HCX - Transport Analytics reports a low Available Bandwidth value
search cancel

HCX - Transport Analytics reports a low Available Bandwidth value

book

Article ID: 323355

calendar_today

Updated On:

Products

VMware HCX

Issue/Introduction

To identify a known issue where the HCX Transport Analytics feature may report a lower Available Bandwidth then is actually available between connected sites

Symptoms:
Starting with HCX Manager software 4.8.0, the Transport Analytics feature may report a lower Available Bandwidth then is actually available between connected sites

HCX Manager UI >> Transport Analytics >> Uplinks Overview > RUN TEST The HCX Manager app.log found in /common/logs/admin will show the HCX appliance selected for the Transport Analytics on-demand test and the result of that test. The test result in app.log will match that shown in the HCX Manager Transport Analytics UI
 
In the example below the HCX Appliance selected for the on-demand test is NE-I6 and the Available Bandwidth reported by the test is "downloadKbps":130181,"uploadKbps":102441
 
2024-02-26 21:21:57.774 UTC [InterconnectService_SvcThread-51105, IX:########-####-####-####-##########43, J:b85a1b25, , TxId: ########-####-####-####-##########b7] INFO  c.v.h.i.u.InterconnectApplianceUtil- getIP : applianceName SM_localxxxxx-14bc6-NE-I6 networkId network-########-####-####-####-##########9f IP 172.xx.xx.11

2024-02-26 21:21:59.269 UTC [InterconnectService_SvcThread-51140, IX:########-####-####-####-##########43, J:b85a1b25, , TxId: ########-####-####-####-##########b7] INFO  c.v.h.i.u.InterconnectUnderlayUplinkUtil- OnDemand test Url : https://172.xx.xx.202:9443/uc/ondemand/start requestObject {"localAddress":"172.xx.xx.11","remoteAddress":"172.xx.xx.12"} responseBody : {"measureTime":1708982481,"status":2,"error":"","localAddress":"172.xx.xx.11","remoteAddress":"172.xx.xx2.12","downloadKbps":0,"uploadKbps":0,"lossRate":0,"latency":"","txPmtu":0,"rxPmtu":0}

2024-02-26 21:23:40.374 UTC [InterconnectService_SvcThread-51147, IX:########-####-####-####-##########43, J:b85a1b25, , TxId: ########-####-####-####-##########b7] INFO  c.v.h.i.u.InterconnectUnderlayUplinkUtil- OnDemand test result  Url : https://172.xx.xx.202:9443/uc/ondemand/results requestObject {"localAddress":"172.xx.xx.11","remoteAddress":"172.xx.xx.12"} responseBody : {"measureTime":1708982481,"status":0,"error":"","localAddress":"172.xx.xx.11","remoteAddress":"172.xx.xx.12","downloadKbps":130181,"uploadKbps":102441,"lossRate":0,"latency":"2.930\/4.898\/48.518","txPmtu":1400,"rxPmtu":1400}
 


When manually running the performance test "perftest fou" on the same HCX appliance NE-I6 however, an accurate and larger value is reported for available bandwidth when compared to the values reported by the Transport Analytics UI
 
In the below output a 468Mbps downlink and 436Mbps uplink available bandwidth are reported by "perftest fou"
  

[admin@host:SM_localxxxxx-NE-I6] perftest fou
++++++++++ StartTest ++++++++++

---------- Fou-0 [te_0, 192.168.2.63 >>> 192.168.2.72] ----------
Connecting to host 192.168.2.72, port 5201
[  5] local 192.168.2.63 port 45703 connected to 192.168.2.72 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-30.00  sec  1.64 GBytes   470 Mbits/sec  1157   1.53 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-30.00  sec  1.64 GBytes   470 Mbits/sec  1157             sender
[  5]   0.00-30.02  sec  1.64 GBytes   468 Mbits/sec                  receiver 

iperf Done.

++++++++++ StopTest ++++++++++
++++++++++ StartTest ++++++++++

---------- Fou-0 [te_0, 192.168.2.63 <<< 192.168.2.72] ----------
Connecting to host 192.168.2.63, port 5201
[  5] local 192.168.2.72 port 44645 connected to 192.168.2.63 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-30.00  sec  1.53 GBytes   438 Mbits/sec   56   1.18 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-30.00  sec  1.53 GBytes   438 Mbits/sec   56             sender
[  5]   0.00-30.08  sec  1.53 GBytes   436 Mbits/sec                  receiver

iperf Done.

++++++++++ StopTest ++++++++++ 



Cause

HCX Manager software defect

Resolution

The issue will be resolved in HCX Manager 4.9.0 software release or higher

Workaround:
On the same HCX appliance used by Transport Analytics (or any other HCX IX/NE appliance associated with the same Connected Sites) manually run the performance test "perftest fou" to determine the available bandwidth between the sites
  
Check the receiver Bitrate in both directions. In the example below it shows 468Mbps downlink >>> and 436Mbps uplink <<< 
  

ssh as admin into the HCX Manager
ccli
list
go <HCX_appliance_ID#>
perftest fou 

  
Example:
 

admin@host [~]$ ccli
Welcome to HCX Central CLI
[admin@host] list
|----------------------------------------------------------------------------|
| Node                      | Id | Address            | State     | Selected |
|----------------------------------------------------------------------------|
| SM_localxxxxx-14bc6-IX-I1 | 3  | 172.xx.xx.204:9443 | Connected |          |
|----------------------------------------------------------------------------|
<snip>
|----------------------------------------------------------------------------|
| SM_localxxxxx-14bc6-NE-I6 | 1  | 172.xx.xx.202:9443 | Connected |          | ------> here !!!!
|----------------------------------------------------------------------------|
| SM_localxxxxx-14bc6-NE-I7 | 2  | 172.xx.xx.206:9443 | Connected |          |
|----------------------------------------------------------------------------|
<snip>
[admin@host] go 1
Switched to node 1.
[admin@host:SM_localxxxxx-14bc6-NE-I6]
[admin@host:SM_localxxxxx-14bc6-NE-I6] perftest fou
++++++++++ StartTest ++++++++++

---------- Fou-0 [te_0, 192.168.2.63 >>> 192.168.2.72] ----------
Connecting to host 192.168.2.72, port 5201
[  5] local 192.168.2.63 port 45703 connected to 192.168.2.72 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-30.00  sec  1.64 GBytes   470 Mbits/sec  1157   1.53 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-30.00  sec  1.64 GBytes   470 Mbits/sec  1157             sender
[  5]   0.00-30.02  sec  1.64 GBytes   468 Mbits/sec                  receiver

iperf Done.

++++++++++ StopTest ++++++++++
++++++++++ StartTest ++++++++++

---------- Fou-0 [te_0, 192.168.2.63 <<< 192.168.2.72] ----------
Connecting to host 192.168.2.63, port 5201
[  5] local 192.168.2.72 port 44645 connected to 192.168.2.63 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-30.00  sec  1.53 GBytes   438 Mbits/sec   56   1.18 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-30.00  sec  1.53 GBytes   438 Mbits/sec   56             sender
[  5]   0.00-30.08  sec  1.53 GBytes   436 Mbits/sec                  receiver

iperf Done.

++++++++++ StopTest ++++++++++ 




Additional Information

Impact/Risks:
Starting with HCX Manager software 4.8.0, the HCX Transport Analytics feature may report a lower Available Bandwidth then is actually available between connected sites. This in turn will impact the migration insights reported by the Transport Analytics UI. This is a cosmetic reporting issue and there will be no impact to any actual migration or network extension workflows