× NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Orbi WiFi 7 RBE973
Reply

Re-post: Bug on cable test? "Short Circuit"

tempaccount
Tutor

Re-post: Bug on cable test? "Short Circuit"

This is a re-post of https://community.netgear.com/t5/Smart-Plus-Click-Switches/Bug-on-cable-test-quot-Short-Circuit-quot... which got automatically moved to spam.

 

Hi,

i'm currently using two GS108Ev3 with the latest available firmware 2.00.12 and the following setup:

Router <-> Ethernet Jack in Wall Plate <-> 10 m Cat 7 installation cable <-> Patch panel <-> 0.9 m Cat 6 Ethernet Cable <-> GS108Ev3 No.1 <-> 0.9 m Cat 6 Ethernet Cable <-> Patch panel <-> 8 m Cat 7 installation cable <-> Ethernet Jack in Wall Plate <-> 8 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

- The cable test on the GS108Ev3 No.1 is not working at all for the link between that switch and the GS108Ev3 No.2. Instead of a result im getting an half-broken HTML source code back like shown at the attached screenshot
- The cable test on the GS108Ev3 No.2 is showing a "Short circuit" at 23 m distance
- The link speed between both switches at Management -> Port status is showing with 100M instead of the expected 1000M

If im replacing both GS108Ev3 with a network cable tester everything is green, not short circuits found.
Furthermore if im replacing the 8 m Cat 6 Ethernet Cable between the Ethernet Jack and the GS108Ev3 No. 2 with a shorter cable (1 m) the described three behaviors above are gone as well. To make sure that the single 8m cable is not the problem ive replaced this with 3 other working cables as well and the issue still happens.

Especially as the first GS108Ev3 No.1 is showing such a broken response for the cable test this looks to me like some sort of bug triggered in the firmware and the shown short circuit might be just an effect of this.


Is anyone aware of such issues with such a setup? Is there anything i could test or work around so i got a stable connection between both switches and a full 1000M speed as expected? Anything else i have missed / done wrong in this setup?

Thanks for your replies.

*Edit*

Note sure but could be related to https://community.netgear.com/t5/Smart-Plus-Click-Switches/Short-Circuit-on-Cable-Test/td-p/1304659

Model: GS108Ev3|ProSAFE 8-port Gigabit Plus Switch, GS108Ev3|ProSAFE 8-port Gigabit Plus Switch
Message 1 of 3

Accepted Solutions
tempaccount
Tutor

Re: Re-post: Bug on cable test? "Short Circuit"

Just to clarify:

 

There is currently no support needed from the community. I'm currently not able to fullfill the requirements of the Netgear support to escalate this to a higher level.

 

This post here is just there to avoid that the collected information are lost if any other user is ever stumbling over the same issue.

View solution in original post

Message 3 of 3

All Replies
tempaccount
Tutor

Re: Re-post: Bug on cable test? "Short Circuit"

In the meantime i was in contact with the Netgear support and had collected additional info:

--------------

1. Perform factory reset of both switches.
-> A factory reset for both switches was done

2. Remove the switches from the entire network.
-> This was done together with step 3. below

3. Connect first switch to the second switch locally using the8 m Cat 6 Ethernet Cable and perform the cable test to check if we got the same results.

Both switches are now connected locally to have the following setup:

GS108Ev3 No.1 <-> 8 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

Running the cable tests now on both switches are giving back an OK and the port speed is 1000M again.

4. Please check if flow control is enabled on uplink ports.
-> flow control is disabled for all ports (this is the default after a factory reset)

--------------

 

and

 

--------------

The 8 M Cat 6 Ethernet Cable was switched with a 15 M Cat 6 Ethernet Cable which is showing the same behavior.

 

I have downgraded the firmware on GS108Ev3 No. 2 to 2.00.09 and retested the connection, the behavior stays the same. As an experiment i've also downgraded the firmware on GS108Ev3 No. 1 in a second step an the behavior is staying the same as well.


So to sum-up i did the following setups which shows different behaviors and could show a possible issue with overlong cable length between the two uplinks:

Setup 1
---------------------
Router <-> Ethernet Jack in Wall Plate <-> 10 m Cat 7 installation cable <-> Patch panel <-> 0.9 m Cat 6 Ethernet Cable <-> GS108Ev3 No.1 <-> 0.9 m Cat 6 Ethernet Cable <-> Patch panel <-> 8 m Cat 7 installation cable <-> Ethernet Jack in Wall Plate <-> 8 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

Behavior: Short circuit in cable test at GS108Ev3 No.1 and broken response on cable tester at GS108Ev3 No.2, 100M link speed on uplink
---------------------

Setup 2
---------------------
Router <-> Ethernet Jack in Wall Plate <-> 10 m Cat 7 installation cable <-> Patch panel <-> 0.9 m Cat 6 Ethernet Cable <-> GS108Ev3 No.1 <-> 0.9 m Cat 6 Ethernet Cable <-> Patch panel <-> 8 m Cat 7 installation cable <-> Ethernet Jack in Wall Plate <-> 1 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

Behavior: No short circuit, cable test in both GS108Ev3 switches are OK. 1000M link speed on uplink
---------------------

Setup 3
---------------------
GS108Ev3 No.1 <-> 8 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

Behavior: No short circuit, cable test in both GS108Ev3 switches are OK. 1000M link speed on uplink
---------------------

Setup 4
---------------------
GS108Ev3 No.1 <-> 15 m Cat 6 Ethernet Cable <-> GS108Ev3 No.2 <-> 1 m Cat 6 Ethernet Cable <-> Laptop

Behavior: No short circuit, cable test in both GS108Ev3 switches are OK. 1000M link speed on uplink
---------------------

--------------

 

and

 

-------------

I think with the current info i have gathered we currently need to distinguish between two different issues:

1. Device is answering with an invalid HTTP response (see attached pcap) where some parts of the headers and body are stripped.

2. The devices might have some issues if they are connected together with overlong cable length >= 25m. In this case i found something similar as well here:

 

https://community.netgear.com/t5/Smart-Plus-Click-Switches/Short-Circuit-on-Cable-Test/m-p/1304659

 

The user is also reporting such a short circuit. For some reason i'm not allowed to reply there with my account but it might be interesting to ask him for the length of the cable between his devices.

Message 2 of 3
tempaccount
Tutor

Re: Re-post: Bug on cable test? "Short Circuit"

Just to clarify:

 

There is currently no support needed from the community. I'm currently not able to fullfill the requirements of the Netgear support to escalate this to a higher level.

 

This post here is just there to avoid that the collected information are lost if any other user is ever stumbling over the same issue.

Message 3 of 3
Top Contributors
Discussion stats
  • 2 replies
  • 3472 views
  • 0 kudos
  • 1 in conversation
Announcements