wow… that’s worrysome though… if you were up to, it would be great if you tried to replicate that. I run tests that create users and it seems to work fine.
Any details of your configuration? any other things you did or manually configured/installed on the Raspberry Pi?
Ok, i I opened a new Issue on Github:
Maybe this helps…
/
/
thank you 
Hi,
thanks for all the work you put into this. And, your setup is EXACTLY what I was trying to achieve manually for months now. So, last night I downloaded NCPi 12.3 as a torrent, put it on my RPi’s card and booted -it starts fine. I saw a couple of “[FAILED]” fly buy but didn’t think much of it.
I went into raspi-config to set locale, timezone, keyboard etc, but none of it worked, all of them came up with a general failed message. Ah well, I’ll do that manually later, right? So I try to update the Pi from the config, but get a fail as well.
Turns out, the Pi get’s no network, ifconfig shows only loopback but no other devices. And upon checking, lo and behold, one of the [FAILED] on boot was “couldn’t activate dhcpcd for all devices”.
Huh… So I manually set a static IP in /etc/dhcpcd.conf and rebooted, but same issue.
Should I open an issue on GH for this? What kind of info can I give you to help? Personally, I’m baffled by this.
The RPi is just 4 weeks old, the SD was bought along with it (no no-name junk). I’ll try and find out how to check for read/write errors.
There’s two HDDs attached via a USB-extender, that’s got it’s own PSU.
I guess I can see if things improve with a different PSU for the PI itself.
Thing is, before I installed your image, the PI was running just fine, same hardware, but vanilla raspbian.
I build on plain raspbian, so stuff in raspi-config is unmodified. That’s what gets me suspicious
I agree, first instinct is: Read / write problems. So, I’m testing the card following
this.
I created a random image file, and dd’ed it onto the card (I skipped the speed check).
dd if=/dev/urandom of=rnd_init_data bs=1024 count=16000000
16000000+0 records in
16000000+0 records out
16384000000 bytes (16 GB, 15 GiB) copied, 192,667 s, 85,0 MB/s
then
sudo dd if=rnd_init_data of=/dev/sde bs=1024 count=16000000
dd: error writing '/dev/sde': No space left on device
15558145+0 records in
15558144+0 records out
15931539456 bytes (16 GB, 15 GiB) copied, 2879,82 s, 5,5 MB/s
The “no space left” message is a bit odd, as I have a 16GB card. Plus, only 5,5 MB/s is low, then again I’m using an old USB 2.0 card reader, might well be the bottle neck.
Cutting this short: Ultimately, the result of both the data on the card and the trunctated original is identical (I used slightly different names):
6091a2906782b702d097f48afc81467c rnd_init_data_read
6091a2906782b702d097f48afc81467c rnd_init_data_short
Ultimately, I put your image back onto the card, ran fsck, comes out clean…
Well, riddle me this… inserted the dard into the RPi again, ran it, it works. So…huh…
Off to explore what you’ve built there. Again, thanks
well that’s a mistery… hopefully it will work well 
So, happy to report, it works splendidly. Thanks a lot for your fabulous work.
Question: Anything I should be aware of when adding new services to my Pi? Namely, I would possibly run a blog off of it (which I assume won’t be a big issue) and might want to add PiHole to my RPi as well (THAT could be a challenge, as PiHole uses lighttpd and has to run on port 80). Would it help if I ran PiHole as a Docker container?
That’s all for now
Hi,
Happy to hear that.
Well, definitely there can be problems mixing web services. Look for instance at this
For that reason, it is interesting to run stuff in docker containers (I am almost done with updating the NCP docker image).
The problem with this is that it can be a little bit of a waste in resources to have several web servers running in parallel. Obviously it will also depend on the usage. If you only ever use them to access the admin page could be ok.
I suggest you open a new thread and ask people. Maybe somebody has already done it 
On a different note:
When runnning the nc-backup through the webpanel it times out at
Maintenance mode enabled
backup database...
backup base files...
backup data files...
I checked via ssh and there is a .tar file created in the correct folder (on a usb HDD, mounted via nc-automount) that keeps growing and growing, even after the webpanel reported
Something went wrong. Try refreshing the page
I assume that the webpanel just times out while the backup process still runs, right? Anything I can do about this?
The last time I ran the backup it created a 91GB tar file (I don’t remember if I somehow stopped the process). I know, that might not exactly be small change, but it’s not excessively huge either, is it?
That is a big backup. I have not been reported this, but it is possible.
If you can reproduce it, I suggest you open an issue with the relevant information (version, and the rest).
There’s people that have tested backups as big as 100GB and 500GB without issue.
I’ll check that out, thank you. Almost looks like the last few automatic backups worked, but I’ll look into it.
Different (or related?) issue: The sync client currently fails with this message:
The downloaded file is empty despite the server announced it should have been 2,2GB
Strangely, checking on both server and client, the file is there, and it shows identical size on both. Any idea?
Unfortunately, until NC13 32bit systems (such as the Raspberry Pi) do not support uploads over 2GB.
I guess that’s what it is
Hmm… yeah, that would be a reason. Not a huge loss, as all of those I can redownload anytime. Thanks