VLAN in this profile. Could anyone tell me why am I facing this and what is the alternative?
>> >
>> >
>> > --
>> > 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 visit https://groups.google.com/d/msgid/powder-users/23c7dd44-a845-4d18-8e5c-5412a56942dcn%40googlegroups.com.
>
> --
> 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 visit https://groups.google.com/d/msgid/powder-users/d5a56e8e-c693-4ec3-8cfb-33febf22da65n%40googlegroups.com.

Kirk Webb

unread,

Nov 13, 2024, 12:13:00 PM11/13/24

to powder...@googlegroups.com

Hi,

Can you please elaborate on how you instantiated this experiment?
Providing a screenshot of the profile parameters you have set/selected
would be useful. The set of available resources - as presented to the
mapping algorithm yesterday - were incomplete, which is unexpected. We
are trying to understand how that might have happened. After taking
the screenshot of your parameters, please go ahead and try to
instantiate again, and then let us know how that goes.

-Kirk

Hariharan Venkatraman

unread,

Nov 13, 2024, 3:07:14 PM11/13/24

to Powder Users

These are my setup and still i am getting the same error.

srslte-shvlan-oran setup.png

oran setup.png

Hariharan Venkatraman

unread,

Nov 13, 2024, 4:29:14 PM11/13/24

to Powder Users

This is the error i am getting right now. This has been the case for past 4 days. Could anyone check this out and tell me what could be the solution?

profile error.png

Hariharan Venkatraman

unread,

Nov 18, 2024, 3:00:52 PM11/18/24

to Powder Users

Is this issue sorted out by the powder team?

Best,
Hariharan Venkatraman

Kirk Webb

unread,

Nov 18, 2024, 5:47:55 PM11/18/24

to powder...@googlegroups.com

The PhantomNet attenuator matrix has been occupied (by user
experiments), so I haven't been able to test/reproduce your issue. I
would recommend that you make a reservation for `nuc1`, `nuc2`, and
one d430 node (for the O-RAN/5G piece). This will ensure the resources
are available during a reservation window of your choosing, and we can
also have a look if things aren't working at that time. I don't
believe you had a reservation when you were recently trying to
instantiate.

-Kirk

On Mon, Nov 18, 2024 at 1:01 PM Hariharan Venkatraman

> To view this discussion visit https://groups.google.com/d/msgid/powder-users/a379309a-acf9-4234-bcb4-3c3228080ca4n%40googlegroups.com.

Hariharan Venkatraman

unread,

Nov 19, 2024, 7:50:46 PM11/19/24

to Powder Users

Could you tell me precisely which profile I need to use to perform my above task?
if possible the setup?
is the srslte-shvlan-oran profile is working?

Best,
Hariharan Venkatraman

Kirk Webb

unread,

Nov 20, 2024, 9:20:04 AM11/20/24

to powder...@googlegroups.com

I would suggest you sign up for an office hours slot (coming up this
Friday) so we can go over the details more readily:

https://calendly.com/powderwireless

-Kirk

On Tue, Nov 19, 2024 at 5:51 PM Hariharan Venkatraman

> To view this discussion visit https://groups.google.com/d/msgid/powder-users/0661bada-f11b-4055-a8f4-fa0ce914eb83n%40googlegroups.com.

Hariharan Venkatraman

unread,

Nov 23, 2024, 10:03:36 PM11/23/24

to Powder Users

I want to know what data Grafana is using to display all the graphs of the powder experiment. Could you please share any documents regarding that? It is very hard to debug with limited information.

Best,
Hariharan Venkatraman

David M Johnson

unread,

Nov 24, 2024, 4:34:59 PM11/24/24

to powder...@googlegroups.com

I'm assuming you are trying to run the NexRAN demo instructions and view
the data in the O-RAN profile's Grafana instance?

If so: the details of what is logged is not documented. However, the
NexRAN code is easy to read and shows you which metrics are being
logged:
https://gitlab.flux.utah.edu/powderrenewpublic/nexran/-/blob/master/src/nexran.cc?ref_type=heads#L189
. Both per-slice and per-UE metrics are being logged.

If not, please provide links to your experiments and provide a bit more
context... thanks.

David

> >> >>>> > >
> >> >>>> > > --
> >> >>>> > > 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 visit

> >> >
> >> > --
> >> > 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 visit

> >
> > --
> > 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 visit

>
> --
> 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 visit

Hariharan Venkatraman

unread,

Nov 25, 2024, 1:51:00 PM11/25/24

to Powder Users

The experiments I am currently doing are:

1. Powder - Show Profile

2. Powder - Show Profile

Best,
Hariharan Venkatraman

David M Johnson

unread,

Nov 25, 2024, 5:13:37 PM11/25/24

to powder...@googlegroups.com

Thanks. These are pointers to profiles that can be used to instantiate
experiments, but not the experiments themselves. Once you have these
running as experiments, if you are having a problem, then please provide
links to each experiment's status page, and more context about the
problem you're facing.

Did the code pointer I sent before resolve your question about what
NexRAN logs to Grafana?

David

On 11/25/24 11:51, Hariharan Venkatraman wrote:
> The experiments I am currently doing are:
> 1. Powder - Show Profile

> <https://www.powderwireless.net/show-profile.php?uuid=30d3e13c-f939-11ea-b1eb-e4434b2381fc>

> 2. Powder - Show Profile

> <https://www.powderwireless.net/show-profile.php?uuid=1bd95656-5b60-11eb-b1eb-e4434b2381fc>

>
>
> Best,
> Hariharan Venkatraman
> On Sunday, November 24, 2024 at 4:34:59 PM UTC-5 john...@flux.utah.edu
> wrote:
>
> I'm assuming you are trying to run the NexRAN demo instructions and
> view
> the data in the O-RAN profile's Grafana instance?
>
> If so: the details of what is logged is not documented. However, the
> NexRAN code is easy to read and shows you which metrics are being
> logged:

> > >> >>>> > >
> > >> >>>> > > --
> > >> >>>> > > 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 visit
> >

> > >> >
> > >> > --
> > >> > 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 visit
> >

> > >
> > > --
> > > 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 visit
> >

> >
> > --
> > 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 visit
> >

Hariharan Venkatraman

unread,

Nov 26, 2024, 12:54:50 PM11/26/24

to Powder Users

Right now I am performing these demos to see whether my RL code is optimizing the resources correctly (that's the goal), based on this later will design my own experiment.
Right now, When i try to see the dashboard in Grafana, there aren't enough metrics to better understand the graphs that it is showing.
1. what kind of dashboard is it (USE/REDS, drill down type, or anything else)?
2. Apart from the visible partial information, how exactly is that specific graph logging and what are its other attributes(units, threshold if any) to better understand the performance?
3. Am I allowed to edit it according to my wish (let's say I want the data visualized in MB and not in bytes)? If so, how do I do it?

I will attach some sample graphs I got through my experiment.

Best,
Hariharan Venkatraman

65291d0537ac5ff50db3cffa7640b1efcc575362.jpg

5e40c9db09456add92697cf5192e164df5314c51.jpg

David M Johnson

unread,

Nov 26, 2024, 3:13:27 PM11/26/24

to powder...@googlegroups.com

On 11/26/24 10:54, Hariharan Venkatraman wrote:
> Right now I am performing these demos to see whether my RL code is
> optimizing the resources correctly (that's the goal), based on this
> later will design my own experiment.

