Odd IVS false alerts

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
Damn it.. spoke too soon. :banghead:

Log section below. You can see how it comes in a cluster usually. I see nothing unusual preceding it. I've double checked everything above again as far as motion detection beyond IVS. Nothing moving at all to even trigger any kind of motion detection. I'd upload the clip but all you'd see is a static image as in the pictures.

I'm kind of lost as far as where to look now. Guess I'll check the power and run a continuous ping to see if I ever see it dropping faster than maybe BI or the cam logs pick up. Beyond that not sure. :idk:

3 7/24/2022 11:43:33.853 PM Car EXTERNAL
0 7/24/2022 11:43:34.686 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:43:44.837 PM Server [::]: Connected
0 7/24/2022 11:43:45.108 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:44:58.676 PM Car EXTERNAL
0 7/24/2022 11:44:59.520 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:45:09.897 PM Server [::]: Connected
0 7/24/2022 11:45:10.147 PM Car Push: OK to Mike's iPhone 13 Pro
10 7/24/2022 11:45:20.034 PM Server [::ffff:192.168.2.240]: Connected
10 7/24/2022 11:45:20.081 PM mla1 [::ffff:192.168.2.240]: Login
0 7/24/2022 11:45:20.685 PM Geofence iPhone 13 Pro: Inside
3 7/24/2022 11:45:29.660 PM Car EXTERNAL
0 7/24/2022 11:45:30.321 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:45:41.898 PM Server [::]: Connected
0 7/24/2022 11:45:42.144 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:46:16.715 PM Car EXTERNAL
0 7/24/2022 11:46:18.658 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:46:28.302 PM Server [::]: Connected
0 7/24/2022 11:46:28.553 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:47:21.662 PM Car EXTERNAL
0 7/24/2022 11:47:22.562 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:47:32.872 PM Server [::]: Connected
0 7/24/2022 11:47:32.976 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:47:37.279 PM Car EXTERNAL
0 7/24/2022 11:47:37.869 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:47:48.837 PM Server [::]: Connected
0 7/24/2022 11:47:48.941 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:47:51.505 PM Car EXTERNAL
0 7/24/2022 11:47:52.128 PM Car Web: HTTP/1.1 200 OK
3 7/24/2022 11:48:02.588 PM Car EXTERNAL
10 7/24/2022 11:48:02.621 PM Server [::]: Connected
0 7/24/2022 11:48:02.718 PM Car Push: OK to Mike's iPhone 13 Pro
3 7/24/2022 11:48:40.580 PM Car EXTERNAL
0 7/24/2022 11:48:41.247 PM Car Web: HTTP/1.1 200 OK
10 7/24/2022 11:48:51.728 PM Server [::]: Connected
0 7/24/2022 11:48:51.988 PM Car Push: OK to Mike's iPhone 13 Pro
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
Should I be seeing the yellow box around the vehicle with regular motion off, SMD off, and target filter turned off for IVS?

I don't typically but I suppose the rain is causing some difference.


5442-IVS7.png
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
On another side, my switch tracks power events and network errors. I see nothing that looks unusual as far as power and 0 errors on the networking side. I'd expect to see some events/errors at least if it was dropping offline all the time.
 

SpacemanSpiff

Known around here
Joined
Apr 15, 2021
Messages
1,467
Reaction score
2,470
Location
USA
So motion detection (MD) does NOT need to be enabled for IVS rules to trigger recording? Is this only when the cam is connected to BI, or does it also apply to DVR and memory card recording destinations?

Past experiences with other IP cams I've come to know this as a global setting that affected all motion settings of the respective make/model. With the years of improvements that have happened since learning that... I think I need to revisit this options' affect on other makes/models. :embarrassed:
 

looney2ns

IPCT Contributor
Joined
Sep 25, 2016
Messages
15,606
Reaction score
22,833
Location
Evansville, In. USA
So motion detection (MD) does NOT need to be enabled for IVS rules to trigger recording? Is this only when the cam is connected to BI, or does it also apply to DVR and memory card recording destinations?

Past experiences with other IP cams I've come to know this as a global setting that affected all motion settings of the respective make/model. With the years of improvements that have happened since learning that... I think I need to revisit this options' affect on other makes/models. :embarrassed:
No, if you use IVS that is all that should be enabled, otherwise you are defeating the purpose of using IVS.
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
Yeah, motion detection needs to be disabled otherwise you're using the basic motion detection and you'll get a ton of falses. Which I am now but shouldn't be that way with IVS.

Thanks re the yellow box. Seems to be a permanent addition now since I re-did everything. So that is different.

Not sure what's up with this. I'm really lost for any explanation.
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
OK, this is kind of driving me nuts now. I've turned off EVERYTHING EVERYWHERE that I can find and the cam STILL is generating seemingly random events.

