Slow SMB Transfer Speeds Around 5MB/s

[Simon Sparks]

I am tempted to install FN 11 in a new boot environment and see if the problem persists.

 

[SweetAndLow]

I am tempted to install FN 11 in a new boot environment and see if the problem persists.

lol the problem is with how you configured things not the version of freenas you have. You should try removing variables now, direct connect your client to the server. This will remove the switch from the scenario. Then run your iperf tests.

 

[Pheran]

Instead of just the running config, post the "show interface" output for the ports your server is connected to. Getting 25 Mbps from iPerf is a gigantic, glaring network problem.

 

[Simon Sparks]

There has been NO network changes the ONLY change was the upgrade to Build FreeNAS-9.10.2-U5 (561f0d7a1) prior to the upgrade the SMB transfer rate was always consistently 110MB/s

 

[Simon Sparks]

GigabitEthernet1/0/9 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet, address is 9caf.ca76.4189 (bia 9caf.ca76.4189)
Description: FreeNAS - Primary Management
MTU 9198 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseTX
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:10, output 00:00:00, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 97
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 6000 bits/sec, 3 packets/sec
5 minute output rate 35000 bits/sec, 22 packets/sec
9089522392 packets input, 707314654328 bytes, 0 no buffer
Received 299527 broadcasts (282590 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 282590 multicast, 0 pause input
0 input packets with dribble condition detected
1066728122 packets output, 1507046236936 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out

GigabitEthernet1/0/21 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet, address is 9caf.ca76.4195 (bia 9caf.ca76.4195)
Description: FreeNAS - Primary Management
MTU 9198 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseTX
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 02:14:59, output 00:00:00, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 25000 bits/sec, 17 packets/sec
4241501649 packets input, 366763261970 bytes, 0 no buffer
Received 284587 broadcasts (282262 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 282262 multicast, 0 pause input
0 input packets with dribble condition detected
12867581440 packets output, 19312300940807 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out

GigabitEthernet2/0/9 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet, address is 0026.cbc4.2f89 (bia 0026.cbc4.2f89)
Description: FreeNAS - Primary Management
MTU 9198 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseTX
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 02:15:20, output 00:00:09, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 25000 bits/sec, 15 packets/sec
617574703 packets input, 64319683971 bytes, 0 no buffer
Received 283445 broadcasts (281635 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 281635 multicast, 0 pause input
0 input packets with dribble condition detected
72524224 packets output, 13831413681 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out

GigabitEthernet2/0/21 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet, address is 0026.cbc4.2f95 (bia 0026.cbc4.2f95)
Description: FreeNAS - Primary Management
MTU 9198 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseTX
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 02:15:33, output 00:00:02, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 25000 bits/sec, 16 packets/sec
24593787 packets input, 15254569150 bytes, 0 no buffer
Received 307535 broadcasts (281766 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 281766 multicast, 0 pause input
0 input packets with dribble condition detected
139071803 packets output, 18065276258 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out








GigabitEthernet3/0/38 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet, address is 58bc.2789.afa6 (bia 58bc.2789.afa6)
Description: Office PC
MTU 9198 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseTX
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:08:18, output 00:00:09, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 97
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 25000 bits/sec, 40 packets/sec
5 minute output rate 107000 bits/sec, 59 packets/sec
4724084 packets input, 2814968851 bytes, 0 no buffer
Received 467546 broadcasts (339935 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 339935 multicast, 0 pause input
0 input packets with dribble condition detected
116044181 packets output, 21839030965 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out

 

[Pheran]

Well that wasn't too helpful, no obvious issues there. If you believe that the problem was caused by an upgrade, then why not revert to the previous FreeNAS boot environment and see if it goes away?

 

[Jacopx]

I had the similar problem from a Apple Mac and I have solved with this:

Code:

printf "[default]\nsigning_required=no\n" | sudo tee /etc/nsmb.conf >/dev/null

The source is this: https://dpron.com/os-x-10-11-5-slow-smb/

 

[Pheran]

I had the similar problem from a Apple Mac and I have solved with this:

Code:

printf "[default]\nsigning_required=no\n" | sudo tee /etc/nsmb.conf >/dev/null

The source is this: https://dpron.com/os-x-10-11-5-slow-smb/

That is not going to fix this problem. His baseline network performance is horrible.

 

[Jacopx]

That is not going to fix this problem. His baseline network performance is horrible.

Uhmmmm I admit that I haven't read with too much attention the problem... Sorry! :/

 

[Simon Sparks]

I believe the issue maybe caused by the auto tune feature.

 

[Simon Sparks]

I am doing a fresh install of 9.10.2 Update 4

 

[SweetAndLow]

I am doing a fresh install of 9.10.2 Update 4

You are not educated enough to touch the auto tune stuff. Please don't bother with it until you know how each of them affect the system. You should also turn off jumbo frames, they have veyr little impact on a 1gbps network.

 

[Simon Sparks]

@SweetAndLow Do you have any actual useful comments.

If you had read all of the posts you would see that jumbo frames is NOT enabled on the 1GbE links only on the 10GbE iSCSI VLAN interfaces.

 

[Pheran]

I am doing a fresh install of 9.10.2 Update 4

OK, I'll bite. Why are you reinstalling instead of booting the previously-working environment?

 

[Simon Sparks]

For some reason the upgrade from U4 to U5 did NOT create a new boot environment not sure why as it has done previously and it did on my backup freenas box.

 

[Stux]

FYI your iperf results show your network is running at less than 3% efficiency.

That's about the same speed as trying to communicate across an air gap.

Have you confirmed you don't have a broken Ethernet cable?

PS: you can communicate across an air gap via cross talk.

 

[anodos]

FYI your iperf results show your network is running at less than 3% efficiency.

That's about the same speed as trying to communicate across an air gap.

Have you confirmed you don't have a broken Ethernet cable?

PS: you can communicate across an air gap via cross talk.

+1 In my experience the majority of networking problems are cable-related.

 

[Pheran]

+1 In my experience the majority of networking problems are cable-related.

The problem with this theory is that if it was a cabling problem, we'd most likely see CRC or other errors in the show interface output, and there are none. Clearly there is something wrong with the network, but I don't know what at this point. If he can in fact return to a working configuration on an earlier FreeNAS version, it could be a driver issue, but I'm skeptical because I don't believe there were any driver changes between U4 and U5.

 

[Simon Sparks]

If you look at the values that the auto configuration put in place I think one of them altered the networking in such a way that it actually made the 1GbE link worse

 

[m0nkey_]

The problem with this theory is that if it was a cabling problem, we'd most likely see CRC or other errors in the show interface output, and there are none. Clearly there is something wrong with the network, but I don't know what at this point. If he can in fact return to a working configuration on an earlier FreeNAS version, it could be a driver issue, but I'm skeptical because I don't believe there were any driver changes between U4 and U5.

There were no driver updates in U5. All it did was patch Samba and a SSH CVE.