At first apologize for my english.
We have purchased the new Equllogic PS-M4110 (Firmware Revision 9.0.2).
We also have 3 Blade: 2xPowerEdge M620, 1xPowerEdge M630 with OracleLinux 6.8 installed.
When a new matrix is added to an existing group, the connections to host volumens are broken.
Configuring MXL 10 / 40GbE switch ports where Equallogic's are located:
Ports 0 / 13,0 / 14,0 / 15,0 / 16,1 / 13,1 / 14,1 / 15,1 / 16 the same configuration
TenGigabitEthernet interface y / xx
No ip address
Mtu 12000
Switchport
Flowcontrol rx on tx off
Spanning-tree mstp edge-port
Spanning-tree rstp edge-port
Spanning-tree 0 portfast
Spanning-tree pvst edge-port
Remote-fault-signaling rx off
No shutdown
!
Configuring MXL 10 / 40GbE ports where Blade is located
Ports 0 / 1,0 / 2,0 / 3,1 / 1,1 / 2,1 / 3 the same configuration
TenGigabitEthernet interface y / x
No ip address
Mtu 12000
Portmode hybrid
Switchport
Flowcontrol rx on tx off
Spanning-tree rstp edge-port
No shutdown
Other switch parameters
No dcb enable
!
Interface Vlan 100
Description "SAN"
No ip address
Mtu 12000
Tagged TenGigabitEthernet 0 / 1-3
Tagged TenGigabitEthernet 1 / 1-3
Untagged TenGigabitEthernet 0 / 13-16
Untagged TenGigabitEthernet 1 / 13-16
No shutdown
!
Switch firmware nersion 9.11
Host Configuration NICs:
# 10GbE TUNING
Net.ipv4.tcp_sack = 0
Net.ipv4.tcp_timestamps = 0
Net.ipv4.tcp_rmem = 8192 87380 16777216
Net.ipv4.tcp_wmem = 4096 65536 16777216
Net.ipv4.tcp_mem = 10000000 10000000 10000000
Net.core.rmem_max = 16777216
Net.core.wmem_max = 16777216
Net.core.rmem_default = 262144
Net.core.wmem_default = 262144
Net.core.optmem_max = 524287
Net.core.netdev_max_backlog = 300000
Net / ipv4 / conf / bond1 / arp_ignore = 1
Net / ipv4 / conf / bond1 / arp_announce = 2
Net / ipv4 / conf / bond1 / rp_filter = 2
Net / ipv4 / conf / bond1.100 / arp_ignore = 1
Net / ipv4 / conf / bond1.100 / arp_announce = 2
Net / ipv4 / conf / bond1.100 / rp_filter = 2
You may have ideas what is wrong and the session breaks with volumes on the "old" matrix