Arlo|Smart Home Security|Wireless HD Security Cameras|NETGEAR

The Arlo Community

Learn, explore, ask questions, and connect with our community of customers and experts.

Top Contributors
0 Kudos

recordings and filename time stamp

There have been more than a few people asking for continuous recordings while movement is detected, and then being able to set the period of time to record after movement has stopped; I'm guessing here that one of the limitations is going to be the file size of the recordings.  Yet I have 'cheap' action cam's for my motorbike which record no problem by just creating multiple files instead of just one huge file, so if a $50 cam can do this I can't see why Arlo can't take the same approach to splitting up continuous recordings into more manageable small files.


However, the two issues I'm having are around accessing the recordings.  


Having attached a USB drive to the base station for local recordings, there is no 'easy way' to access these.  having to take the drive off-line, then wade though subdirectories trying to find the correct file is hardly user friendly.


Could we at the very least get an index file (plain text minimum, csv preferred) with the time/date stamp of the recording, camera name and then filename of the recording?  Have an App to access the recordings in the same was as the cloud recordings.  Or better still, be able to access the recordings in the same way at the cloud recordings without having to take the USB drive off line.


The suggested method of attaching a second USB drive to allow one to be taken off line isn't helpful, as this then creates a headache when trying to find a recording later on, as the recordings for the same day can be split across different USB drives.


On the subject of accessing recordings in the cloud, if I find via the web a recording I wish to save, I can download to my PC/Mac however the downloaded file does not give the recordings time/date stamp to match the time and date of the event, but the time date the file was downloaded for some obscure reason.


Do you know the file name is the date/time in Epoch ( computer time ) and can easily be converted to human time


exp; 1535356785139.mp4

is  Assuming that this timestamp is in milliseconds:
GMT: Monday, August 27, 2018 7:59:45.139 AM
Your time zone: Monday, August 27, 2018 3:59:45.139 AM GMT-04:00 DST
Relative: 5 months ago


Thanks for that, if I was converting a single filename from a single camera source to get the time and date stamp that maybe would be liveable, however does the file's name relate to the start of the recording or end time?


I just checked the USB drive for the 11th of Jan and there are 173 files in one subdirectory and about a dozen in another subdirectory with the "Date modified" date matching 11 Jan.  So, you can't even find a single day in the same subdirectory not alone then having to check the first few files last modified on the 12th of Jan for those that were started to be recorded on the 11th, but were upto the current maximum of 3 or 5 minutes long (depending on camera), thus puting the date/time stamp on the modified into the first few minutes of the 12th.


Eitherway, there is still no relationship of the filename to the camera that recorded it.  Since I have a mix of Arlo and Arlo Pro 2 cameras, I can guess that the smaller files are from the original cameras, but this ignores the fact that I have them all set to record while movement is detected so some of the smaller files may well of come from the newer cameras.


If, we get what a lot of users are asking for which is longer record times in place of the current maximum recording time of only a few minutes then this would make matters worse.




OK, found there is a relationship with the filename on the USB drive and now have a simple script to separate these out in each of the Arlo created subdirectories, and then place the files into new subdirectories for each room:-


mkdir kitchen
mv 4N7278S56DB82_*.* kitchen/ 
mkdir hallway
mv 4N727BS8A0CE1*.* hallway/

and so on....

Still hasn't solved the issue of timestamp but this will do for now.