Thanks. I'm still trying to understand your goals, but it sounds like
you are trying to extract a dataset from this demo and build an RL model
for optimizing RAN sharing? And integrate that into an xApp?

> Right now, When i try to see the dashboard in Grafana, there aren't
> enough metrics to better understand the graphs that it is showing.
> 1. what kind of dashboard is it (USE/REDS, drill down type, or anything
> else)?

It is a purpose-built dashboard to graph the metrics I pointed you to
before:
https://gitlab.flux.utah.edu/powderrenewpublic/nexran/-/blob/master/src/nexran.cc?ref_type=heads#L189
. These are a variety of PDCP, MAC, and high-PHY metrics, for uplink
and downlink, sent to influxdb for each UE per time quanta (1024ms by
default), and for each slice (aggregated across UEs in the slice). The
dashboards that show these metrics are closest in concept to the USE
approach, but it's not a perfect correspondence. Higher-level metrics
such as byte counters represent PDCP throughput. Lower-level metrics
like CQI, PMI, RI, MCS function as channel quality/status indicators.

> 2. Apart from the visible partial information, how exactly is that
> specific graph logging and what are its other attributes(units,
> threshold if any) to better understand the performance?

Counter/rate metrics (bytes, PRBs, errors, brate, pkts) all have
threshold zero and the unit corresponds to the field name (for brate,
you can see how srslte computes those here:
https://gitlab.flux.utah.edu/powderrenewpublic/srslte-ric/-/blob/oran-ric/srsenb/src/metrics_csv.cc?ref_type=heads#L78).
For the lower-level metrics like CQI, PMI, RI, MCS, etc, you can refer
to LTE specs or summaries to understand the value ranges and why they
are useful indicators of channel quality.

> 3. Am I allowed to edit it according to my wish (let's say I want the
> data visualized in MB and not in bytes)? If so, how do I do it?

Yes, you can edit these dashboards and create your own that more closely
match the USE pattern if you like. When you create an experiment from
the O-RAN profile, it is stood up from bare metal and only accessible by
yourself and members of your project. You can change any aspect of the
software in your experiment as required.

> I will attach some sample graphs I got through my experiment.
>
> Best,
> Hariharan Venkatraman

David

> On Monday, November 25, 2024 at 5:13:37 PM UTC-5 john...@flux.utah.edu
> wrote:
>
> Thanks. These are pointers to profiles that can be used to instantiate
> experiments, but not the experiments themselves. Once you have these
> running as experiments, if you are having a problem, then please
> provide
> links to each experiment's status page, and more context about the
> problem you're facing.
>
> Did the code pointer I sent before resolve your question about what
> NexRAN logs to Grafana?
>
>
> David
>
> On 11/25/24 11:51, Hariharan Venkatraman wrote:
> > The experiments I am currently doing are:
> > 1. Powder - Show Profile
> >

> <https://www.powderwireless.net/show-profile.php?uuid=30d3e13c-f939-11ea-b1eb-e4434b2381fc <https://www.powderwireless.net/show-profile.php?uuid=30d3e13c-f939-11ea-b1eb-e4434b2381fc>>

> > 2. Powder - Show Profile
> >

> <https://www.powderwireless.net/show-profile.php?uuid=1bd95656-5b60-11eb-b1eb-e4434b2381fc <https://www.powderwireless.net/show-profile.php?uuid=1bd95656-5b60-11eb-b1eb-e4434b2381fc>>

> >
> >
> > Best,
> > Hariharan Venkatraman
> > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> john...@flux.utah.edu
> > wrote:
> >
> > I'm assuming you are trying to run the NexRAN demo instructions and
> > view
> > the data in the O-RAN profile's Grafana instance?
> >
> > If so: the details of what is logged is not documented. However, the
> > NexRAN code is easy to read and shows you which metrics are being
> > logged:
> >

> > > >> >>>> > >
> > > >> >>>> > > --
> > > >> >>>> > > 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 visit
> > >
> >

> > > >> >
> > > >> > --
> > > >> > 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 visit
> > >
> >

> > > >
> > > > --
> > > > 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 visit
> > >
> >

> > >
> > > --
> > > 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 visit
> > >
> >

Hariharan Venkatraman

unread,

Nov 27, 2024, 11:17:38 AM11/27/24

to Powder Users

thanks for the above explanation, I will get back to you if I have any doubts.
The issue i am currently facing is while establishing ue. What is the IMSI and rf device name i need to have in my code
(the one now i am using to establish the ue is:
sudo IP netns add ue1
sudo srsue --gw.netns=ue1 --usim.algo=xor --usim.imsi=001010123456789 --rf.device_name=zmq --rf.device_args="tx_port=tcp://*:2001,rx_port=tcp://localhost:2000,id=ue,base_srate=23.04e6"
)

experiment: ( Powder - Experiment Status ) performing this according to instructions in the profile and instead of kpimon xapp I am using the nexran xapp ( instruction from Powder - Experiment Status ). Instead, of the demo script I have my Python script with RL for optimization.

so this is the experiment i am doing and i couldn't able to establish the ue and nexran xapp and the iperf client shows "bad file descriptor" error even after multiple reload/reboot of nodes.

could you help me find the correct imsi and RF tag to set up the ue?

Best,
Haiharan Venkatraman

Hariharan Venkatraman

unread,

Dec 2, 2024, 12:21:58 PM12/2/24

to Powder Users

can anyone clarify this?

Best,
Hariharan Venkatraman

David M Johnson

unread,

Dec 2, 2024, 5:31:57 PM12/2/24

to powder...@googlegroups.com

On 12/2/24 10:21, Hariharan Venkatraman wrote:
> can anyone clarify this?

Sorry, we were not able to look at the experiment you provided, due to
the holiday weekend. Additional responses below, inline...

> Best,
> Hariharan Venkatraman
>
> On Wednesday, November 27, 2024 at 11:17:38 AM UTC-5 Hariharan
> Venkatraman wrote:
>
> thanks for the above explanation, I will get back to you if I have
> any doubts.
> The issue i am currently facing is while establishing ue. What is

I just tested the demo instructions that you mention, and those worked
for me. One thing that sometimes goes wrong is if the EPC user
configuration (/etc/srslte/user_db.csv) is updated while srsepc is
running. srsepc rewrites the file when it exits, so any changes you
make in the interim will be overwritten.

> the IMSI and rf device name i need to have in my code
> (the one now i am using to establish the ue is:
> sudo IP netns add ue1

In this fragment, `IP` should be `ip` -- perhaps the issue is that
simple? It is `ip` in the profile instructions.

> sudo srsue --gw.netns=ue1 --usim.algo=xor
> --usim.imsi=001010123456789 --rf.device_name=zmq
> --rf.device_args="tx_port=tcp://*:2001,rx_port=tcp://localhost:2000,id=ue,base_srate=23.04e6"
> )
>
>
> experiment: ( Powder - Experiment Status

> <https://www.powderwireless.net/status.php?uuid=c14a08ce-ab5d-11ef-af1a-e4434b2381fc> ) performing this according to instructions in the profile and instead of kpimon xapp I am using the nexran xapp ( instruction from Powder - Experiment Status <https://www.powderwireless.net/status.php?uuid=f4cb0007-ab55-11ef-af1a-e4434b2381fc> ). Instead, of the demo script I have my Python script with RL for optimization.

>
> so this is the experiment i am doing and i couldn't able to
> establish the ue and nexran xapp and the iperf client shows "bad
> file descriptor" error even after multiple reload/reboot of nodes.
>
> could you help me find the correct imsi and RF tag to set up the ue?
>
>
> Best,
> Haiharan Venkatraman

David

> On Tuesday, November 26, 2024 at 3:13:27 PM UTC-5
> john...@flux.utah.edu wrote:
>
> On 11/26/24 10:54, Hariharan Venkatraman wrote:
> > Right now I am performing these demos to see whether my RL
> code is
> > optimizing the resources correctly (that's the goal), based
> on this
> > later will design my own experiment.
>
> Thanks. I'm still trying to understand your goals, but it sounds
> like
> you are trying to extract a dataset from this demo and build an
> RL model
> for optimizing RAN sharing? And integrate that into an xApp?
>
> > Right now, When i try to see the dashboard in Grafana, there
> aren't
> > enough metrics to better understand the graphs that it is
> showing.
> > 1. what kind of dashboard is it (USE/REDS, drill down type,
> or anything
> > else)?
>
> It is a purpose-built dashboard to graph the metrics I pointed
> you to
> before:

> . These are a variety of PDCP, MAC, and high-PHY metrics, for
> uplink
> and downlink, sent to influxdb for each UE per time quanta
> (1024ms by
> default), and for each slice (aggregated across UEs in the
> slice). The
> dashboards that show these metrics are closest in concept to the
> USE
> approach, but it's not a perfect correspondence. Higher-level
> metrics
> such as byte counters represent PDCP throughput. Lower-level
> metrics
> like CQI, PMI, RI, MCS function as channel quality/status
> indicators.
>
> > 2. Apart from the visible partial information, how exactly is
> that
> > specific graph logging and what are its other attributes(units,
> > threshold if any) to better understand the performance?
>
> Counter/rate metrics (bytes, PRBs, errors, brate, pkts) all have
> threshold zero and the unit corresponds to the field name (for
> brate,
> you can see how srslte computes those here:

> > > 2. Powder - Show Profile
> > >
> >

> > >
> > >
> > > Best,
> > > Hariharan Venkatraman
> > > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> > john...@flux.utah.edu
> > > wrote:
> > >
> > > I'm assuming you are trying to run the NexRAN demo
> instructions and
> > > view
> > > the data in the O-RAN profile's Grafana instance?
> > >
> > > If so: the details of what is logged is not documented.
> However, the
> > > NexRAN code is easy to read and shows you which metrics are
> being
> > > logged:
> > >
> >

> > > > >> >>>> > >
> > > > >> >>>> > > --
> > > > >> >>>> > > 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 visit
> > > >
> > >
> >

> > > > >> >
> > > > >> > --
> > > > >> > 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 visit
> > > >
> > >
> >

> > > > >
> > > > > --
> > > > > 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 visit
> > > >
> > >
> >

> > > >
> > > > --
> > > > 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 visit
> > > >
> > >
> >

> > >
> > > --
> > > 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 visit
> > >
> >

> >
> > --
> > 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 visit
> >

Hariharan Venkatraman

unread,

Dec 3, 2024, 6:57:09 PM12/3/24

to Powder Users

i haven't changed anything with epc
i just followed the profile instructions and once the epc is set the next is nodeb
with "IP" correction, i did with small letters only (it was typo in my previous conversation)

1. is my imsi and rf tag is correct? i don't think zmq is correct if not then what is the correct tag values for these (from core infrastructure)
2. could you share me the logic (what data is logged, how it is visualized (single value/aggregate)), behind the computation of few of the graphs that i got from grafana ? (attached) --> while performing the " Running the NexRAN xApp and the first RAN slicing demo " from "Profile PowderProfiles/O-RAN"

Note: i am just doing the demo experiment and instead of demo script i have my own python with RL script for resource allocation

Best,
Hariharan Venkatraman

65291d0537ac5ff50db3cffa7640b1efcc575362.jpg

49d4adb6256d4109e46776a725e724a13e1a7f2b.jpg

5e40c9db09456add92697cf5192e164df5314c51.jpg

David M Johnson

unread,

Dec 3, 2024, 10:16:16 PM12/3/24

to powder...@googlegroups.com

On 12/3/24 16:57, Hariharan Venkatraman wrote:
> i haven't changed anything with epc
> i just followed the profile instructions and once the epc is set the
> next is nodeb
> with "IP" correction, i did with small letters only (it was typo in my
> previous conversation)
>
> 1. is my imsi and rf tag is correct? i don't think zmq is correct if not
> then what is the correct tag values for these (from core infrastructure)

Your imsi matches the profile instructions. However, your srsue command
in the prior message doesn't include
`--usim.k=00112233445566778899aabbccddeeff --usim.imei=353490069873310`
. Maybe that is the issue.

> 2. could you share me the logic (what data is logged, how it is
> visualized (single value/aggregate)), behind the computation of few of
> the graphs that i got from grafana ? (attached) --> while performing
> the " Running the NexRAN xApp and the first RAN slicing demo " from
> "Profile PowderProfiles/O-RAN"

I tried to explain this in an earlier reply; please see
https://groups.google.com/g/powder-users/c/7s6bB3tv1aA/m/LS_UM7jNAAAJ .
You can see exactly what is being sent to influxdb (both per-ue and
per-slice data is being sent for each slice/ue in each measurement
report), and then you can examine the grafana dashboard queries to see
how the per-ue and per-slice data is being loaded.

> Note: i am just doing the demo experiment and instead of demo script i
> have my own python with RL script for resource allocation

The command lines you are reporting don't match the demo instructions in
some of your messages. Maybe double-check when you have problems, where
you've made changes or customized them?

> Hariharan Venkatraman

David

> <https://www.powderwireless.net/status.php?uuid=c14a08ce-ab5d-11ef-af1a-e4434b2381fc <https://www.powderwireless.net/status.php?uuid=c14a08ce-ab5d-11ef-af1a-e4434b2381fc>> ) performing this according to instructions in the profile and instead of kpimon xapp I am using the nexran xapp ( instruction from Powder - Experiment Status <https://www.powderwireless.net/status.php?uuid=f4cb0007-ab55-11ef-af1a-e4434b2381fc <https://www.powderwireless.net/status.php?uuid=f4cb0007-ab55-11ef-af1a-e4434b2381fc>> ). Instead, of the demo script I have my Python script with RL for optimization.

