Switching problems, switch needs frequent reboots/disconnection

[lmpf]

Hello.

I'm having a very weird issue. Recently I started having packet loss to some of the clients in the network.

I suspect this has something to do with the switch but I am unsure how to reach a decent conclusion.

This is the switching setup at the moment:

• HPE ProCurve 1800-24G Switch (J9028B)
• D-Link DGS-1016D

They are bridged using a CAT6 cable. When I detect the packet loss, I have 2 options which take immediate effect and resolve the issue for another 2 or 3 hours until it happens again:

1.  Disconnect the cable that bridges both switches and immediatly reconnect it again;
2.  Restart the HPE switch (it's the only one that is managed);

I suppose that the obvious thing is that one of the switches is faulty, unfortunately I don't have another switch at hand at this moment. Another thing worth mentioning is that this issue started pretty much a month ago, when we got our ISP to install a fiber optics connection with their brand new router. Because the ISP's router was losing a lot of packets, we decided to bridge it with an old Draytek that we had here, which despite still being affected by this packet loss, it loses much less packets, allowing me to connect remotely and restarting the switch, which then stabilizes everything for a couple hours.

NOTE: The packet loss is also present if I ping the network from a WAN connection

Any ideas? Thank you.

[nikunjbhatia]

---

@lmpf wrote:

Hello.

I'm having a very weird issue. Recently I started having packet loss to some of the clients in the network.

I suspect this has something to do with the switch but I am unsure how to reach a decent conclusion.

This is the switching setup at the moment:

• HPE ProCurve 1800-24G Switch (J9028B)
• D-Link DGS-1016D

They are bridged using a CAT6 cable. When I detect the packet loss, I have 2 options which take immediate effect and resolve the issue for another 2 or 3 hours until it happens again:

1.  Disconnect the cable that bridges both switches and immediatly reconnect it again;
2.  Restart the HPE switch (it's the only one that is managed);

I suppose that the obvious thing is that one of the switches is faulty, unfortunately I don't have another switch at hand at this moment. Another thing worth mentioning is that this issue started pretty much a month ago, when we got our ISP to install a fiber optics connection with their brand new router. Because the ISP's router was losing a lot of packets, we decided to bridge it with an old Draytek that we had here, which despite still being affected by this packet loss, it loses much less packets, allowing me to connect remotely and restarting the switch, which then stabilizes everything for a couple hours.

NOTE: The packet loss is also present if I ping the network from a WAN connection

Any ideas? Thank you.

---

Reply for your query:

Hi

As per your observations that after restarting for couple of hours it works fine and again issue restarted
Please check for ARP table utilization.
This can be reasons if the table is full then it will start dropping packet.
you can also check whether device reaching maximum arp capacity due to which packet loss is happening.