Monitoring Status of Cameras

Craigsweb

n3wb
Oct 25, 2018
3
0
South Africa
Hi guys

I am looking for suggestions on the best way to monitor the status of cameras. ie is the camera/ channel healthy. My client has 100s of cameras and I am busy looking at ways to notify him that a camera is not working in anymore before his client phones him... I was hoping to use SNMP but I cant find any information on the NVRs MIB. Does anyone know if it is possible to monitor the status of channels or cameras with SNMP
 
Hi Sorry, my client mostly has hikvision and dahua, I build web apps and currently building a platform for my client to monitor the status of all his systems. Most of the systems are not monitored and only used when there is a incident. I was hoping to poll the NVRs with SNMP and determine if a camera has gone down. Been searching all day for information on if this is even possible and its not looking to optimistic.
 
Hi Sorry, my client mostly has hikvision and dahua, I build web apps and currently building a platform for my client to monitor the status of all his systems. Most of the systems are not monitored and only used when there is a incident. I was hoping to poll the NVRs with SNMP and determine if a camera has gone down. Been searching all day for information on if this is even possible and its not looking to optimistic.

My 5216-16P-4KS2 has alarm output, which you can direct "IVS" event to, eg video loss signal can be triggered there. Then you can wire a physical alarm system to it. Or you can have an email sent out on these events. Then you have to monitor only that mailbox. Lots of choices!
 
I have not actually tried this - so I don't know how well supported this API is - but maybe it's worth a look :

Good point @alastairstevenson ! Dahua also has these API's, but then you'll need to construct your "alarm" app around it. I remembered seeing a nodejs gem (from @nayr ) - here it is: NodeJS Module: node-dahua-api

This nodejs module can read out alarm states of dahua stuff, which means that if the alarm is already logged, it can harvest this information.

That may help to find a solution for TS use case.

Good luck!
CC
 
Nagios can notify you after doing intelligent deeper inspection via HTTP, SNMP, SSH, Telnet, etc. More usefully, I use it to LOG INTO the camera interface using credentials every 5-30 minutes depending on camera and location. I've found if the UI can't load through the authentication process, the video feed is also tapped and needs service. You can also schedule actions based on down alerts etc. I don't use SNMP to do monitoring on the cams, but use it for other services fairly well.

You can tap this into the NVRs and other servers to predict usage and get health checks on pretty much anything/everything. If you want to step up your game in the monitoring realm, this would be where to start. ;-)

Here are screen shots of my personal test machine:

WSIPYUs.jpg
g9Dj2TG.jpg
DvAsglF.jpg
U8DULE4.jpg
 
  • Like
Reactions: alastairstevenson
Hi gmaster
Thank you for the great response. Could you give me more information on what plugins you used on nagios. I am finding the nagios website kinda hard to navigate. secondly are you saying that a http login with a 200 return and a ping is sufficient for status monitoring?
 
Hi gmaster
secondly are you saying that a http login with a 200 return and a ping is sufficient for status monitoring?

I can chime in on the latter: Yes, a HTTP login (with authentication) is enough to know that 1) the ip is reachable, 2) the HTTPd server is running, 3) the account (user/pass) is working (otherwise you'd receive a 403 Forbidden) nor a HTTP 500 (error). Testing out the plugins (to actually view a stream) would be a bridge to far, however an NVR could detect a "video loss" signal. If you combine these http logins and a regular ping request (then you'll have actual round trip timings), you have a pretty good idea of the status of your cams. Back in the script kiddies days, people used to dump "top" and "uptime" statistics in cgi scripts, and let crawlers pull that information inside monitoring dashboards.

Hope this helps!
CC