Caennanu

So recently i switched to this docker from the spaceinvader one, simply cause its more up to date. Am running this docker for 1 - 2 weeks now. However, since yesterday it seems to have gone haywire. I log my events to SQL so i can more precisely pinpoint the motion. The timestamps of event logging are now 9 hours off and i am on UTC+1. The wallclock is showing the proper time and i've set 'use wallclock' to yes. What is happening here?? Next to this, the motion detection seems to have gone haywire too. All of a sudden it is making 2 - 24 hour vids, while timer is set to only record 5 minutes. And since i am recording 5mp at 25fps, these files are . . . quite large, also since i'm using RTMP and not using a variable bitrate. Any1 know where i can even start to analyze these issues? Here a screenshot of literally the last moments. Also the motion detected seems it is outside of the motion area, like it is completely neglecting it.

I have succesfully extracted the Vbios from a HD 5430 mobility Radeon, But am unable to boot an VM with it, so far --update, i have succesfully booted an VM with this Vbios. --update2, i have 3 VMs, but only 1 of them is able use the passthrough, the others keep throwing errors the device is bussy (1 VM active at a time, so not trying to share the resource) And switching it on and off a couple times between VM's tends to crash Unraid alltogether. AMD5430.rom

So fast forward a 'couple' days. For the Dahua camera i still have issues, i tried putting in the RTMP address (or compose one anyhow) but have failed so far. It all works, motion detection and the works, but for some reason the monitor stream (the substream) keeps blacking out. If motion occurs at that time, the vid is still made tho, so essentially no issue just quite annoying. The Reolink camera's i have switched to RTMP (uses a bit more bandwidth . . .). Unfortunately, this does not solve the monitor's going black. In terms of monitor's going black, i see nothing that could explain it. No errors are throw, nothing written to SQL. Both client side and server have plenty of resources available. Below a screenshot of the blacking out. As you can see, no errors, nothing. Heck they don't even do it all at the same time. And then the stream starts up again. no messages, no errors, no nothing. . . If anyone has any idea's, i'd love to hear them!

Interesting, can't wait for an explenation. Hadn't seen myself that the SpaceInvader one hadn't been updated in a while

i been looking around to get between a 1050 and 1660. but the GPU market right now is completely out of whack. The 1030 has no encoding streams, where the 1050 (ti) and 'up' have 3 encoding streams simultaneously. (there are some more cards, check this link.) Now when all of your camera's are set to copy, it won't use much cpu / gpu and won't need encoding. Unless you use object / motion detection. This will result into higher CPU usage if you do not have a GPU doing it for you. (read more here) This is the reason i'm going for a card that can encode 3 streams, so i can set object detection and offload it to the GPU. I hope this helps.

Thanks for the reply. I have not tried this, but this looks like a good option i might explore. Right now, one of my 410's is acting up, losing connection constantly. Cable tested just fine (with the klein VDV501, or v3 short) so i'm thinking ima need to RMA it or the operating temps are currently . . . over its limit (lots of snow here in holland). In terms of the Dahua, its a 1.3mp bullet cam. nothing fancy as i wanted to just pick up one quick and test out if the quality was good enough. I have changed from Poseidon to HLS this stopped most the blacking out. After this i have found out how to enable the secondary stream. This seems to have solved most my issues. For Dahua i have turned on VBR (i also stream so ofcourse i turned on VBR as i knew the benefit).

So after a bit of testing, i have come to a somewhat conclusion, and i hope others can confirm i'm right or know a bit more tweaks, as i'm not out of the woods yet. I was using poseidon as the codec, when switching to HLS i seem to have atleast a near constant stream on the dashboard. The C410's are running about a minuten behind on reality and the Dahua is running about 5 minutes behind (the substream refuses to work now?) However, also with FLV i seem to have issues. Every so often (its sporadic) the camera's go black. The recordings are fine and do not have this issue. Any1 know why the camera on the dashboard / monitor go black? or why my substream on the dahua doesn't want to work?

There is an option for this yes. If you look under the monitor settings, you will find a 'presets' section. In this you can schedule the presets via a schedule. Simply make a preset, add the camera's to them and apply an schedule.

Gday all, Recently started installing camera's and was planning on using shinobi docker on unraid (for now without hardware acceleration). Have been having some issues, primarily with setting up the monitor. The feeds just keep freezing up, runs for 10 seconds roughly. This happens both on the mainstream and substream, and both on a Dahua and my 8 Reolink C410's, all are powered via PoE. Main stream is set to 3mp resolution with 3000kb/ps bandwidth, codec set to main on the Reolink cam's. Substream is set to main other settings unchanged. Anyone have an idea why the live views from the Shinobi monitor keep freezing? logging in directly, the streams work just fine. Also happens both on local client, and a VM running on unraid itsself, using VNC. CPU assigned cores to the VM hit about 50% with 5 camera's showing. //Caen