> > . These are a variety of PDCP, MAC, and high-PHY metrics, for
> > uplink
> > and downlink, sent to influxdb for each UE per time quanta
> > (1024ms by
> > default), and for each slice (aggregated across UEs in the
> > slice). The
> > dashboards that show these metrics are closest in concept to the
> > USE
> > approach, but it's not a perfect correspondence. Higher-level
> > metrics
> > such as byte counters represent PDCP throughput. Lower-level
> > metrics
> > like CQI, PMI, RI, MCS function as channel quality/status
> > indicators.
> >
> > > 2. Apart from the visible partial information, how exactly is
> > that
> > > specific graph logging and what are its other attributes(units,
> > > threshold if any) to better understand the performance?
> >
> > Counter/rate metrics (bytes, PRBs, errors, brate, pkts) all have
> > threshold zero and the unit corresponds to the field name (for
> > brate,
> > you can see how srslte computes those here:
> >

> > > > 2. Powder - Show Profile
> > > >
> > >
> >

> > > >
> > > >
> > > > Best,
> > > > Hariharan Venkatraman
> > > > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> > > john...@flux.utah.edu
> > > > wrote:
> > > >
> > > > I'm assuming you are trying to run the NexRAN demo
> > instructions and
> > > > view
> > > > the data in the O-RAN profile's Grafana instance?
> > > >
> > > > If so: the details of what is logged is not documented.
> > However, the
> > > > NexRAN code is easy to read and shows you which metrics are
> > being
> > > > logged:
> > > >
> > >
> >

> > > > > >> >>>> > >
> > > > > >> >>>> > > --
> > > > > >> >>>> > > 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 visit
> > > > >
> > > >
> > >
> >

> > > > > >> >
> > > > > >> > --
> > > > > >> > 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 visit
> > > > >
> > > >
> > >
> >

> > > > > >
> > > > > > --
> > > > > > 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 visit
> > > > >
> > > >
> > >
> >

> > > > >
> > > > > --
> > > > > 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 visit
> > > > >
> > > >
> > >
> >

> > > >
> > > > --
> > > > 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 visit
> > > >
> > >
> >

> > >
> > > --
> > > 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 visit
> > >
> >

Hariharan Venkatraman

unread,

Dec 10, 2024, 3:05:39 PM12/10/24

to Powder Users

I couldn't establish the UE network even after using the tags as you suggested. Also i am using NexranXapp instead of KPIMON Xapp.
here is what i did:
e2-term from node-o
epc and enodeb instantiate in m-nb-1 node
ue instantiate in m-ue-1 node
The ue code i am using right now (rest all are same as in profile):

sudo ip netns add ue1

sudo srsue --gw.netns=ue1 --usim.algo=xor --usim.imsi=001010123456789 --rf.device_name=zmq --rf.device_args="tx_port=tcp://*:2001,rx_port=tcp://localhost:2000,id=ue,base_srate=23.04e6" --usim.k=00112233445566778899aabbccddeeff --usim.imei=353490069873310

still couldn't get the ue connected status.

my experiment profiles: Powder - Experiment Status Powder - Experiment Status

Best,
Hariharan Venkatraman

image_720.png

Hariharan Venkatraman

unread,

Dec 11, 2024, 1:10:18 PM12/11/24

to Powder Users

could anyone clarify this

Best,
Hariharan Venkatraman

David M Johnson

unread,

Dec 11, 2024, 1:50:36 PM12/11/24

to powder...@googlegroups.com

Hariharan,

I've been tied up with an event the past two days, but this ends soon.
I think it would be most effective to do a quick office hour since we
were unable to meet last week, and since you have resources allocated.
I am available today at 1-3pm MT; please contact me off-list with a
30-min slot in that range, if something in there works for you.

In the meantime, I see two experiments below: an O-RAN experiment, and
SDRs connected to our RF matrix environment. The UE command you are
using is still running in simulated mode. When you use real radios that
are "connected" via the RF attenuator matrix, you must remove the
--rf.device_name and --rf.device_args (or tailor them to the SDRs on the
radios). This may solve your problem.

Thanks,

David

On 12/11/24 11:10, Hariharan Venkatraman wrote:
> could anyone clarify this
>
> Best,
> Hariharan Venkatraman
>
> On Tuesday, December 10, 2024 at 3:05:39 PM UTC-5 Hariharan Venkatraman
> wrote:
>
> I couldn't establish the UE network even after using the tags as you
> suggested. Also i am using NexranXapp instead of KPIMON Xapp.
> here is what i did:
> e2-term from node-o
> epc and enodeb instantiate in m-nb-1 node
> ue instantiate in m-ue-1 node
> The ue code i am using right now (rest all are same as in profile):
>
> sudo ip netns add ue1
>
> sudo srsue --gw.netns=ue1 --usim.algo=xor
> --usim.imsi=001010123456789 --rf.device_name=zmq
> --rf.device_args="tx_port=tcp://*:2001,rx_port=tcp://localhost:2000,id=ue,base_srate=23.04e6" --usim.k=00112233445566778899aabbccddeeff --usim.imei=353490069873310
>
> still couldn't get the ue connected status.
>
> my experiment profiles: Powder - Experiment Status

> <https://www.powderwireless.net/status.php?uuid=9797a539-b66b-11ef-af1a-e4434b2381fc> Powder - Experiment Status <https://www.powderwireless.net/status.php?uuid=e5477f70-b66b-11ef-af1a-e4434b2381fc>

> https://groups.google.com/g/powder-users/c/7s6bB3tv1aA/m/LS_UM7jNAAAJ <https://groups.google.com/g/powder-users/c/7s6bB3tv1aA/m/LS_UM7jNAAAJ> .

> > > > > 2. Powder - Show Profile
> > > > >
> > > >
> > >
> >

> > > > >
> > > > >
> > > > > Best,
> > > > > Hariharan Venkatraman
> > > > > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> > > > john...@flux.utah.edu
> > > > > wrote:
> > > > >
> > > > > I'm assuming you are trying to run the NexRAN demo
> > > instructions and
> > > > > view
> > > > > the data in the O-RAN profile's Grafana instance?
> > > > >
> > > > > If so: the details of what is logged is not documented.
> > > However, the
> > > > > NexRAN code is easy to read and shows you which metrics
> are
> > > being
> > > > > logged:
> > > > >
> > > >
> > >
> >

> > > > > > >> >>>> > >
> > > > > > >> >>>> > > --
> > > > > > >> >>>> > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > >> >
> > > > > > >> > --
> > > > > > >> > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > >
> > > > > > > --
> > > > > > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > >
> > > > > > --
> > > > > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > >
> > > > > --
> > > > > 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 visit
> > > > >
> > > >
> > >
> >

> > > >
> > > > --
> > > > 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 visit
> > > >
> > >
> >

Hariharan Venkatraman

unread,

Dec 11, 2024, 3:17:34 PM12/11/24

to Powder Users

i am ready to meet anytime today. Let me know which is right time for you and send the meet link.

Best,
Hariharan Venkatraman

Hariharan Venkatraman

unread,

Dec 12, 2024, 8:17:10 PM12/12/24

to Powder Users

I uninstalled the Nexran Xapp using "/local/setup/oran/dms_cli uninstall --xapp_chart_name=nexran --namespace=ricxapp" and reinstalled with "/local/setup/oran/dms_cli install --xapp_chart_name=nexran --version=0.1.0 --namespace=ricxapp"

When I restarted everything from the beginning, I couldn't establish the iperf server (attached image), stating it could allocate me that IP address. How can I rectify this?

Best,
Hariharan Venkatraman

error with iperf server.png

David M Johnson

unread,

Dec 12, 2024, 8:38:08 PM12/12/24

to powder...@googlegroups.com

Hi Hari,

If you can't start the iperf server because the address is not
available, that means your srsepc is not running. srsepc creates a
virtual interface with the 192.168.0.1 address when it starts up.

David

> > > > > > 2. Powder - Show Profile
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > >
> > > > > >
> > > > > > Best,
> > > > > > Hariharan Venkatraman
> > > > > > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> > > > > john...@flux.utah.edu
> > > > > > wrote:
> > > > > >
> > > > > > I'm assuming you are trying to run the NexRAN demo
> > > > instructions and
> > > > > > view
> > > > > > the data in the O-RAN profile's Grafana instance?
> > > > > >
> > > > > > If so: the details of what is logged is not documented.
> > > > However, the
> > > > > > NexRAN code is easy to read and shows you which metrics
> > are
> > > > being
> > > > > > logged:
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > >> >>>> > >
> > > > > > > >> >>>> > > --
> > > > > > > >> >>>> > > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > >> >
> > > > > > > >> > --
> > > > > > > >> > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > >
> > > > > > > > --
> > > > > > > > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > >
> > > > > > > --
> > > > > > > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > >
> > > > > > --
> > > > > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > >
> > > > > --
> > > > > 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 visit
> > > > >
> > > >
> > >
> >

Hariharan Venkatraman

unread,

Dec 13, 2024, 12:11:35 PM12/13/24

to Powder Users

but my srsepc is up and working when i instantiate it and then i got the user connected status as well. But i got the same issue while setting up iperf server.

Best,
Hariharan Venkatraman

David M Johnson

unread,

Dec 14, 2024, 7:19:40 AM12/14/24

to powder...@googlegroups.com

Hi Hari,

Perhaps then you are running on the UE node instead of the NB node?
Check to see if there is an interface with that hardware. If the IP
address exists on the node on which you're running the command, maybe
there is a stale iperf server running bound to the port; kill it if so.

David

> > > > > > > 2. Powder - Show Profile
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > >
> > > > > > >
> > > > > > > Best,
> > > > > > > Hariharan Venkatraman
> > > > > > > On Sunday, November 24, 2024 at 4:34:59 PM UTC-5
> > > > > > john...@flux.utah.edu
> > > > > > > wrote:
> > > > > > >
> > > > > > > I'm assuming you are trying to run the NexRAN demo
> > > > > instructions and
> > > > > > > view
> > > > > > > the data in the O-RAN profile's Grafana instance?
> > > > > > >
> > > > > > > If so: the details of what is logged is not documented.
> > > > > However, the
> > > > > > > NexRAN code is easy to read and shows you which metrics
> > > are
> > > > > being
> > > > > > > logged:
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > > >> >>>> > >
> > > > > > > > >> >>>> > > --
> > > > > > > > >> >>>> > > 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 visit
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > > >> >
> > > > > > > > >> > --
> > > > > > > > >> > 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 visit
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > > >
> > > > > > > > > --
> > > > > > > > > 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 visit
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > > >
> > > > > > > > --
> > > > > > > > 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 visit
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > > >
> > > > > > > --
> > > > > > > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > > >
> > > > > > --
> > > > > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

> > > > >
> > > > > --
> > > > > 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 visit
> > > > >
> > > >
> > >
> >

> > > >
> > > > --
> > > > 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 visit
> > > >
> > >
> >

> https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com> <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com>> <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com?utm_medium=email&utm_source=footer> <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/68ece741-28e6-4ba7-9f11-f02e15227c29n%40googlegroups.com?utm_medium=email&utm_source=footer>>>.

> https://groups.google.com/d/msgid/powder-users/a8c55742-2d17-4960-af2f-d562ea7370c7n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/a8c55742-2d17-4960-af2f-d562ea7370c7n%40googlegroups.com> <https://groups.google.com/d/msgid/powder-users/a8c55742-2d17-4960-af2f-d562ea7370c7n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/a8c55742-2d17-4960-af2f-d562ea7370c7n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

> https://groups.google.com/d/msgid/powder-users/d4906ffd-d0da-484b-bfd9-cfd1d73f69cdn%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/d4906ffd-d0da-484b-bfd9-cfd1d73f69cdn%40googlegroups.com?utm_medium=email&utm_source=footer>.

Hariharan Venkatraman

unread,

Dec 15, 2024, 5:06:21 PM12/15/24

to Powder Users

Made a reservation again (because even after rebooting the experiment after
initiating the O-RAN experiment, the IP address not connected problem comes
up while starting enodeB. Also, this issue started because i tried to
delete the nexran xapp slice and create it and try to bound it again. will
try as it is this time (after the reservation is approved).

Best,
Hariharan Venkatraman

On Saturday, December 14, 2024 at 7:19:40 AM UTC-5 john...@flux.utah.edu

Hariharan Venkatraman

unread,

Dec 18, 2024, 11:03:36 AM12/18/24

to Powder Users

Experiment: shvlan-srslte-oran (using nexranxapp steps in o-ran profile
instead of kpimon xapp)

epc, enodeb and ue is up and working with correct imsi
changed imsi via cd /local/repository/demo for nexran xapp and it is up and
working (getting logs) --> are there any other ways to confirm this i have
a strong doubt in this
iperf server and client is up and working
but couldn't get the data from influx db
after the initial steps upto starting ue when i try to go into node 0 again
for nexran xapp and perform whatsoever command i am getting this wierd
disconnection in nodeb and epc

and no matter what i perform the next steps i couldn't able to get the data

Best,
Hariharan Venkatraman

On Sunday, December 15, 2024 at 5:06:21 PM UTC-5 Hariharan Venkatraman

image_720.png

David M Johnson

unread,

Dec 18, 2024, 1:00:54 PM12/18/24

to powder...@googlegroups.com

On 12/18/24 09:03, Hariharan Venkatraman wrote:
> Experiment: shvlan-srslte-oran (using nexranxapp steps in o-ran profile
> instead of kpimon xapp)
>
> epc, enodeb and ue is up and working with correct imsi
> changed imsi via cd /local/repository/demo for nexran xapp and it is up
> and working (getting logs) --> are there any other ways to confirm this

What is in the nexran xapp log? Each time the enb sends an E2 KPM
indication (metrics report), nexran will log it before sending it to
influx. Check via `kubectl logs -f -n ricxapp -l app=ricxapp-nexran`
while your eNodeB is running.

> i have a strong doubt in this
> iperf server and client is up and working
> but couldn't get the data from influx db
> after the initial steps upto starting ue when i try to go into node 0
> again for nexran xapp and perform whatsoever command i am getting this
> wierd disconnection in nodeb and epc

Is the UE disconnecting immediately? Do you have the logfile for that?
Are you able to get any data across before it disconnects? You can
start a `ping` (to the epc) running on the UE host before you fire up
`srsue`, just to load the link as soon as the connection comes up.

It doesn't sound like this is the issue, but you might consider
redeploying your RIC; sometimes the `submgr` service seems to allow
"dangling" subscriptions to be left in place when the eNodeB re-attaches
to the `e2term` service... and this results in the xapp thinking the
subscription is in place, but the eNodeB doesn't know about it. I've
mentioned this before; just reminding.

