Apollo Rebuild Part II
Part 2 ended up being more software based and an SSD installation, surprisingly this took up the entire weekend. Let's go into it!
If you missed the first part of the Apollo rebuild you can check it out here.
This build tackled a very worn SSD, replacing it with a 500GB SSD from Samsung. Since I was replacing the drive I figured I may as well mix up the software and give it some much needed attention.
Previously this build ran Ubuntu Server 14.04 LTS, which has since stopped being supported. It also ran a terribly outdated Linux kernel to support OpenVZ, my first foray into virtualisation a few years ago. I can't remember off the top of my head now, but I think OpenVZ running on top of Ubuntu would only run a Linux kernel of 2.6.x - TWO full versions behind currently development.
While OpenVZ was great, it also provided some admin issues - since an entire OS had to be virtualised I ended up maintaining 13 (at a time) OS versions of Ubuntu, all sharing an old kernel.
Needless to say it got fairly messy, so I decided to dive head first into Docker. I'd spent an afternoon tinkering with it a few years back but it broke my brain, I simply couldn't understand it.
Anyway, enough muttering. Here's part 2 of the build for Apollo.
Seeing what VMs need backing up
CTID NPROC STATUS IP_ADDR HOSTNAME
101 - stopped 192.168.1.40 dns1
102 - stopped 192.168.1.41 dns2
103 76 running 192.168.1.42 smokeping
104 953 running 192.168.1.50 unifi
106 47 running 192.168.1.45 ops
107 57 running 192.168.1.46 deluge
108 21 running 192.168.1.47 couchpotato
109 - stopped 192.168.1.48 bitcoin
110 62 running 192.168.1.51 data
111 - stopped 192.168.1.52 requests
112 - stopped 192.168.1.53 mono
113 - stopped 192.168.1.54 newznabMost of these had been stopped for various reasons. I'll run through the list very quickly:
dns1 & dns1 - At a time were dnsmasq instances, handling local DNS
smokeping - Ran pings to various destinations to monitor the (now replaced, woo!) ADSL
unifi - A UniFi controller for my Ubiquiti networking gear
ops - A small VM with remote access for remote testing, ran various things like timestamped traceroutes, MTRs, pings etc via cron
deluge - Torrents
couchpotato - For that wanted movies list
bitcoin - A, soon to be returned, very well performing Bitcoin node
data - Held InfluxDB and Grafana
requests - Plex Requests. This was already migrated to Docker but hosted on a NAS
mono - Testing Mono with ASP on Linux. It was a terrible idea. Don't do this. Microsoft and Linux don't play well together, and when they do anything with ASP is a resource hog
newznab - Dev environment for a Newznab indexer
I plan to move all of these, potentially a little more, over to Docker.
/etc/fstab was also backed up with various NFS volumes.
Also, ^^ Markup learning is happening!
UUID=2ce029f6-2465-49d5-94b2-52791ca43a4a /mnt/storage ext4 defaults 0 2
#
192.168.1.181:/volume1/movies /mnt/movies nfs rw,hard,intr,nolock 0 0
192.168.1.181:/volume1/series /mnt/series nfs rw,hard,intr,nolock 0 0
192.168.1.181:/volume1/dropbox /mnt/dropbox nfs rw,hard,intr,nolock 0 0
192.168.1.181:/volume1/music /mnt/music nfs rw,hard,intr,nolock 0 0```
At this stage I should mention that I'm waiting to order a modular PSU to clean up the cable shit show currently in this chassis as well as proper drive bay mounts for both these drives.
The old faithful Seagate drive is also being replaced with a 2TB WD Black drive.
Also a sneaky view with a desk that I'm currently kitting out - more on this soon!
Let's make sure everything is there and mounted correctly once we've restored everything:
jeff@apollo:~$ df -h
Filesystem Size Used Avail Use% Mounted on
udev 7.8G 0 7.8G 0% /dev
tmpfs 1.6G 1.2M 1.6G 1% /run
/dev/sda2 457G 5.9G 428G 2% /
tmpfs 7.8G 0 7.8G 0% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 7.8G 0 7.8G 0% /sys/fs/cgroup
/dev/loop0 91M 91M 0 100% /snap/core/6350
/dev/sda1 511M 6.1M 505M 2% /boot/efi
tmpfs 1.6G 0 1.6G 0% /run/user/1000
/dev/sdb1 2.7T 233G 2.4T 9% /mnt/storage
192.168.1.181:/volume1/movies 21T 9.3T 12T 45% /mnt/movies
192.168.1.181:/volume1/series 21T 9.3T 12T 45% /mnt/series
192.168.1.181:/volume1/dropbox 21T 9.3T 12T 45% /mnt/dropbox
192.168.1.181:/volume1/music 21T 9.3T 12T 45% /mnt/music```Awesome, looks good!
Plex was the first thing to be re-installed. This happened with a .deb since I can't seem to find a reliable Docker image, plus I get Plex Pass downloads so figured I'd install it straight onto the host.
Plex downtime lasted 41 minutes, and was fully restored. Phew! SLA met :D
Then the fun started, time to get Docker up and running with some containers. This was quite a learning curve. Volumes, port mappings, aliasing etc still really baffles me.
I managed to get the basics up and running with a few needed tweaks, some of the documentation on these images confused the shit out of me - so I'll probably write a blog on how I got some of these up and running.
Here's my current container list, all of these are fully functional, albeit with a few permission issues writing to NFS shares.
jeff@apollo:~$ docker ps -a
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
0881a1496a6c influxdb "/entrypoint.sh infl…" 25 hours ago Up 25 hours 0.0.0.0:8086->8086/tcp influxdb
86d4fb931609 grafana/grafana "/run.sh" 35 hours ago Up 34 hours 0.0.0.0:3000->3000/tcp grafana
d818dd4f5977 linuxserver/couchpotato "/init" 47 hours ago Up 47 hours couchpotato
cfac2b7f71e6 linuxserver/sonarr "/init" 47 hours ago Up 47 hours sonarr
427772346c72 linuxserver/deluge "/init" 2 days ago Up 47 hours deluge
10ce4a90247a linuxserver/sabnzbd "/init" 2 days ago Up 24 hours sabnzbd```That's it for now. Part 3 should include a little more Docker knowledge along with a PSU reinstall and drives mounted properly.
Until next time.