User avatar
Posts: 362
Joined: Sun Dec 31, 2017 4:35 pm
Location: UK

Well that was a chore

Sun Jul 14, 2019 8:21 pm

After setting up my Pi4 as an NFS/PiHole server to replace a 3B+ which was doing the job, and serving the OSs for the other PIs via NFS, and realising my 500GB external drive was getting close to capacity, I decided to migrate everything to a 1TB (this was last week). My lesson from all this was don;t do stuff like this at 7am on a Sunday morning after you've been out Saturday evening.

I managed to trash not only the entire data on the 500GB, but also my backup, so music, videos and photos all gone, as well as the OS folders for the other PIs. Now, I've managed to recover from other sources (mobile phones have their uses, it seems), a couple of videos, but since were most ripped from my own DVD collection, apart from leg work, no problem, and I've even recovered my most precious photos.

However, the PIs have had to be re-built from scratch. One in particular I was turning over to a Nextcloud server exclusively. However, imaging to the NFS drive and updating repeatedly failed (no idea why, one of the other 3B+s is just fine), and I've eventually settled on the NextcloudPi image - now working but still on MicroSD. Waiting to migrate to NFS, but even getting this far has been a chore.

Suppose it's not helped by the media etc. being on the NFS drive connected to the Pi4, but even so, I seem to remember it being much easier 1st time around.
Headless PI. OMG, someone cut it's head off. Oh, hang on. it didn't have one to start with.


Re: Well that was a chore

Sun Jul 14, 2019 8:50 pm

Feel for you :o :cry:

I put my PiHole on a zero and do not touch it as I once found myself baffled why I could ping by IP and not name having taken the server down :oops:

Glad to read you've been able to get the precious pics back - printing had an advantage (well for 70 years or so).

Posts: 3318
Joined: Sat Mar 17, 2018 7:29 am
Location: Anime Heartland

Re: Well that was a chore

Sun Jul 14, 2019 10:39 pm

The lesson I learned when I encountered a similar situation was:

"....never mess with a fully running functional system until the replacement system has been certified reliable....."

To this my steps are:
1. build a new system
2. copy the data from the old system (access as read-only) to the new system
3. run the two systems in parallel for a few days (or weeks), mirroring the data from the old system to the new
4. slowly migrate to the new system
5. once fully migrated to the new system, remove (but not dismantle) the old system.
6. the old system will now serve as "offline system backup"
7. the offline backup system's data capacity is upgrade to match the capacity of the new system, with the old drive becoming "backup"

been doing with my SBC NAS systems, starting from RPi1B+, then to RPi2B (in parallel with NanopiNeo2, RPi1B+ re-tasked), and then to Rock64-1G (in parallel with the RPi2B, nanopineo2 re-tasked)
will do the same with the arrival of RPi4B-1G (to run in parallel with Rock64-1G)

Pi-hole has always been relegated to another low-power SBC.
"Don't come to me with 'issues' for I don't know how to deal with those
Come to me with 'problems' and I'll help you find solutions"

Some people be like:
"Help me! Am drowning! But dont you dare touch me nor come near me!"

Return to “Networking and servers”