|
Caged posted:I don't see a need for a spinning disk in a desktop / laptop at all. For home use you have stuff like the WD Mybook or a NAS if you're a bit nerdier, and enterprise hasn't been storing anything locally for a while (ideally). Spinning disks will be around as long as they are cheaper per GB than flash with acceptable performance, but hidden away in a datacentre. Same here, all my OS disks are SSDs because the speed is just awesome and anything I need to store goes on the NAS running ZFS. With Gbit ethernet and some fast drive 100MB/sec is plenty fast.
|
![]() |
|
![]()
|
# ? Jun 10, 2024 21:03 |
|
D. Ebdrup posted:The short answer is, who the gently caress knows? Anyone claiming to know, are most probably pulling bullshit out of their asses. Personally, I prefer SSDs for OS and using a server with platter disks for all other storage. The one thing that this doesn't cover is the fact that your cost/GB on hard drives doesn't hold up down to the sizes of smaller SSDs; you can get 120GB SSDs on sale for around $70-$80 sometimes, but you'll pay almost that much for a 120GB HDD if you can find one new still. And for a lot of people, 100-200GB or so of usable storage is enough to handle their OS and applications and even their personal stuff aside from large amounts of locally stored video / photos / audio. We're getting pretty close to the point where a big-enough SSD is as cheap as the smallest, cheapest platter drives.
|
![]() |
|
Platter density has stalled for a bit. That'll likely change in the next year or so.
|
![]() |
|
I've also read somewhere that there could be atomic limits to how big SSDs can go, but I'm sure that's been said about every computer technology so it probably doesn't mean much.
|
![]() |
|
FISHMANPET posted:I've also read somewhere that there could be atomic limits to how big SSDs can go, but I'm sure that's been said about every computer technology so it probably doesn't mean much. Theres lots of room for more chips though in SSD's. Even if they reached a limit to how much space they can fit in a square inch they have lots of square inches to work with. This is 1tb ![]()
|
![]() |
|
The mass platter-based storage consumers of the next 15 years are probably going to be businesses moreso than end users partly because everyone seems totally cool with the whole cloud computing thing. While storage for home 4k video and DVRing may be present, I don't foresee that being a big deal exactly either for most users because the content providers and cable companies will donate all their profits to the NSA before they actually give everyone 4k video let alone 1080P. So this leaves "power users" like video editors and engineers, and I don't foresee that market eagerly letting go of their local mass storage.UndyingShadow posted:As a hardware geek, all that's super exciting, but at the same time kinda frustrating, because it will always be cheaper to throw a bunch of hard drives and an off the shelf mobo in a desktop case, and get a great FreeNAS setup. As much as I keep backups, I despise the thought of having to re-rip CDs & movies and having to re-organize everything into my media software all over again. I literally spent 3 years ripping CDs nearly every day, I'm not going to let my data die without a healthy dose of paranoia. D. Ebdrup posted:PassMark has a complete set of CPU benchmarks, consisting of integers crunching, compression tests, encryption tests, and more that you can read about here, which produce a completely arbitrary but mostly-reliable number which gives an ide of relative cpu preformance. ![]()
|
![]() |
|
ZFS question: I created a zpool on Freenas and then decided I want to try Ubuntu Server instead. The zpool exported OK, but it will only let me import it as read only because the zpool ZFS version is higher than the linux ZFS version. Can anyone walk me through the steps necessary to create a new zpool from the data using only the linux system and the 2 disks the zpool is currently on? The zpool is a mirrored pair of disks and I don't need to change that. I want everything to be the same, just writable.
|
![]() |
|
You'll have to create a new pool with one of the disks, copy the data over, then add the old disk into the new mirror. There's no way to downgrade ZFS versions.
|
![]() |
|
Thanks for replying! Care to baby a noob through the zfs commands to do that thing you said?
|
![]() |
|
What I would do first is unplug one of the drives so you don't accidentally nuke both drives Then you'd create the new pool with just the single drive: zfs create pool devicename1 (you may need to use -f to force this, since it will detect an existing pool on the device) Reboot and plug in the other drive. Mount your other pool which will be read only and degraded, but since it's a mirror that's fine. Then copy the data over from the old read only pool to the new pool. Attach the second drive to make a mirror zpool attach pool devicename1 devicename2 What you're doing here is attaching the second disk to the first one to make a mirror. Make sure that the existing device is the first paramter, and the new drive you're attaching to the mirror is the second parameter. You'll probably also have to force that with -f since it will detect an existing pool. I've only done it on Solaris but I assume the commands aren't all backwards in Linux or anything like that.
