NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
M4300 switch: port is down after reboot
Hello community.
I'm facing the following issue: I have 3x M4300 switches in a circle. After rebooting (reload) one of the switches, some of those ports connecting to the other 2x switches are down. If I "shutdown" and "no shutdown" such a port, everything is up and running. For example, port 1/0/11:
(SWITCH) #show interfaces status all
Link Physical Physical Media Flow Control
Port Name State Mode Status Type Status
--------- ---------------------------- ------ ---------- ---------- ------------------ ------------
1/0/7 Up Auto 1000 Full 1000SX Inactive
1/0/8 Down 10G Full Inactive
1/0/9 Down Auto Inactive
1/0/10 Down Auto Inactive
1/0/11 SWITCH2 Down Auto Inactive
1/0/12 Down Auto Inactive
1/0/13 Down Auto Inactive
(SWITCH) #configure
(SWITCH) (Config)#interface 1/0/11
(SWITCH) (Interface 1/0/11)#shutdown
(SWITCH) (Interface 1/0/11)#no shutdown
(SWITCH) (Interface 1/0/11)#end
(SWITCH) #show interfaces status all
Link Physical Physical Media Flow Control
Port Name State Mode Status Type Status
--------- ---------------------------- ------ ---------- ---------- ------------------ ------------
1/0/7 Up Auto 1000 Full 1000SX Inactive
1/0/8 Down 10G Full Inactive
1/0/9 Down Auto Inactive
1/0/10 Down Auto Inactive
1/0/11 SWITCH2 Up Auto 1000 Full 1000SX Inactive
1/0/12 Down Auto Inactive
1/0/13 Down Auto Inactive
How can I troubleshoot this further?
M4300-24X24F ProSAFE 24-port 10GBASE-T and 24-port 10G SFP+
M4300-12X12F ProSAFE 12-port 10GBASE-T and 12-port 10G SFP+
Software: 12.0.17.13
Bootcode: B1.0.0.17
This is the only stuff I've seen in the logs:
(SWITCH) #show logging buffered
Buffered (In-Memory) Logging : enabled
Buffered Logging Wrapping Behavior : On
Buffered Log Count : 81
Buffered Log Threshold (lines) : 160
<13> Jul 11 08:43:55 SWITCH-1 TRAPMGR[dot1s_task]: traputil.c(795) 538 %% Spanning Tree Topology Change Received: MSTID: 0 1/0/11
<13> Jul 11 08:43:53 SWITCH-1 TRAPMGR[dot1s_task]: traputil.c(795) 537 %% Spanning Tree Topology Change Received: MSTID: 0 1/0/11
<13> Jul 11 08:43:53 SWITCH-1 TRAPMGR[dot1s_task]: traputil.c(795) 536 %% Spanning Tree Topology Change Initiated: 0, Interface: 1/0/11
<13> Jul 11 08:43:53 SWITCH-1 TRAPMGR[dot1s_task]: traputil.c(795) 535 %% Spanning Tree Topology Change: 0, Unit: 1
<13> Jul 11 08:43:53 SWITCH-1 TRAPMGR[trapTask]: traputil.c(753) 534 %% Link Up: 1/0/11
Any hints for me?
Thanks a lot.
Johannes
Hallo Johannes,
A little bit thin insights to provide an answer and discuss in the community.
In case there are 10G/1G SFP+ modules plugged, it's extremely likely these try to establish a 10G link - what might be not reliable when connecting to a DWDM port operating only on 1000M, since there is (different from copper Ethernet) no link negotiation. Suspect more a "random" effect of the "down" and "up".
Strongly recommend to use the appropriate SFP+ module (a pure SFP 1000BASE-FX for this fiber link) , and not to depend on some "auto"magic.
I'd suggest to get in contact with the Netgear ProAVDesign team (with LaurentMa at the helm), refer to this thread please, and chime back on the findings. This team is much nearer to the engineering than the community moderators or other "old crows" like me here.
Grüsse aus der Schweiz,
-Kurt.
8 Replies
Hello webernetz
And welcome to the NETGEAR Community! 🙂Looks like your firmware is not updated. You may want to try updating the firmware to the latest and see if there will be any difference. If the issue is the same after the update of firmware you could try opening a ticket with our support team.
Have a lovely day,
Erwin
Netgear TeamHey Erwin. Thanks for your answer. In the meantime I upgraded all switches to the most current version (12.0.19.7), but the issue remains.
We suspect the DWDM system between the data centres, or at least some kind of interaction between the SFPs. Do you have any known issues with such systems?
Hallo Johannes,
A little bit thin insights to provide an answer and discuss in the community.
In case there are 10G/1G SFP+ modules plugged, it's extremely likely these try to establish a 10G link - what might be not reliable when connecting to a DWDM port operating only on 1000M, since there is (different from copper Ethernet) no link negotiation. Suspect more a "random" effect of the "down" and "up".
Strongly recommend to use the appropriate SFP+ module (a pure SFP 1000BASE-FX for this fiber link) , and not to depend on some "auto"magic.
I'd suggest to get in contact with the Netgear ProAVDesign team (with LaurentMa at the helm), refer to this thread please, and chime back on the findings. This team is much nearer to the engineering than the community moderators or other "old crows" like me here.
Grüsse aus der Schweiz,
-Kurt.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
