questions about Synology SHR ....

[zero7404]

And yes, synology allows you to 'grow' existing arrays - so you could put the current 2 * 16tb in a 4-bay as a raid1 array, then later add another 16tb to go to raid5 ( taking it from 16tb usable to 32tb usable ) and then add a 4th to either go bigger again ( 48tb ) or increase redundancy by going from raid5 to raid6 ( although raid6 on a 4 drive array doesn't make much sense - might as well just run raid10 as it'll yield the same space and have better write performance )

this is one compelling reason to head in the synology direction. i think it was asked within the thread but I did not address: do I really need boat-loads of storage ? -- answer: no, absolutely not. at least not in the context you might see based on your experience.

i'm a small-time home user, with a family. and maybe 2 or 3 users at home at most are queuing the NAS for something, mostly media content. MOST times, it's sitting idle and NEVER will see anywhere near the 24/7 use of a data center. that's not to say that the NAS isn't putting the drives to work at all when no one is accessing data, but most times it's crickets in the NAS.

if i threw 2x 16TB + 2x 8TB + 2x 6TB into a larger NAS, and came away with a single 16TB volume, I'm happy with that, knowing I can expand later on and not have to fiddle with drive swaps or waiting days to copy all my data from one enclosure to another when upgrading. as it stands right now, the WD EX2 Ultra is so barebones, it does not offer any kind of upgrade subroutines for drive swaps. you simply backup your junk, put new drives in the NAS, and restore your junk back to it from the backup. if we are talking 16TB worth of junk over usb3.0 connection .... it amounts to the rysnc job taking about 4 days to complete.

imo the synology would be an upgrade by at least a few clicks

 

[John Davis]

this is one compelling reason to head in the synology direction. i think it was asked within the thread but I did not address: do I really need boat-loads of storage ? -- answer: no, absolutely not. at least not in the context you might see based on your experience.

i'm a small-time home user, with a family. and maybe 2 or 3 users at home at most are queuing the NAS for something, mostly media content. MOST times, it's sitting idle and NEVER will see anywhere near the 24/7 use of a data center. that's not to say that the NAS isn't putting the drives to work at all when no one is accessing data, but most times it's crickets in the NAS.

if i threw 2x 16TB + 2x 8TB + 2x 6TB into a larger NAS, and came away with a single 16TB volume, I'm happy with that, knowing I can expand later on and not have to fiddle with drive swaps or waiting days to copy all my data from one enclosure to another when upgrading. as it stands right now, the WD EX2 Ultra is so barebones, it does not offer any kind of upgrade subroutines for drive swaps. you simply backup your junk, put new drives in the NAS, and restore your junk back to it from the backup. if we are talking 16TB worth of junk over usb3.0 connection .... it amounts to the rysnc job taking about 4 days to complete.

imo the synology would be an upgrade by at least a few clicks

the ability to grow arrays isn't unique to Synology - it's part of the linux md driver - but the ability to do so whilst mixing drive sizes IS synology only ( in a user friendly form at least, you can do it from the commandline on most linux distros as well, it's not the underlying technology that is synology specific, it's the UI to make it easy to do so that is unique to them)

Just remember - those N300 drives are NAS specific units - so if you do get a synology make sure you change the settings on the NAS to NOT spin down the disks when idle (default is to spin down after a certain period idle) - enterprise/NAS drives are rated for 24/7 use, and spinning them up/down is actually a good way to shorten their lifespan

 

[zero7404]

I went thru the Seagate 3TB mess back when I was managing core network hosts at telecom provider (modest size for the US, only 7M subs) - I had a 30 percent failure rate on those drives.

Spinning rust does eventually die - and since one of the key things about a NAS is to centrally host most of the files one would like to keep around...

Backups are handy - and something to consider with any NAS, as RAID is not a backup...

regarding Seagate spinning platters, i recently had one 2.5" form-factor Seagate fail on me, it came out of an Xbox One. could not run tests on it because it was in bad enough shape where CyrstalDisk Info could not load it up.
i also have an older 1.5TB (2x750GB) external usb backup drive (2.5 form factor as well), which has been around for some years now and is still alive and kicking.

other than that, not much experience with Seagate. I'm sure there are many crowds that will swear by Seagate, or WD, Toshiba, HGST, etc.... until they succeed in making you believe that pigs fly.

