How do you partition your linux installation?
200 MB boot
25 GB root
435 GB home
4 GB swap
post your configurations
Depends: hdd or ssd ?
SSD: I would recommend using some ramdisk (tmpfs) for /tmp and /var (/var/log <- depends if you need the log files after reboot.)
I would also avoid swap on an SSD. Just buy plenty of ram.
This is not yet the case, I want to try it someday though:
One btrfs partition, /OS /boot /home subvolumes, "OS" mounted on /.
One swap partition equal in size to physical ram.
This way I wouldn't have to worry about running out of space on separate volumes.
Why? What happens when you use your 4gb of ram? When i overflow my 2gb of ram, my 8gb swap gives me a nice buffer to shut a couple programs without straught up crashing my computer
>4gb of swap
>tiny root partition
>huge home partition
>4 fucking gb's of swap
>ever needing a separate home partition
>4 godfuckingdamned gb's of swap fucking hell
Separate home partition is only useful if you need it encrypted. And who in their right mind needs more than 512mb of swap, even on a goddamn netbook.
Also who ZFS here?
>tfw delicious zfs goodness
Shame the only way to access it through Windows is through a convoluted VM setup or through another machine
Although ZFS without ECC is asking for trouble
ZFS's corruption prevention needs ECC; not using ECC is worse than not using ZFS
> TL;DR version of the scenario: ZFS is on a system with non-ECC RAM that has a stuck bit, its user initiates a scrub, and as a result of in-memory corruption good blocks fail checksum tests and are overwritten with corrupt data, thus instantly murdering an entire pool. As far as I can tell, this idea originates with a very prolific user on the FreeNAS forums named Cyberjock, and he lays it out in this thread here. It’s a scary idea – what if the very thing that’s supposed to keep your system safe kills it? A scrub gone mad! Nooooooo!
If you read that same article a little further you'd notice he debunks the idea completely. It only corrupts the data of a block in a rare specific scenario, and to have a scrub-of-death requires it to consistently happen for all blocks. Which is simply not how ram works.
The most that will happen is zfs logging an error about an unreadable block because it cannot verify checksums. A lot of unreadable, but not corrupted blocks. Fix the ram and you're good to go again.
128 MB boot
rest root because I'm not fucking autistic and my machine has 8 GB of RAM and runs fucking Linux so why the fuck would I compile swap support into my kernel in the first place?