February 23, 201115 yr ...so this is what I do (for no other reason than "because I can:)... Feb 23 11:26:43 unRAID emhttp: Spinning down all drives... Feb 23 11:26:43 unRAID kernel: mdcmd (49): spindown 0 Feb 23 11:26:44 unRAID kernel: mdcmd (50): spindown 1 Feb 23 11:26:44 unRAID kernel: mdcmd (51): spindown 2 Feb 23 11:26:45 unRAID kernel: mdcmd (52): spindown 3 Feb 23 11:26:45 unRAID kernel: mdcmd (53): spindown 4 Feb 23 11:26:46 unRAID kernel: mdcmd (54): spindown 5 Feb 23 11:26:46 unRAID kernel: mdcmd (55): spindown 6 Feb 23 11:26:46 unRAID kernel: mdcmd (56): spindown 7 Feb 23 11:26:46 unRAID kernel: mdcmd (57): spindown 8 Feb 23 11:26:46 unRAID kernel: mdcmd (58): spindown 9 Feb 23 11:26:46 unRAID kernel: mdcmd (59): spindown 10 Feb 23 11:26:46 unRAID kernel: mdcmd (60): spindown 11 Feb 23 11:26:46 unRAID kernel: mdcmd (61): spindown 12 Feb 23 11:26:46 unRAID kernel: mdcmd (62): spindown 13 Feb 23 12:17:03 unRAID emhttp: Spinning up all drives... Feb 23 12:17:03 unRAID kernel: mdcmd (63): spinup 0 Feb 23 12:17:03 unRAID kernel: mdcmd (64): spinup 1 Feb 23 12:17:03 unRAID kernel: mdcmd (65): spinup 2 Feb 23 12:17:03 unRAID kernel: mdcmd (66): spinup 3 Feb 23 12:17:03 unRAID kernel: mdcmd (67): spinup 4 Feb 23 12:17:03 unRAID kernel: mdcmd (68): spinup 5 Feb 23 12:17:03 unRAID kernel: mdcmd (69): spinup 6 Feb 23 12:17:03 unRAID kernel: mdcmd (70): spinup 7 Feb 23 12:17:03 unRAID kernel: mdcmd (71): spinup 8 Feb 23 12:17:03 unRAID kernel: mdcmd (72): spinup 9 Feb 23 12:17:03 unRAID kernel: mdcmd (73): spinup 10 Feb 23 12:17:03 unRAID kernel: mdcmd (74): spinup 11 Feb 23 12:17:03 unRAID kernel: mdcmd (75): spinup 12 Feb 23 12:17:03 unRAID kernel: mdcmd (76): spinup 13 Feb 23 12:18:00 unRAID emhttp: Spinning down all drives... Feb 23 12:18:00 unRAID kernel: mdcmd (77): spindown 0 Feb 23 12:18:01 unRAID kernel: mdcmd (78): spindown 1 Feb 23 12:18:01 unRAID kernel: mdcmd (79): spindown 2 Feb 23 12:18:02 unRAID kernel: mdcmd (80): spindown 3 Feb 23 12:18:03 unRAID kernel: mdcmd (81): spindown 4 Feb 23 12:18:03 unRAID kernel: mdcmd (82): spindown 5 Feb 23 12:18:03 unRAID kernel: mdcmd (83): spindown 6 Feb 23 12:18:03 unRAID kernel: mdcmd (84): spindown 7 Feb 23 12:18:03 unRAID kernel: mdcmd (85): spindown 8 Feb 23 12:18:03 unRAID kernel: mdcmd (86): spindown 9 Feb 23 12:18:03 unRAID kernel: mdcmd (87): spindown 10 Feb 23 12:18:03 unRAID kernel: mdcmd (88): spindown 11 Feb 23 12:18:04 unRAID kernel: mdcmd (89): spindown 12 Feb 23 12:18:04 unRAID kernel: mdcmd (90): spindown 13 ;D Anyone else finding themselves doing the same. Or, should I have myself committed? John
February 23, 201115 yr Don't get me wrong, I completely understand being bored at work. In fact, if it weren't for the draconian firewalls at my gov't job I probably never would have gotten into unraid like I have. I've learned so much from just reading these forums. That said, I have to be a downer and rap your knuckles. Spinning up and down drives like you are doing is not good for them. Many drives will actually last longer if you keep them spun up 24/7 instead of spinning them up and down several times per day. I suggest you find a less destructive way to kill time Sent from my DROID2 using Tapatalk
February 23, 201115 yr Author Many drives will actually last longer if you keep them spun up 24/7 instead of spinning them up and down several times per day. Is this true? I have been aching for a reason to never spin my drives down just for response time sake. The problem is I have 8x older Maxtor 250gb drives that just love to get warm...dare I say hot. I have moved everything to the unfinished part of our basement. I think I am going to set spin down to never and see how the temps maintain. John