I've had WD drives kick dirt in my face (3 failed drives -- all of them SMR's), but did get good advice about Toshiba brand HDD's from within a WD forum, of all places .... ironically.

i used to run (for a long time) with a single external backup of my NAS, and both these devices were WD's affordable options .... all the while not realizing that these units contained drives that could take a crap on me which they eventually did in time.

that helped me get to where i am now: raid1 backup volume physically separated from the raid0 NAS. still in risky territory with raid0 .... but at least i've gone from 1 duplicate of my data to 2 duplicates. had i known more about advance raid offerings by better NAS vendors, I would not have even considered WD again.

to make things even worse from a customer's perspective - their forum has 'bots' masquerading as 'WDStaff', one of them answered a post of mine with a solution that has nothing to do with my problem.

 

[zero7404]

the ability to grow arrays isn't unique to Synology - it's part of the linux md driver - but the ability to do so whilst mixing drive sizes IS synology only ( in a user friendly form at least, you can do it from the commandline on most linux distros as well, it's not the underlying technology that is synology specific, it's the UI to make it easy to do so that is unique to them)

Just remember - those N300 drives are NAS specific units - so if you do get a synology make sure you change the settings on the NAS to NOT spin down the disks when idle (default is to spin down after a certain period idle) - enterprise/NAS drives are rated for 24/7 use, and spinning them up/down is actually a good way to shorten their lifespan

just turned off the drive 'sleep' function of my NAS. thanks for that advice .... there are N300's in the NAS, but the external backup is MG08's - that enclosure doesn't park the heads on the drives, so i think it keeps platters spinning. i do only power-cycle it when i need it. perhaps i should always leave it on ? it is an OWC Mercury Elite Pro Dual.

interesting to know that can be done via command line ... even if i consider myself competent enough to handle command line work from time to time, i would not venture to managing/creating raid arrays .... as we all know, command line can be somewhat dangerous to the life of data if some argument or switch mistakenly added in the command is executed ....

if these NAS offerings where stratospheric in price, i would then button down and make damn sure i know the linux commands to use with a 'lesser' NAS .... but the prices of QNAP and Synology are within reach.

 

[sfx2000]

the ability to grow arrays isn't unique to Synology - it's part of the linux md driver - but the ability to do so whilst mixing drive sizes IS synology only ( in a user friendly form at least, you can do it from the commandline on most linux distros as well, it's not the underlying technology that is synology specific, it's the UI to make it easy to do so that is unique to them)

It's more along the line of LVM (Logical Volume Manager) that lets you grow the array.

IIRC - Syno also does BTRFS, where, similar to ZFS, MD/LVM are not used... rather BTRFS handles the MD and LVM functionality within it's own stack...

 

[John Davis]

It's more along the line of LVM (Logical Volume Manager) that lets you grow the array.

IIRC - Syno also does BTRFS, where, similar to ZFS, MD/LVM are not used... rather BTRFS handles the MD and LVM functionality within it's own stack...

it uses md to change raid levels however

also whilst btrfs supports filesystem level raid, synology doesn't use it ( you can see that examining things from the shell using mdadm - even with btrfs the raid is still handled by md )

 

[zero7404]

It's more along the line of LVM (Logical Volume Manager) that lets you grow the array.

IIRC - Syno also does BTRFS, where, similar to ZFS, MD/LVM are not used... rather BTRFS handles the MD and LVM functionality within it's own stack...

would a synology nas lock you into btrfs if you chose shr or shr2 ?

i think it also offers ext4 fs onboard

 

[sfx2000]

it uses md to change raid levels however

also whilst btrfs supports filesystem level raid, synology doesn't use it ( you can see that examining things from the shell using mdadm - even with btrfs the raid is still handled by md )

mdadm does manage the raidset across the available disks, and what type of raid.

LVM allows one to combine raidsets, and it also defines the actual size of the array.

 

[John Davis]

would a synology nas lock you into btrfs if you chose shr or shr2 ?

i think it also offers ext4 fs onboard

nope - a lot of their smaller units with limited memory still use ext4, and even on the big units you can still opt for ext4 rather than btrfs if you so wish

 

[zero7404]

nope - a lot of their smaller units with limited memory still use ext4, and even on the big units you can still opt for ext4 rather than btrfs if you so wish

thanks John,

my thinking is that an ext4 bare drive can be easier to work with in case of a recovery along the lines of pulling the disks from a dead unit.

unless the data integrity of btrfs is that much better than ext4 ?