|
![]() |
|
FISHMANPET Thank you for the lesson! Now it's time to go wait for the long, long copy. Stupid moving parts...
|
![]() |
|
I currently have a SmartOS machine with 1x OS (ie, 'zones') platter drive and a 8x2TB RAIDZ2 data array. I'm considering an SSD for the OS drive, but would like to also have a small partition as a ZIL on that same drive for the data array to improve synchronous write performance. Is this a bad idea?
|
![]() |
|
FISHMANPET posted:What I would do first is unplug one of the drives so you don't accidentally nuke both drives I'm newish to ZFS. When he's ready to copy from one pool to the next, would a 'zfs send' be appropriate to use?
|
![]() |
|
Uuuuum, it wouldn't be inappropriate to use. It's probably a good idea, as you'll have a lesser chance of corrupting any data during the copy (though rsyncing would eliminate that chance as well). In a case like this where I doubt there are any snapshots or advance file system layouts, it's probably a wash. ZFS send really shines when you've got snapshots and lots of filesystem properties to shuffle around.
|
![]() |
|
FISHMANPET posted:Uuuuum, it wouldn't be inappropriate to use. It's probably a good idea, as you'll have a lesser chance of corrupting any data during the copy (though rsyncing would eliminate that chance as well). In a case like this where I doubt there are any snapshots or advance file system layouts, it's probably a wash. ZFS send really shines when you've got snapshots and lots of filesystem properties to shuffle around. Good to know. I'm just diving into ZFS and still learning. Thanks!
|
![]() |
|
Hi XPEnology fans. I have used UNRAID for years. Love(loved) it, but the lack of development kills me, mainly the performance on it sucks and there isnt a dual parity option. I currently run a 6 disk Array including 1 parity drive. I picked unraid because it let me swap in larger disks to replace smaller ones and it also lets me expand out the array just by adding another disc. Nothing on the market really could do that at the time. My issue is that even though my mobo and case supports ~15 discs, I am scared to go past 6 because of only one parity drive and I need to shell out like 80 bucks to upgrade my "key". hmmph. If I rolled an XPEnology server would I still be able to increase my pool size by adding discs or upgrading discs and ALSO be able to run dual parity? Looks like yes, but I'd rather find out before I go on a disc buying spree ha. Does XPEnology have a good plugin system? I run a suite of apps such as sabnzbd and sickbeard and I hope its a painless process to get those going I hope? Is it pretty much identical to whatever stock synology boxes can handle? Thanks all!
|
![]() |
|
Minty Swagger posted:Hi XPEnology fans. Yes you can run 2 disc redundancy (shr2) and yes there are lots of apps.
|
![]() |
|
Yeah I put it on myself to actually read about it and it sounds like what UNRAID would have grown into if the sole developer had more incentive to work on it. Sounds like my next challenge is to find out if my hardware supports it. If I run XPEnology out of the box can I install lets say a windows 7 virtual machine on it and cool stuff like that, or does XPEnology need to be installed as a virtual machine inside of some other OS to do that?
|
![]() |
|
If you like unraid you could also consider snapraid. Same sort of idea, with your disks of data and dedicated parity drives, but with (very) active development, block checksums, scrubbing, etc. It's snapshot raid, so you do your parity calculations when you want to, which some people like (save it until the end of the day and run it at night, for example). It's also free, and you can use as many parity drives as you want. Command line or (old and outdated) program though, but it's very easy. http://snapraid.sourceforge.net/ As for xpenology: it works best either on it's own, or running as a virtual machine guest where you pass the drives through to said guest. Works alright in ESXi, not so much in HyperV
|
![]() |
|
Hm, so I could for example boot windows off an SSD and then run snapraid just on top of everything. Do you just set everything up as JBOD and then it layers on top? Any downsides rolling an XPEnology build? They both sound great!
|
![]() |
|
Minty Swagger posted:Hm, so I could for example boot windows off an SSD and then run snapraid just on top of everything. Do you just set everything up as JBOD and then it layers on top? Any downsides rolling an XPEnology build? They both sound great! That's one way of doing it. It also work with linux and freebsd and osx and everything. You don't set it up as a JBOD though, you leave your drives as-is, then use the snapraid configuration to set drives as data or parity. You can keep your existing data on those drives, too. Only downsides to xpenology are hardware support, which isn't an issue if you're buying new hardware or use supported hardware, and being limited to whatever packages in the OS you want; for most people (and for a dedicated server) it isn't a problem, since you probably won't need to use it as a dns server or a domain controller, but it's something to look at. It's also why people run it as a virtual machine guest under esxi, so they can make more virtual machines for those other tasks.
|
![]() |
|
I bought a cheap ($20) D-Link DNS-323 on eBay because someone in this thread said that they're pretty neat. What cool stuff can I do with it when I get it? I'm a tinkerer, it's a bad habit.
|
![]() |
|
QPZIL posted:I bought a cheap ($20) D-Link DNS-323 on eBay because someone in this thread said that they're pretty neat. I owned one. The DLNA server is cool, but it's not powerful enough to transcode, so you'll need to make sure the files are in a format natively supported by your player. It has an itunes server, but it never worked properly for me. My library might have been too large, dunno. Open up FTP services and port it through your router on a strange port for remote access. I guess it might support bittorrent? I never tried it. It's a slow box, though. It brags gigabit, but I never broke 8MB/s or so, connected to a gig router. My N40L used the same disks and hit 80MB/s. Don't expect miracles.
|
![]() |
|
QPZIL posted:I bought a cheap ($20) D-Link DNS-323 on eBay because someone in this thread said that they're pretty neat. Fixed, and me too. Don't feel bad; all the fun and interesting hobbies are expensive and time-consuming. ![]() Here's an entire wiki dedicated to cool stuff to do with the DNS-323, and here's a Debian installation guide if you're so inclined. Embedded Linux machines are great for tinkering.
|
![]() |
|
Wild EEPROM posted:That's one way of doing it. It also work with linux and freebsd and osx and everything. You don't set it up as a JBOD though, you leave your drives as-is, then use the snapraid configuration to set drives as data or parity. You can keep your existing data on those drives, too. I think my biggest push away from snapraid is that its still a lot of disks with parity so you still have to manage your disk space per drive still unless you utilize another app like stablebit drivepool or something. That seems like bad news layering on so many different types of tech though in my opinion!
|
![]() |
|
Civil posted:It's a slow box, though. It brags gigabit, but I never broke 8MB/s or so, connected to a gig router. My N40L used the same disks and hit 80MB/s. Don't expect miracles. No worries, I'm not expecting miracles. But I have no backup solution right now apart from a couple 500gb USB3 drives, so for $20 I really can't lose here. SamDabbers posted:Here's an entire wiki dedicated to cool stuff to do with the DNS-323, and here's a Debian installation guide if you're so inclined. Embedded Linux machines are great for tinkering. Cool ![]()
|
![]() |
|
IOwnCalculus posted:That's the great thing about standards, there's so many to choose from! I was under the impression that at least some HW RAID companies had broad compatibility across their own products - so a RAID created on one card would be accessible to any same-brand card of that generation or newer. Farmer Crack-Ass fucked around with this message at 19:33 on Jan 20, 2014 |
![]() |
|
Minty Swagger posted:I think my biggest push away from snapraid is that its still a lot of disks with parity so you still have to manage your disk space per drive still unless you utilize another app like stablebit drivepool or something. That seems like bad news layering on so many different types of tech though in my opinion! There's a new feature in new versions which include disk pooling. Not sure how it works since I haven't used it.
|
![]() |
|
Minty Swagger posted:Does XPEnology have a good plugin system? I run a suite of apps such as sabnzbd and sickbeard and I hope its a painless process to get those going I hope? Is it pretty much identical to whatever stock synology boxes can handle? Minty Swagger posted:If I run XPEnology out of the box can I install lets say a windows 7 virtual machine on it and cool stuff like that, or does XPEnology need to be installed as a virtual machine inside of some other OS to do that?
