OpenALPR Webhook Processor for IP Cameras

I ended up reverting the snapshot, running on the old version again. On the newer version for some reason I have no docker container logs, so I can't really even tell what its doing

With it being a VM, if I increase the disk size even more to do the migration, there is no nice easy way to shrink it back down again

I think the plan now is to increase the size of the current VM, hopefully let it finish the migration and then move it all to a new VM with a reasonable about of space
 
  • Like
Reactions: CamCrazy
I ended up reverting the snapshot, running on the old version again. On the newer version for some reason I have no docker container logs, so I can't really even tell what its doing

With it being a VM, if I increase the disk size even more to do the migration, there is no nice easy way to shrink it back down again

I think the plan now is to increase the size of the current VM, hopefully let it finish the migration and then move it all to a new VM with a reasonable about of space
Sounds like a good plan, mine is ticking over fine today on Windows 11 machine, not decided whether to retain existing database or start fresh.
 
I would start fresh, but I find the historical information very useful
 
Alright, 700GB of space added, and off we go again. Hopefully its enough this time
 
  • Like
Reactions: CamCrazy
Still no logs for the container for some reason, but the phase of 100% CPU usage on 1 core while the DB increases in size seems to have completed, and now we're onto some kind of part where its multithreaded

Still no Web UI yet, hopefully it comes up eventually
 
It finished! And now I also have Docker logs

So far, so good! I'm on 5.2.0 as going to 5.3.0 gave me an error, so now I'm on 5.2.0 I'm going to make sure its all good before moving to 5.3.0
 
Going to 5.3.0 seems to work, and the Webhooks are getting through because I see the HTTP 405 errors since I never updated the URL, but logging in gives me "Internal Server Error" in the top left

I reverted the snap back to 5.2.0, and will try again tomorrow. Very happy to finally be on 5.2.0
 
sqlite doesn't handle drop/create operations very well. while your db size has double, half of it is "unused" sqlite doesn't give up storage space unless you run a vacuum. you can run it yourself if you want to shrink the db
https://stackoverflow.com/questions/18126997/how-to-vacuum-sqlite-database
Try running a browser incognito and see if it works then, you might have some frontend code cached.

Should I be running this from within the container? I tried and get an error that sqlite3 is not found

# sqlite3 /app/config/processor.db 'VACUUM;'
sh: 11: sqlite3: not found
 
Running outside the container now, wish me luck boys!

1718332226689.png

Initial size, 329gb. It doubled after the upgrade to v5
 
Ended up taking about 11 hours in total, back down to 164GB, so what it was before the upgrade

So now I'm on 5.2.0 with a compacted DB. Maybe later on I will be pushing forward to 5.3.0
 
  • Like
Reactions: CamCrazy
Ah shoot! I have a problem

Currently on 5.2.0, and its getting new plates, and I get the little alert for every new plate with the WebUI up, however the plate doesn't show unless I refresh the page

Also, the search does not work, and if I try and add a new user, it doesn't save. Anyone got any ideas of the problem? Tried clearing cache and all that jazz, nothing

If I upgrade to 5.3.0, I got an error in the Docker logs about user auth, so I'm thinking there may be an issue with the user account, I was going to make a temp user to test, but can't save it
 
Last edited:
Well, I don't know what fixed it. I was getting frustrated and moved to 5.1.1, still broken. Move to 5.1.0, still broken, 5.0.0. Back to 5.2.0, still broken

Now, it just works...

Still can't edit users, maybe I'll wait a few more hours and that will fix itself too!
 
@CamCrazy did you say you have an issue with plates being counted twice? Because now with my fully working setup, I have the same issue!


1718394730927.png

1718394759217.png
 
@CamCrazy did you say you have an issue with plates being counted twice? Because now with my fully working setup, I have the same issue!


View attachment 196456

View attachment 196457
Sorry, been tied up with work, yes, I did have duplicate plates but that was due to me running the system on original machine at the same time, once disabled on original machine, duplicates stop. Sorry, that wasn't helpful for you. With the latest release, in theory you may need to check permissions on the db file also.

I tried moving my original db to the new install (should that just work!?) and docker fails to run, logs show SQLite Error 11: 'malformed database schema (RawPlateGroups) - invalid rootpage. IP address on the new machine is different, all else identical, I checked permissions on the db file.

Sorry for slow reply :(
 
Sorry, been tied up with work, yes, I did have duplicate plates but that was due to me running the system on original machine at the same time, once disabled on original machine, duplicates stop. Sorry, that wasn't helpful for you. With the latest release, in theory you may need to check permissions on the db file also.

I tried moving my original db to the new install (should that just work!?) and docker fails to run, logs show SQLite Error 11: 'malformed database schema (RawPlateGroups) - invalid rootpage. IP address on the new machine is different, all else identical, I checked permissions on the db file.

Sorry for slow reply :(

Interesting, did the original count on all the existing plates go back down?

Are you sure that was the cause? I can't see how running 2 completely independent systems would double the count on one

I'm starting to wonder if I should just roll back before all of this back to a v4 release
 
Do both duplicates show the image or a ? symbol??

Its the other way around for me, for example I see 11 entries which is correct, but it reports that the plate has been seen 22 times

If I find a net new plate, it reports that its been here 2 times, but when I search, it only brings back 1 time
 
Interesting, did the original count on all the existing plates go back down?

Are you sure that was the cause? I can't see how running 2 completely independent systems would double the count on one

I'm starting to wonder if I should just roll back before all of this back to a v4 release
Yes, it was the cause, sorry, just seen your comment about 11 captures but 22 times! that is bizarre? how is that happening.
 
I am wondering if I need to upgrade my original version 4.x.x to newer release before moving the database? wonder if that would help, thought I was winning there for a minute :)

**Update, dropped back to release 5.0.0 and on running the container it is sitting at 18-20% CPU so hopefully doing its thing now...will wait for CPU to drop then try opening.
 
Last edited: