Problems acccessing SG300-28P via management interface

[adam-goldberg]

I have a new SG300-28P, and have had occasional issues with being unable to connect to it via anything other than the serial port.  I have connectivity between my machine and the switch (tested with ping each way), and in fact, have the same problem if I take a laptop to the switch and connect them directly.

What happens is that though the switch is operating normally, http, https, ssh and telnet attempts to access all fail in one way or another.  Ssh and telnet either yields no response or a refused connection (even though those services are enabled).  For http and https, I'll occasionally get enough of the web page to be able to tell what it is ... but attempts to log in just don't work.

While this is happening, the CPU and packet load on the switch is very, very low.

Rebooting didn't help entirely, though it may have made it better.  Resetting to factory defaults and then reconfiguring makes it work.

This is using the latest firmware: 1.2.7.76.

Searching the web for this sort of failure doesn't yield any results -- maybe I'm the only one to see this?

I don't know what else I can do to diagnose ..... I've got it working without trouble now...

[bhackbarth]

Hi Adam,

  Could you do another factory reset, then do a ping -t

and let it run for 10 minutes or more and see if you see any dropped packets?

[adam-goldberg]

Ok... I got no lost packets.  Most of the responses were 1ms or <1ms (periodically, 2-5ms).  Occasionally, they were quite longer.  The maximum was 89ms.

But again, it's not exhibiting the problem now.

[Clayton Sill]

Hello Adam,

After doing a factory reset how long does it take untill you start experiencing this issue? I've never heard of a issue that you are experiencing. Has it always done this since you have gotten the switch? Wouldn't hurt to open a case with Cisco Small Business support just to see if they have seen a similar issue like this before. I would try reflashing the firmware, maybe something got corrupted in the firmware.

I hope that helps,

Thanks,

Clayton Sill

[pavel_kotelva]

I have the same problem: several dozens of SF300-24 with the equal  problem. Just after reboot the switch responds to ping, ssh and telnet. After some time (from 15 min to 7 hours) ping dissapear, telnet/ssh too. Connecting through console to the switch in this state can see, that there is no ARP and MAC addresses of the gate or servers on the management vlan. While rrying to ping the gate get the next message:

PING: unable-to-send

But  if I reload the switch by power cycle or 'reload' command, the ping and  access to gate and servers returns. All switches have 1.3.0.62 firmware. All switches are "from the box", purcheased few weeks ago. The 'factory reset' didn't helps at all.

[adam-goldberg]

Sorry, I solved my problem and moved on without noting it here.

The problem was entirely my fault, I'm afraid.  It was solved by carefully checking DHCP-supplied DNS and default gateway settings (oops!).

[browan]

I have this problem too.  It seems to have started from either when I upgraded to the latest firmware and/or changed the management interface from the default (vlan 1) to vlan 11.  It will stay up and pinging for anywhere from a few minutes to 3 hours, then I lose all connectivity until I reboot the device.

switch5782a5#show inventory

NAME: "1"   DESCR: "SG300-10P 10-Port Gigabit PoE Managed Switch"  

PID: SRW2008P-K9   VID: V01   SN: PSJ1522063N  

switch5782a5#sh ver

SW version    1.3.5.58 ( date  10-Oct-2013 time  17:15:41 )

Boot version    1.3.5.06 ( date  21-Jul-2013 time  15:12:10 )

HW version    V01