> and no matter what i perform the next steps i couldn't able to get the data

> Hariharan Venkatraman

David

Hariharan Venkatraman

unread,

Dec 21, 2024, 8:02:19 PM12/21/24

to Powder Users

1. enode b successfully showing the xml file after the nexran xapp applied over it
2. no problem with ue setup
3. the data receiving confirmation on 192.168.0.11 --> so no problem within the above steps
4. iperf client and server is up and working
5. couldn't receive data in dashboard also the reward is same throughout
6. confirmation that all the pods for ric is running
7. confirmation that nexran database exists (checked thorugh the influxDB URL)
8. when i try to print the last 10 values (txbrate and rxbrate) i am getting the same values (PROBLEM and no graphs are visible in the dashboard)
9. when tried to check the status of influxdb it shows not found

what is the reason behind the same data logging and how to resolve this

adding images for all the above points (except the iperf server on nb-1 and iperf client on ue-1 working status --> it is giving the response) and my python file.
Note: i used the correct NExran IP and influx db URL that i got through ". /local/repository/demo/get-env.sh"

Best,
Hariharan Venkatraman

WhatsApp Image 2024-12-21 at 19.06.52_f2bff234.jpg

WhatsApp Image 2024-12-21 at 19.49.43_d95717a7.jpg

WhatsApp Image 2024-12-21 at 19.50.39_1f81a6de.jpg

WhatsApp Image 2024-12-21 at 19.04.37_d7f24b54.jpg

WhatsApp Image 2024-12-21 at 19.57.32_33b3bfb8.jpg

WhatsApp Image 2024-12-21 at 19.05.33_8fe5c799.jpg

WhatsApp Image 2024-12-21 at 19.04.13_155aeee0.jpg

WhatsApp Image 2024-12-21 at 19.09.04_b674bfbf.jpg

Hariharan Venkatraman

unread,

Dec 21, 2024, 8:09:23 PM12/21/24

to Powder Users

forget to add my python script.

Best,
Hariharan Venkatraman

q_learning_avg_upprb_dlprb.py

David M Johnson

unread,

Dec 21, 2024, 8:45:54 PM12/21/24

to powder...@googlegroups.com

On 12/21/24 18:02, Hariharan Venkatraman wrote:
> 1. enode b successfully showing the xml file after the nexran xapp
> applied over it
> 2. no problem with ue setup
> 3. the data receiving confirmation on 192.168.0.11 --> so no problem
> within the above steps
> 4. iperf client and server is up and working
> 5. couldn't receive data in dashboard also the reward is same throughout
> 6. confirmation that all the pods for ric is running
> 7. confirmation that nexran database exists (checked thorugh the
> influxDB URL)
> 8. when i try to print the last 10 values (txbrate and rxbrate) i am

> getting the same values (*PROBLEM and no graphs are visible in the
> dashboard*)

Yes, if you don't see data in the dashboard, your script is unlikely to
find any in influxdb, either.

> 9. when tried to check the status of influxdb it shows not found

influxdb is running in kubernetes, not in the host root context:

kubectl -n ricxapp get pods ricxapp-influxdb-0

> what is the reason behind the same data logging and how to resolve this
>
> adding images for all the above points (except the iperf server on nb-1
> and iperf client on ue-1 working status --> it is giving the response)
> and my python file.
> Note: i used the correct NExran IP and influx db URL that i got through
> ". /local/repository/demo/get-env.sh"

I verified that your NexRAN xapp instance has the proper influxdb config:

. /local/repository/demo/get-env.sh
curl -L -X GET http://$NEXRAN_XAPP:8000/v1/appconfig ; echo

Then I went to check your nexran xapp log, to make sure it was receiving
data from the eNodeB via the RIC (it dumps logs when KPM indications
arrive) but it's not outputting anything, since your eNodeB is not running.

I would advise you to re-run all the steps we ran on the call. Tear
everything down: run the nexran demo cleanup script (the -shvlan.sh
variant), undeploy nexran via dms_cli, restart the RIC components, wait
until the e2term service is finished restarting, redeploy nexran with
dms_cli, configure nexran influxdb IP, restart the RAN and UE processes,
and re-run the -shvlan.sh variant of the demo script. The problem could
be a stale RIC subscription, as we've discussed.

> Hariharan Venkatraman

David

> --
> 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 visit

Message has been deleted

Hariharan Venkatraman

unread,

Dec 27, 2024, 5:49:57 PM12/27/24

to Powder Users

i am getting "ssh: connect to host pc497.emulab.net port 22: Connection timed out" while trying to log in to the node0
If i reboot both the exp will that make any changes (that we made during the call)? if so whats the other way to resolve this?

Also i have already re-run the entire step that we discussed in call and there is no change in the error.

Best,
Hariharan Venkatraman

Hariharan Venkatraman

unread,

Dec 27, 2024, 5:50:17 PM12/27/24

to Powder Users

ssh: connect to host pc497.emulab.net port 22: Connection timed out

this is what it is showing when try to access the node-o and if i reboot the node, does that make any changes that we did during the call? If so how to resolve this?

Also i did the whole re-run of what we did in the call, still the same error. I have made a post on this 2 days back, but it was deleted, I don't know why.

Best,
Hariharan Venkatraman

Hariharan Venkatraman

unread,

Dec 28, 2024, 3:24:57 PM12/28/24

to Powder Users

I reloaded the node and now everything up and working (receiving data). But couldn't bind the slice with the enb.
what i did:
1. created another copy of run-nexran-slicing.sh (since the one you created was deleted in reloading) and replaced the id: 1 in line 29.

I remember we did one more change and i am unable to find that. So is the problem. Could you help me with that.

Best,
Hariharan Venkatraman

David M Johnson

unread,

Dec 29, 2024, 1:40:01 PM12/29/24

to powder...@googlegroups.com

Hi Hariharan,

I've attached a diff that shows the changes I believe we made, as well
as full versions of the scripts we changed. I believe that the only
change we made was to use a different enb_id in both the demo script and
the cleanup script.

Thanks,

David

On 12/28/24 13:24, Hariharan Venkatraman wrote:
> I reloaded the node and now everything up and working (receiving data).
> But couldn't bind the slice with the enb.
> what i did:
> 1. created another copy of run-nexran-slicing.sh (since the one you
> created was deleted in reloading) and replaced the id: 1 in line 29.
>
> I remember we did one more change and i am unable to find that. So is
> the problem. Could you help me with that.
>
> Best,
> Hariharan Venkatraman
>
>
> On Friday, December 27, 2024 at 5:50:17 PM UTC-5 Hariharan Venkatraman
> wrote:
>

> ssh: connect to host pc497.emulab.net <http://pc497.emulab.net> port

nexran-profile.diff

run-nexran-slicing-shvlan.sh

cleanup-nexran-shvlan.sh

Hariharan Venkatraman

unread,

Jan 9, 2025, 7:01:55 PMJan 9

to Powder Users

Did that. Now everything is working except the influxdb. i can see the kpm in enodeb as well the nexran xapp is correctly deployed along with correct slices (Fast and slow) and i used your run-nexran-slicing-shvlan.sh for demo. But getting the same data from influxdb and i couldn't able to see any graphs in the grafana as well. I am using the same influxdb url which i got through (. /local/repository/demo/get-env.sh)

there is caution msg in my ue node (added image).
Could you give me a solution for this.

Best,
Hariharan Venkatraman

Screenshot 2025-01-09 185809.png

David M Johnson

unread,

Jan 10, 2025, 10:50:11 AMJan 10

to powder...@googlegroups.com

On 1/9/25 17:01, Hariharan Venkatraman wrote:
> Did that. Now everything is working except the influxdb. i can see the
> kpm in enodeb as well the nexran xapp is correctly deployed along with
> correct slices (Fast and slow) and i used your
> run-nexran-slicing-shvlan.sh for demo. But getting the same data from
> influxdb and i couldn't able to see any graphs in the grafana as well. I
> am using the same influxdb url which i got through (.
> /local/repository/demo/get-env.sh)

Did you check the nexran xapp logs to verify that KPM metrics are
arriving? Did you check the nexran xapp config to ensure the influxdb
URL is correct? (you just do a GET to the /config endpoint instead of a
PUT) If KPM metrics are not arriving, most likely there is a stale
subscription in the RIC, and you will need to tear everything down and
restart the RIC deployments in Kubernetes.