|
![]() |
|
I have this problem with my Synology Diskstation where even after disabling facial recognition in PhotoStation the facerecognition process keeps eating up cpu cycles. Same as described here and here. In those threads an apparently pretty simple solution is suggested, but I'm some sort of idiot that can't make sense of it. This is what I tried: 1. enabled SSH and Telnet service using the UI. 2. Stop PhotoStation 2. using Putty, log in to the diskstation using my admin account 3. enter the stuff for changing directories and renaming the tempfile What happens then is that I get a message that renaming failed because I don't have permission to do that. So what did I miss? A second thing is that the procedure calls for rebooting the Diskstation and I'm ashamed to admit I have no idea how to do that, apart from maybe pressing the on/off button or cutting the power. Although I guess that's something I can probably google. Any help immensly appreciated. (Talk to me like I know nothing, because I really don't)
|
![]() |
|
Flipperwaldt posted:2. using Putty, log in to the diskstation using my admin account
|
![]() |
|
Reboot is on the fake Start menu thing in DSM
|
![]() |
|
FISHMANPET posted:Uuuuum, it wouldn't be inappropriate to use. It's probably a good idea, as you'll have a lesser chance of corrupting any data during the copy (though rsyncing would eliminate that chance as well). In a case like this where I doubt there are any snapshots or advance file system layouts, it's probably a wash. ZFS send really shines when you've got snapshots and lots of filesystem properties to shuffle around. I used 'cp -npRv' (I'm only copying media/photos). Your directions worked like a charm! Resilvering now. ![]()
|
![]() |
|
Sub Rosa posted:Tried logging in as root instead? Same password as admin. Caged posted:Reboot is on the fake Start menu thing in DSM EDIT all this worked and I'm happy as a clam! Flipperwaldt fucked around with this message at 02:35 on Jan 18, 2014 |
![]() |
|
The guy with all the XPenology questions should just download it and install in an vm to putz with. That's what I've been doing, works fine in vmware workstation. Anyone else read this article a few days ago on Ars? http://arstechnica.com/information-technology/2014/01/bitrot-and-atomic-cows-inside-next-gen-filesystems/
|
![]() |
|
This is half NAS related and half home networking related, but I'm curious as to what configuration and performance people get from their NAS or file servers on Mac OS X (if there are many at all who have some a setup). I use ZFS on Linux on Ubuntu Server 12.04 LTS on my file server (currently just my last desktop with a Phenom II X3 720 and 4GB of DDR2 repurposed, with 6 2TB drives in 2 RAID-Z vdevs - it's due for an update and will be replaced by a Haswell Xeon E3 build with ECC soon), and the one aspect that I've never been satisfied with is file sharing with Macs. Until OS X 10.9 Mavericks, I used NFS (because SMB performance was beyond terrible), which after way too much time spent on researching, testing, and "tuning" I was able to use async mode and larger buffer size to get OK performance (peak transfer rates of ~80MB/s over GbE). This was still not great (by comparison a Windows client would be able to reach 120MB/s via SMB), but it was usable enough, even if performance wasn't terribly consistent, and there were other issues (could never get user mapping to work, so security is limited to simply only allowing connections from my subnet, and sometimes file renames would just fail, or some files wouldn't appear in Finder). After upgrading to OS X 10.9 however, I discovered that Mavericks pretty much broke NFS auto-mounting. It did however also get upgraded Samba file sharing, and after a lot more time spent on researching, testing, and tuning (force SMB2, force TCP_NODELAY, set specific send/receive buffer), I'm able to get usable performance out of SMB (peak transfer rates of ~70MB/s), and that's what I'm currently using. It's far from ideal though, as in addition to the subpar performance (compared to SMB via Windows client), there's also always a weird delay when copying files using Finder - as far as I can tell, the transfer progress would always be stuck at 0 for the first hundred or so MB of data, and then suddenly catch up. This only seems to happen on transfers between the Mac and the file server - transfers between the Mac and a Windows file share works fine. And this happens with both my Hackintosh (i3 3225/GA-H77N using onboard Realtek GbE) and my 2012 rMBP (using the TB-GbE adapter), so doesn't seem to be a network hardware specific thing on the Mac client side. I have not tried AFP, mostly because I hate the idea of having to set up yet another protocol for just one OS (even though that's what NFS ended up being in my case, at least it would also be usable if I added another *nix box), not to mention it's really basically a dead protocol at this point - Time Machine backups are literally the only reason to use AFP now, as even Apple is pushing SMB(2) as of 10.9. But if I can't figure out how to get SMB to work better, I might just have to when I rebuild my file server. iSCSI is also not an option, as I need the shares to be available to multiple clients (and I don't want to pay for an iSCSI initiator that somehow Apple still hasn't added into OS X). Anyway, TL;DR: What do people use with Macs, and what kind of performance do you get out of them?
|
![]() |
|
I use SMB2 over 802.11n and get ~75mbit to a FreeBSD/ZFS server. Haven't noticed any issues renaming files or getting progress bars to work.
|
![]() |
|
GokieKS posted:This is half NAS related and half home networking related, but I'm curious as to what configuration and performance people get from their NAS or file servers on Mac OS X (if there are many at all who have some a setup). Just get netatalk 3.0, ppa is located at https://launchpad.net/~jofko/+archive/ppa It is drop dead simple to setup and configure and I get ~100MB/s with afp. It has the added bonus of being able to emulate a time machine share which I use for backing up my mac.
|
![]() |
|
Does anyone have updates on whether Newegg is still lovely at shipping drives? I know an earlier post in the thread said they use air cushions now that prevent jostling, but I wanted to confirm that this wasn't just some fluke before I buy a new set of backup drives from them.
|
![]() |
|
![]()
|
# ? Jun 10, 2024 21:03 |
|
I got some WD Reds from them recently that came in the air pocket things, no complaints.
|
![]() |