"no ping reply (NOP-Out) after 5 seconds" kills iSCSI volumes under load? (v11.1)

[kdragon75]

As long as the vlans are all created in the vlans database mode.

Have you added the vlan via the following?

Code:

(CN1610) #vlan database
(CN1610) (Vlan)#vlan  X

 

[Dudleydogg]

Yes I did that initially:
vlan database
vlan 2-5,10
exit
switchport Trunk seemed to be the Key, I now can tag my vkernals with a vlan and they communicate, I have separated iscsi, vmotion, Maintenance, and fault tolerance to separate vlans now.
Changing MTU back to 1500 was the big fix. I have been stable now for over a week.

 

[kdragon75]

Yes I did that initially:
vlan database
vlan 2-5,10
exit
switchport Trunk seemed to be the Key, I now can tag my vkernals with a vlan and they communicate, I have separated iscsi, vmotion, Maintenance, and fault tolerance to separate vlans now.
Changing MTU back to 1500 was the big fix. I have been stable now for over a week.

That's great to hear! I must say I have been happy with the switch too. I wish it was a bit more energy efficient but for $100usd shipped, I can't complain.

 

[Dudleydogg]

That's great to hear! I must say I have been happy with the switch too. I wish it was a bit more energy efficient but for $100usd shipped, I can't complain.

The problem with the ping timeouts is what prompted me to get the switch, initially I had a 4 port 10G in Freenas and created a bridge with a post init script. it worked perfectly other than the timeouts, which now I know were the MTU issue. So if my switch fails OR I could make a redundant Network connection with the 2 spare ports now.