> there is caution msg in my ue node (added image).
> Could you give me a solution for this.

I think that this warning message implies that the ratio for the slice
this UE is in is too low.

> Hariharan Venkatraman

David

> <http://pc497.emulab.net <http://pc497.emulab.net>> port

Hariharan Venkatraman

unread,

Jan 15, 2025, 9:39:04 PMJan 15

to Powder Users

1. I couldn't get the slice-related graphs in my Grafana dashboard although everything is correctly setup (added image )
2. I noticed an error in the ue node, at the end of the experiment (might have occurred earlier), before terminating.

Could you clarify these doubts?
Best,
Hariharan Venkatraman

WhatsApp Image 2025-01-15 at 21.34.19_b183f136.jpg

WhatsApp Image 2025-01-15 at 20.01.35_93d4c394.jpg

David M Johnson

unread,

Jan 16, 2025, 12:43:58 PMJan 16

to powder...@googlegroups.com

On 1/15/25 19:39, Hariharan Venkatraman wrote:
> 1. I couldn't get the slice-related graphs in my Grafana dashboard
> although everything is correctly setup (added image )

Hi Hariharan, did you see my previous replies inline? The image you
sent does show API calls to the xApp, but I asked you to confirm those
things by doing some quick investigations:

>> Did you check the nexran xapp logs to verify that KPM metrics are
>> arriving? Did you check the nexran xapp config to ensure the
influxdb
>> URL is correct? (you just do a GET to the /config endpoint
instead of a
>> PUT) If KPM metrics are not arriving, most likely there is a stale
>> subscription in the RIC, and you will need to tear everything
down and
>> restart the RIC deployments in Kubernetes.

These are still the same things I would check one by one and verify. If
you don't see data in Grafana, you need to make sure it is arriving at
the xApp. The xApp logs are available via kubectl:

kubectl logs -f -n ricxapp -l app=ricxapp-nexran

You can check the xApp config via

. /local/repository/demo/get-env.sh
curl -i -X GET http://${NEXRAN_XAPP}:8000/v1/appconfig

It will tell you which influxdb endpoint it is trying to send metrics to.

> 2. I noticed an error in the ue node, at the end of the experiment
> (might have occurred earlier), before terminating.

As I said in my previous reply, I'm thinking that this warning message
implies that the ratio for the slice containing this UE is too low.
Maybe for the 50PRB case, make sure no slice is limited below a share of
20 or 25%? Let's say you gave your fast slice a share of 1024, and your
slow slice 128. That ends up giving your slow slice an 11% share, which
could be resulting in the UE getting this invalid RBG configuration.

>
> Could you clarify these doubts?
> Best,
> Hariharan Venkatraman

David

Hariharan Venkatraman

unread,

Feb 4, 2025, 1:15:09 PMFeb 4

to Powder Users

I tried to reserve the resource for the above experiment, but it has been canceled.
I continued researching this because I didn't get the expected maximized performance from the slice allocation.
Also facing the RIC stale sub problem again (tried all the combinations of rectifications that we discussed but couldn't able to get the data on influxdb and verified and performed multiple times still the same prob)
I even redeployed and restarted the O-ran, but I am still facing the same issue. Now, I have made some changes to my local script, which will perform the slice allocation.
So is it possible to approve those resources?

Best,
Hariharan Venkatraman

David M Johnson

unread,

Feb 4, 2025, 2:48:53 PMFeb 4

to powder...@googlegroups.com

Hi Hari, I was uncanceling your reservation, but must have done so just
as you deleted it. There was a delay in approving your reservation
yesterday, and it was approved well after it was scheduled to start, and
this caused an unused-cancelation to happen sooner than it should have.
Sorry about that.

I re-created your reservation; the resources are available for you now.

David

> > > >
> > > > --
> > > > 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 visit
> > > >
> > >
> >

Hariharan Venkatraman

unread,

Feb 4, 2025, 6:06:35 PMFeb 4

to Powder Users

correct IMSI and enb id.
getting logs correctly in node 0 terminal.
But couldn't able to see the results in grafana, when i checked the enb terminal , the attached image is what i am getting and i am unsure of what is happening.
could you clarify this?

Best,
Hariharan Venkatraman

Hariharan Venkatraman

unread,

Feb 4, 2025, 6:09:21 PMFeb 4

to Powder Users

error in enb.png

David M Johnson

unread,

Feb 4, 2025, 7:20:10 PMFeb 4

to powder...@googlegroups.com

Hi Hari,

This message from the nodeb indicates that another process (a stale
nodeb, most likely) is still bound to the same IP address and port, and
so is blocking the new nodeb from binding. You'll need to kill all your
nodeb processes and try again.

David

> > > > >
> > > > > --
> > > > > 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 visit
> > > > >
> > > >
> > >
> >

Hariharan Venkatraman

unread,

Feb 17, 2025, 12:48:11 PMFeb 17

to Powder Users

I have requested emulab d710-ota-nuc1- ota-nuc2 (from 3550hz to 3600hz (CBRS ID)) and it got approved.
when i tried to start the srslte-shvlan-oran experiment after launching the O-Ran profile, it says "Warning: You have not specified any frequency request(s) for this experiment. Any transmissions will result in a quarantined experiment and frozen account."

may i know how to rectify this?

Best,
Hariharan Venkatraman

David M Johnson

unread,

Feb 17, 2025, 6:03:05 PMFeb 17

to powder...@googlegroups.com

Hi Hari,

This profile (incorrectly) does not request spectrum in the indoor OTA
case. For today, you can go ahead with your indoor OTA experiment as
per your reservation (which I have just un-canceled). We will need to
fix this profile for the future.

Thanks for checking,

David

> > > > > >
> > > > > > --
> > > > > > 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 visit
> > > > > >
> > > > >
> > > >
> > >
> >

Hariharan Venkatraman

unread,

Feb 17, 2025, 7:05:38 PMFeb 17

to Powder Users

The resources were canceled, and I couldn't start the profile (srslte-shvlan-oran). As mentioned, I already started the O-RAN profile with shvlan.
Also, could you explain what you are referring to as " indoor OTA "?
Best,
Hariharan Venkatraman

David M Johnson

unread,

Feb 17, 2025, 8:15:31 PMFeb 17

to powder...@googlegroups.com

On 2/17/25 17:05, Hariharan Venkatraman wrote:
> The resources were canceled, and I couldn't start the profile
> (srslte-shvlan-oran). As mentioned, I already started the O-RAN profile
> with shvlan.

Hm, not sure why it reverted to the canceled state, but I have
un-canceled it again.

> Also, could you explain what you are referring to as " indoor OTA "?
> Best,
> Hariharan Venkatraman

Please read https://docs.powderwireless.net/hardware.html , especially
the "Indoor OTA lab resources" subsection. Your reservation included
ota-nuc1 and ota-nuc2 ; those are SDRs in the indoor OTA lab.

David

> > > > > > >
> > > > > > > --
> > > > > > > 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 visit
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >

> https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com> <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com>> <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com?utm_medium=email&utm_source=footer> <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/f62e21d0-640f-44b7-99b4-f5c13bd11137n%40googlegroups.com?utm_medium=email&utm_source=footer>>>.

