No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search


To have a better experience, please upgrade your IE browser.

Knowledge Base

The V3 storage low NFS performance issue

Publication Date:  2016-08-29  |   Views:  198  |   Downloads:  0  |   Author:  y90002984  |   Document ID:  EKB1000214146


Issue Description

The speed is low when copy the files between the LUN on the S5600 and NFS shares of 5500 V3. 

Handling Process

For the performance issue, the trouble shooting should start from Host, network and then storage array.


Check if the CPU, Memory and network usage of the host are normal


Check if the CPU, Memory and network usage of the storage are normal


Check if the IO latency and disk usage of the storage are normal


Check the system status by logs

Generally speaking, most of performance issues will fixed by these four steps. But for this case, they are all normal but the performance is still low.

So we decide to proceed below steps:

1. Exclude the S5600's suspect, copy the data from windows host's local disks but the problem still exist.

2. Exclude the file system's suspect, create a new file system and export the share but the problem still exist.

3. Exclude the network, mount the file system by using the P0 port of the controller B. The read speed from local disks is 400MB/s and the write speed is 200MB/s. So we can confirm that the problem was caused by the network.

For the network issue, try below steps to trouble shooting:


Switch to a normal port to check if the negotiation mode of the port is normal.


Switch to a normal port to check if the SFP is normal.


Switch to a normal port to check if the network card is normal.


Direct connect to other port to verify if the port is normal


Check the firwall and route configuration is correct.


Catch the packets to check the location of the root cause.

Excluded the suspect of the cable, SFP, NIC, Port, Firewall and route configuration after these 6 steps.

The problem still exist after exchanged the HBA card, but it is normal when direct connect without switch. So the problem should be caused by switch.

After analyze the packets catched on the storage, we found the storage can answer the ARP message from switch. But there is no response from switch after that. 

Everything become normal after customer replaced the switch, so it confirmed the problem was caused by the switch.


Onsite Solution: Replace the fault switch.

Temporary Solution: Use the other ports to access the storage, don't use the P0 port.

Other solution: 

1. Reboot the switch.

2. Unplug and plug in the cables on the other ports of the switch.