To review what I've done (and I'm double checking each of these again as I list them to be absolutely sure):
  • Reset the cam 3X a few times.
  • Unchecked EVERY enable box for EVERY type of motion detection/events.
    • -- Regular Video Detection not enabled.
    • -- SMD not enabled.
    • -- Audio Detection not enabled.
    • -- Smart Plan light bulb not highlighted.
    • -- IVS completely deleted all rules, under Global Setup tab, Anti-Disturb not enabled. There are no rules.
    • -- Face detection, people counting, heat map, abnormality all not enabled.
  • Installed SD card. With no events (as above) to trigger recording/snapshots, as expected nothing is recorded.
  • Checked the cam logs. Nothing but logins, logouts, config changes, the usual things as I've been playing with it.
  • Checked BI logs, Nothing unusual. No drop offs, 0 loss of signal, etc. Events come in from cam as EXTERNAL.
  • Checked switch stats. No unusual power or data events. 0 network errors for the port for as long as the switch has data (months).
  • Day/night, obvious motion in scene or completely still seems to make no difference. Appears to be completely random. I see no pattern in timing or times of day. IR on/off , Day/Night schedule makes no difference.
  • In BI, made sure that the cam isn't a member of any group that might be triggering it (though those types of events should show as GROUP).
  • In BI deleted and recreated the cam completely from scratch several times.
Camera works great otherwise. Positive detection is fine. I have 6 other Dahau cams running IVS and they all work perfectly. I've never had any issues with any and all are set up in basically the same way.

Again of note too that adds to the puzzle is that I had another 5842 cam in the same location with the same view temporarily and it too did the same. Never really looked at it much since I'd just hung it up to check that it was working waiting for the 5442, but I do recall the same type of random false events happening. That same 5842 now looks at the same scene from a different angle and does not generate the events. Also another 5442 set up in the same way I did this one and looking at the same area from farther away does not do it.

WTF could it be??? Bugs me most because I just don't understand. lol
 
Last edited:

sebastiantombs

Known around here
Joined
Dec 28, 2019
Messages
11,511
Reaction score
27,692
Location
New Jersey
Just for a laugh pull the camera down and test it on the bench. I'm wondering if you're getting some EMI in the CAT cable or there's a problem with that cable. A simple continuity tester may show it as fine, but there's more to it than continuity.
 

tigerwillow1

Known around here
Joined
Jul 18, 2016
Messages
3,844
Reaction score
8,507
Location
USA, Oregon
You seem to have logically eliminated everything as a cause of the unwanted triggers. With the 5442s I went through 5 different firmware versions before finding one where IVS worked the way I expect it to. I have a single 4k-x and its IVS detection is totally lame for my purposes, while it could be perfect for other users. I'm waiting and hoping on a new firmware for that one. Your finding that the 5842 did the same thing blows all the logical conclusions out of the water. I ran a 5442 and 5842 side-by-side for a couple of weeks and the IVS in my favorite 5442 firmware version worked a bunch better than the 5842's IVS. The 5842 missed a lot of real positives that the 5442 caught.
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
One thing that I had not tried was updating the firmware. I did that after I posted earlier to the 2022-02-18 version. So far, so good... but I've said that before. ; )

If that doesn't work then I'll pull the thing down, re-terminate/punch things down again, and swap another 5442 into that position.

If it still does it then, I'll just give the fuck up. lol
 

bigredfish

Known around here
Joined
Sep 5, 2016
Messages
17,322
Reaction score
48,375
Location
Floriduh
Been sorta following this and it is a head scratcher.
One thing I don’t see is any mention of bourbon, it’s part of my standard installation protocol ( as well as anti-covid therapy)

Seriously, the part about the other cam doing it, the shot of the trigger on the lamppost, makes me think it is indeed the lamppost.
Can you block it from view with a big piece of cardboard or somthin and test?

Im not a BI user so can’t add to what others have already mentioned.

I have a stop sign in a scene that a ptz and a 5442 like to think is a human now and then…
 

Mike A.

Known around here
Joined
May 6, 2017
Messages
3,828
Reaction score
6,386
Ahhh... Got it! The answer, as usual, is user error. I'm a dumbass...

I have a motion sensor attached to that light pole to turn the light on/off at dusk/dawn, make it brighter on motion, etc. I'd forgotten that at one point playing around with Home Assistant I tried to use that same sensor to determine if that car was leaving/arriving in the driveway. Ended up going a better route with that and thought that I'd killed off that part of the automation. At the time that I did that, the BI name for that cam was Car. About that same time I also changed cameras around and used another cam to look at the cars which I named Cars (plural). The cam in that position then was turned to look at something else and took another name. So at that point the automation, one part of which was to trigger a BI cam named Car, still was running. But since there no longer was a cam named Car nothing happened when it did and I didn't notice. Fast forward a few months to when I get these new cams... I change things around again, turn that cam back to look at the car, rename it back to Car. So...
:smash:

Yes, it was the motion sensor triggering the cam from part of a Home Assistant automation that I'd completely forgotten about. I even looked at my HA stuff but the way that it was done I didn't notice that part. That explains why two different cams were doing the same thing - they both were named Car in BI.
 
Last edited:

IAmATeaf

Known around here
Joined
Jan 13, 2019
Messages
3,304
Reaction score
3,282
Location
United Kingdom
Similar thing happened with me but with time syncing. All my other 7 cams were time syncing except 1 that would drift.

Checked and rechecked, enabled logging in NTP Time Service, checked the logged, reset NTP in the cams.

Eventually realised that the IP address of the NTP server was 192.168.0.30 when it should have been 192.168.1.30 but the hundreds of times that I must have checked/reviewed the config just couldn’t see or spot it.

Was a proper DOH! moment when I eventually realised.
 
Top