> https://groups.google.com/d/msgid/powder-users/fab18117-4126-45d7-af85-ef7ec79d48e0n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/fab18117-4126-45d7-af85-ef7ec79d48e0n%40googlegroups.com> <https://groups.google.com/d/msgid/powder-users/fab18117-4126-45d7-af85-ef7ec79d48e0n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/fab18117-4126-45d7-af85-ef7ec79d48e0n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

> https://groups.google.com/d/msgid/powder-users/909ca263-d537-4acf-a3b9-ef23b4518d89n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/909ca263-d537-4acf-a3b9-ef23b4518d89n%40googlegroups.com?utm_medium=email&utm_source=footer>.

Hariharan Venkatraman

unread,

Feb 24, 2025, 3:51:12 PMFeb 24

to Powder Users

even after restarting the O-ran, I couldn't understand why I was getting this error while trying to setup an indoor OTA experiment.
could you clarify on this?

Best,
Hariharan Venkatraman

prob with enb while setting up indoor ota.png

David M Johnson

unread,

Feb 24, 2025, 5:49:26 PMFeb 24

to powder...@googlegroups.com

On 2/24/25 13:51, Hariharan Venkatraman wrote:
>
> even after restarting the O-ran, I couldn't understand why I was getting
> this error while trying to setup an indoor OTA experiment.
> could you clarify on this?

Hi Hari, looking at your RAN nodes experiment
(https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc)
and its parameter bindings, I see that you selected dlFreq=3505e6,
ulFreq=3515e6, and prbs=100 . This means your downlink and uplink are
nearly completely overlapping, since 100 PRBs in FDD mode requires 20MHz
each for uplink and downlink bandwidth. Given your current spectrum
reservation of 3500-3520, you will need to set prbs to 50 (10MHz
bandwidth) or lower on your nodeb host. I would start at prbs=25 to
leave a bit of guard between the uplink and downlink ranges, just to
make sure things are working.

> Hariharan Venkatraman

David

Hariharan Venkatraman

unread,

Mar 16, 2025, 5:45:30 PMMar 16

to Powder Users

there was some problem with the influxDB of the O-RAN profile. No matter what, i wasn't able to access the grafana pag,e and there was also the stale state subscription error. For which i tried to redploy the O-RAN and uninstall the nexran and when i tried to check the pods the influxDB is showing as pending and it stays as it is without changing the status. Could you let me know how to fix this, I have tried using --bidir for srslte-shvlan-oran and the data seemed to be pretty noisy. I want to try in this experiment and indoor OTA as well.

Best,
Hariharan Venkatraman

David M Johnson

unread,

Mar 17, 2025, 11:14:19 AMMar 17

to powder...@googlegroups.com

I was unable to finish tracking down the influxdb/grafana issue before
your experiment terminated. I saw something new though, a problem
creating the persistentvolume to back the grafana and influxdb
deployments. I ran a test experiment of my own this morning and did not
see that problem, so you should be fine on a new experiment. If this
happens again, let me know, and I'll see if I can collect more data. It
was not at all obvious to me why the persistentvolumeclaim failed to
resolve into a persistentvolume.

David

On 3/16/25 15:45, Hariharan Venkatraman wrote:
> there was some problem with the influxDB of the O-RAN profile. No matter
> what, i wasn't able to access the grafana pag,e and there was also the
> stale state subscription error. For which i tried to redploy the O-RAN
> and uninstall the nexran and when i tried to check the pods the influxDB
> is showing as pending and it stays as it is without changing the status.
> Could you let me know how to fix this, I have tried using --bidir for
> srslte-shvlan-oran and the data seemed to be pretty noisy. I want to try
> in this experiment and indoor OTA as well.
>
> Best,
> Hariharan Venkatraman
>
> On Monday, February 24, 2025 at 5:49:26 PM UTC-5 john...@flux.utah.edu
> wrote:
>
> On 2/24/25 13:51, Hariharan Venkatraman wrote:
> >
> > even after restarting the O-ran, I couldn't understand why I was
> getting
> > this error while trying to setup an indoor OTA experiment.
> > could you clarify on this?
>
> Hi Hari, looking at your RAN nodes experiment

> (https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc <https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc>)

> and its parameter bindings, I see that you selected dlFreq=3505e6,
> ulFreq=3515e6, and prbs=100 . This means your downlink and uplink are
> nearly completely overlapping, since 100 PRBs in FDD mode requires
> 20MHz
> each for uplink and downlink bandwidth. Given your current spectrum
> reservation of 3500-3520, you will need to set prbs to 50 (10MHz
> bandwidth) or lower on your nodeb host. I would start at prbs=25 to
> leave a bit of guard between the uplink and downlink ranges, just to
> make sure things are working.
>
> > Hariharan Venkatraman
>
> David
>

Hariharan Venkatraman

unread,

Mar 24, 2025, 9:55:58 PMMar 24

to Powder Users

there is some weird error in the O-RAN profile as well the srslte-shvlan profile (kubectl: command not found). I tried using which command still nothing shows up. What might be problem and how to rectify it? I even tried by rebooting the node and still the same issue.

Best,
Hariharan Venkatraman

David M Johnson

unread,

Mar 25, 2025, 1:28:27 AMMar 25

to powder...@googlegroups.com

Hey Hari, sorry, but this looks like a platform core issue that is
affecting experiments that use the platform APIs during setup. We'll
have to investigate more in the morning when more folks are around.

David

> (https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc <https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc> <https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc <https://www.powderwireless.net/status.php?uuid=6f4d9bf0-f2e7-11ef-af1a-e4434b2381fc>>)

> > and its parameter bindings, I see that you selected dlFreq=3505e6,
> > ulFreq=3515e6, and prbs=100 . This means your downlink and uplink
> are
> > nearly completely overlapping, since 100 PRBs in FDD mode requires
> > 20MHz
> > each for uplink and downlink bandwidth. Given your current spectrum
> > reservation of 3500-3520, you will need to set prbs to 50 (10MHz
> > bandwidth) or lower on your nodeb host. I would start at prbs=25 to
> > leave a bit of guard between the uplink and downlink ranges, just to
> > make sure things are working.
> >
> > > Hariharan Venkatraman
> >
> > David
> >
> > --
> > 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 visit
> >

> https://groups.google.com/d/msgid/powder-users/3de1b127-f775-40c9-ab1f-a7e215e75c82n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/3de1b127-f775-40c9-ab1f-a7e215e75c82n%40googlegroups.com> <https://groups.google.com/d/msgid/powder-users/3de1b127-f775-40c9-ab1f-a7e215e75c82n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/powder-users/3de1b127-f775-40c9-ab1f-a7e215e75c82n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

> https://groups.google.com/d/msgid/powder-users/d9f45b1e-2139-4b52-87ff-143d3b13d481n%40googlegroups.com <https://groups.google.com/d/msgid/powder-users/d9f45b1e-2139-4b52-87ff-143d3b13d481n%40googlegroups.com?utm_medium=email&utm_source=footer>.

David M Johnson

unread,

Mar 25, 2025, 9:45:03 AMMar 25

to powder...@googlegroups.com

This has been fixed. Terminate and recreate your experiment and you
should be good to go.

David

Hariharan Venkatraman

unread,

Apr 7, 2025, 11:22:57 AMApr 7

to Powder Users

I have made a reservation to perform the indoor OTA experiment. Could you approve it?

Best,
Hariharan Venkatraman

Reply all

Reply to author

Forward

0 new messages