Time to Burn the Whole System Down

[Ruff.Hi]

I built my first server a few times back in late 2016, early 2017 (is that built ... or rebuilt?). Since then it has basically been my local network storage device and plex server.

Now ... I am pretty sure that I have thoroughly borked my permissions, shares, etc, etc. Plex won't run, creating a new jail says something about permissions being too lax. I have a directory that I can't delete. This is definitely by doing stuff that I shouldn't have done. I also want to move away from RaidZ3 to RaidZ2.

I could nuke it from space (only way to be sure) and start again ... but that would sacrifice my data ... so - time to stand up a new server, copy the data over and then start again.

I imagine that others have done this ... any guide to doing this?

 

[jgreco]

Yeah. You don't necessarily need to sacrifice your data as long as you have access to someplace to store it temporarily. Copy your data somewhere else (this may be the hard part), make sure you copied it correctly, check again that it's all there using a DIFFERENT method. If you can swing doing a new server, though, that's a good option, because then when you're done you can nuke the old server and replicate your new server's data to it.

Since you want to lose any "bad" permissions and other stuff along the way, you probably want to use rsync if possible to copy the data.

I don't have a guide, sorry. Just these few comments.

 

[Ruff.Hi]

Here is where I am starting from ...

Part List

• CPU: Intel Core i3-6100 3.7 GHz Dual-Core Processor
• CPU Cooler: CRYORIG H5 Universal 65 CFM CPU Cooler
  I already have this. Along with about 3 or 4 other air CPU coolers. I might not end up with the H5 as it is probably overkill
  .
• Motherboard: Supermicro MBD-X11SSM-F-O Micro ATX LGA1151 Motherboard
  Similar to my current Supermicro X10SL7-F
  .
• Memory: 2 x Crucial 16 GB (1 x 16 GB) DDR4-2666 Memory
• Storage: 2 x Kingston A400 120 GB 2.5" Solid State Drive
  Mirrored Boot drives
  .
• Storage: 4 x Western Digital Red 4 TB 3.5" 5400RPM Internal Hard Drive
• Case: Corsair 550D ATX Mid Tower Case
  This use to be my PC ... 4 5.25" bays in the front ... just right for a hot swap cage
  .
• Power Supply: SeaSonic FOCUS Plus Gold 550 W 80+ Gold Certified Fully Modular ATX Power Supply
• Custom: ICY DOCK MB324SP-B ExpressCage 4x2.5" SAS/SATA HDD Hot Swap Mobile
  This is for 4 x SSDs ... probably overkill but I really like the cage setup
  .
• Custom: Kingwin HDD/SSHD Hot Swap Hard Drive Enclosure, Internal Four Bay Mobile Rack
  

 

[Ruff.Hi]

Yeah. You don't necessarily need to sacrifice your data as long as you have access to someplace to store it temporarily. Copy your data somewhere else (this may be the hard part), make sure you copied it correctly, check again that it's all there using a DIFFERENT method. If you can swing doing a new server, though, that's a good option, because then when you're done you can nuke the old server and replicate your new server's data to it.

Since you want to lose any "bad" permissions and other stuff along the way, you probably want to use rsync if possible to copy the data.

I don't have a guide, sorry. Just these few comments.

Good point ... copy the file, not the permissions. I typically use BeyondCompare to move this type of stuff. I will google copying (and checking the copy) files.

 

[Ruff.Hi]

I have a (sort of) server up using 2 x 4TB mirrored drives that have failed Long SMART Tests. These got replaced in my production server and I am using them to 'play'. Anyway ... PuTTy was refusing connection. Google threw up this option ...

A PuTTY user manual that appears on the Quest Software website offers a simple explanation for the "Network Error: Connection refused" issue that PuTTY users may encounter from time to time. According to the manual, this error message appears when the application attempts to connect to a server that does not offer the service PuTTY is trying to access. ​

Hmm ... trying to connect to a server that doesn't offer the service? Better check if I have SSH running.

Duh - I didn't. Let me try that again. Well ... further than previously but it will not recognize my root password. Hmm.

Edit: Turned SSH on ... but didn't configure it.

CHECK MARK! needed against ... Log in as root with password

 

[Ruff.Hi]

I have a (sort of) server up using 2 x 4TB mirrored drives that have failed Long SMART Tests. These got replaced in my production server and I am using them to 'play'.

Did I say the HDDs I am using are rubbish?

New alerts:
* Device: /dev/ada3, ATA error count increased from 0 to 5

Current alerts:
* Device: /dev/ada3, 2 Currently unreadable (pending) sectors
* Device: /dev/ada3, ATA error count increased from 0 to 5
* Pool DBPoolTest state is ONLINE: One or more devices has experienced an error resulting in data corruption. Applications may be affected.