February 23, 201115 yr That definately applies more to older drives. I haven't heard of any modern drives failing from too many spin ups and spin downs, but it still seems like something to avoid as it is unneccesary mechanical wear and tear on your drives. I figure if you access your drives often enough that they will spin up and down quite often, then it might be appropriate to set them to never spin down or set the spin down timer really high. You'll have to address heat and power issues that may arise. I would definitely recommend green drives to keep both heat and power consumption to a minimum. I take the opposite approach. I like to keep my drives spun down as much as possible to save power. Therefoe I use cache_dirs as well as some conscious effort in my file organization in an effort to keep my drives spun down. I also try to keep my most often used data on my green drives, and use my 7200 rpm drives primarily for archival. Sent from my DROID2 using Tapatalk
February 23, 201115 yr ...so this is what I do (for no other reason than "because I can:)... ... ;D Anyone else finding themselves doing the same. Or, should I have myself committed? John I dunno what to even label this one. UnRAID Compulsive Design was easy.. This one has me stumped... Raj.. any ideas?
February 23, 201115 yr Many drives will actually last longer if you keep them spun up 24/7 instead of spinning them up and down several times per day. Is this true? I have been aching for a reason to never spin my drives down just for response time sake. The problem is I have 8x older Maxtor 250gb drives that just love to get warm...dare I say hot. I have moved everything to the unfinished part of our basement. I think I am going to set spin down to never and see how the temps maintain. John This used to be the case, I think in today's era, it's not the same. It all depends on your ambient temperatures. if you have hot summers (and do not use air conditioning in the computer area) like I do, then spinning them up and down is better then 24x7. My older drives used to last around 2yrs or so, then start dying off. So far I'm on about 3 years with my drives spinning up and down. In contrast my 10,000 RPM SAS drives have been spinning 24x7 for about 3 years now. But that's in a supermicro chassis designed with really good air flow. Plus I moved my always on vmware instances to an SSD which I'm sure diminishes the wear and tear on the RAID 1 SAS drives.
February 23, 201115 yr ...so this is what I do (for no other reason than "because I can:)... ... ;D Anyone else finding themselves doing the same. Or, should I have myself committed? John I dunno what to even label this one. UnRAID Compulsive Design was easy.. This one has me stumped... Raj.. any ideas? UCBP (unRaid Compulsive Button Push)
February 23, 201115 yr Author 8 of my 14 drives are 250gb 7200 rmp Maxtors that I bought in 2006. They have been running pretty much 24/7 ever since I bought them (albeit the few days here and there for upgrades...like unRAID). Honestly, with as hot as I have seen them get I can't believe they have lasted this long! John
February 25, 201115 yr Hmm, reminds me of a bit of Joe L.'s prose: I stopped shaving... I stopped showering... I have no more friends... even the neighbor's dog keeps its distance. I go to bed mumbling about disks, and connectors, and drive trays... The forum posts are the only company I have at night when I can't sleep. I've become a recluse... I pour over sales catalogs and mailings looking for discount-codes. I live for the monthly parity check. I spend my days copying the same files to and from the disks, over and over again... It has been months since I actually watched a movie... but my server runs like a dream! Taken from here.
