Is it normal for Dahua NVR's to timeout under PING test

Kalan

n3wb
Joined
Apr 17, 2018
Messages
14
Reaction score
6
Location
Australia
Hi there,

Following on from exisiting "problems" with a new Dahua network camera... it's worth posing this question in another forum.

I have a Dahua DHI-NVR5464-16P-4KS2. I believe it's on the latest firmware as I updated it the other week.

Is it normal for these NVR's to timeout when I perform a simple ping to the NVR's address?

Attached is a screenshot of the results I'm receiving and a text edit file with the results from 120 pings. It's around 25-30% packet loss in the past 24 hours I've been monitoring it.

Thanks.
 

Attachments

IAmATeaf

Known around here
Joined
Jan 13, 2019
Messages
3,287
Reaction score
3,252
Location
United Kingdom
There’s something wrong, even when it works the ping times are terrible. I’d start with checking the cabling.
 

catcamstar

Known around here
Joined
Jan 28, 2018
Messages
1,659
Reaction score
1,193
What I do in these cases: take 2 linux devices, put a premade cable between them (eg 1 meter). Run iperf, and measure the available bandwidth. This is your baseline. Now plug these devices between your endpoints (one in the place of the NVR). Rerun the iperf and voilà. You immediately know where the bottleneck is (but not what the root cause may be: bad cabling, wrong setting in switch, ... )

Good luck!
CC
 

SouthernYankee

IPCT Contributor
Joined
Feb 15, 2018
Messages
5,170
Reaction score
5,320
Location
Houston Tx
Take the camera down and test with a short premade store purchased cable.
What are trhe TWO ip address of the NVR ?
Is the camera in the POE ports at the back of the NVR ? It is strange that the NVR subnet is use a 192.168 address.

Private ip addresses. Local IP addresses. These addresses are NOT used by the internet. They are for your local home/business network.
10.0.0.0 to 10.255.255.255
172.16.0.0 to 172.31.255.255
192.168.0.0 to 192.168.255.255

Note there is no reason to redact local ip addresses when posting.
 

Kalan

n3wb
Joined
Apr 17, 2018
Messages
14
Reaction score
6
Location
Australia
What I do in these cases: take 2 linux devices, put a premade cable between them (eg 1 meter). Run iperf, and measure the available bandwidth. This is your baseline. Now plug these devices between your endpoints (one in the place of the NVR). Rerun the iperf and voilà. You immediately know where the bottleneck is (but not what the root cause may be: bad cabling, wrong setting in switch, ... )

Good luck!
CC
I have one Linux device available.... but not two. :(
 

Kalan

n3wb
Joined
Apr 17, 2018
Messages
14
Reaction score
6
Location
Australia
Take the camera down and test with a short premade store purchased cable.
What are trhe TWO ip address of the NVR ?
Is the camera in the POE ports at the back of the NVR ? It is strange that the NVR subnet is use a 192.168 address.

Private ip addresses. Local IP addresses. These addresses are NOT used by the internet. They are for your local home/business network.
10.0.0.0 to 10.255.255.255
172.16.0.0 to 172.31.255.255
192.168.0.0 to 192.168.255.255

Note there is no reason to redact local ip addresses when posting.
The problem is with the NVR, not the new camera (I assume).

Most cameras that we have connected are going to POE switches which are then added into the NVR manually/auto search. 10 of which ARE going into the NVR's POE switch but that's semi-unrelated to the device timing out on a ping test I would think.
 

Kalan

n3wb
Joined
Apr 17, 2018
Messages
14
Reaction score
6
Location
Australia
There’s something wrong, even when it works the ping times are terrible. I’d start with checking the cabling.
I believe I have resolved it. Now I will preface this by saying I never installed the hardware so the exact specification's are only being revealed now.


TL;DR: I moved the network cable of the NVR from a 10/100mbps Edimax POE switch to the main rack switch.


We have an Edimax 8port 4POE switch that's used for connecting the surrounding switches with security cameras attached to them (I assume I'm 99% correct). The Dahua NVR was plugged into that.

First thing I did this morning was re-locate the cable between Edimax switch and the main rack switch. So the NVR is now plugged in directly to our larger switch.

Immediately all packets that were sent got delivered, simple fix... didn't think much more of it other than how easy it was and maybe there was more to it.

As I was in the room, I tinkered with the quality and bitrate, and to my shock the latency of most cameras shot through the roof... 300ms up to like 2500ms. I didn't really know why so I reverted all cameras back to the settings I had noted down. The latency lowered and was back to normal (1.3 to 3ms).

... now I've only just found out but the switch in the server room with all of the other switches connected to it is 10/100mbps... and I assume the other switches with the cameras connected to it are the same kind. The invoice I have from a few years back seems to be likely.

One would suggest that the main hub for all the switches should have more bandwidth to cope with decent amounts of bitrate if I were to increase 40 cameras of VBR 3000kbs to (even) 4096 and from high 4 or higher 5 or highest 6.


Are my findings clear, and does this all make sense? I've learned a bit today, it isn't really my primary interest.
 
Top