February 25, 201115 yr lol hopefully I don't end up like that I can already see myself in the future messing with my server more than actually using what is on it haha I also wouldn't have discovered most of this if I was not bored at work. Anyways, I only have one disk ( for now) and I have sickbeard, sab, and couch potato on the server. Should I just leave it always spun up for now? Once I get a cache drive (where Sab etc was intended to be on) what should I do?
February 26, 201115 yr I don't really know enough about those add-ons to tell you conclusively. Do they access the disks more than once every few hours? If so, then it isn't a bad idea to leave them spun up at all times. If not, then I would probably let them spin down. You could leave the spin down timer at the default setting (1 hour) and then just check your syslog after a few days to see how often your disks spin up and down, then evaluate based on that.
February 26, 201115 yr I don't really know enough about those add-ons to tell you conclusively. Do they access the disks more than once every few hours? If so, then it isn't a bad idea to leave them spun up at all times. If not, then I would probably let them spin down. You could leave the spin down timer at the default setting (1 hour) and then just check your syslog after a few days to see how often your disks spin up and down, then evaluate based on that. good idea. I am pretty sure as long as you are downloading from SABnzbd then it is accessing the disk. So hwo does the timer work... 1 hour of inactivity and it spins down? Maybe that wouldn't be a problem then? As soon as everything is finished downloading if an hour goes by it spins down. When disks are spun down once you access something on the disk that automatically makes them spin up?
March 2, 201115 yr Pretty much just like you said. The only exception will be if the file being accessed is still in unRAID's RAM cache, then the disk may not have to spin up at all. Cache_dirs is a script written by Joe L. that constantly accesses your user share's directory structures in an attempt to keep them in the RAM. This means that when you browse a share that spans several disks, the disks should not have to spin up. Only the single disk that contains the file you open should spin up. That's the ideal case. If you don't have enough RAM, then you still may see multiple disks spinning up as you browse a share.
March 2, 201115 yr Pretty much just like you said. The only exception will be if the file being accessed is still in unRAID's RAM cache, then the disk may not have to spin up at all. Cache_dirs is a script written by Joe L. that constantly accesses your user share's directory structures in an attempt to keep them in the RAM. This means that when you browse a share that spans several disks, the disks should not have to spin up. Only the single disk that contains the file you open should spin up. That's the ideal case. If you don't have enough RAM, then you still may see multiple disks spinning up as you browse a share. is this only when using a cache drive? How much ram is "enough"
March 3, 201115 yr No, a cache drive is a different thing. Cache drive or not, unRAID will use all of the available RAM to buffer transfers and speed things up. How much RAM you need depends on how much and what kind of data you are transferring. 1 GB is pretty much the minimum, though some people still get away with 512 MB. 2 GB is the new standard. Anything above that will help, but may be overkill.
March 3, 201115 yr No, a cache drive is a different thing. Cache drive or not, unRAID will use all of the available RAM to buffer transfers and speed things up. How much RAM you need depends on how much and what kind of data you are transferring. 1 GB is pretty much the minimum, though some people still get away with 512 MB. 2 GB is the new standard. Anything above that will help, but may be overkill. is the script installed by defualt or is there instructions to install
March 3, 201115 yr Not installed by default. Here's one fan control script. Search 'fan control' and you'll find others.
March 3, 201115 yr Not installed by default. Here's one fan control script. Search 'fan control' and you'll find others. ok you lost me... how does that fan script relate to Cache_dirs?
March 3, 201115 yr It doesn't. Both are just different methods of attempting to make your server as low power as possible.
Archived
This topic is now archived and is closed to further replies.