Parity is used by unRAID to protect against data loss. If a drive in the array fails, the data on the other drives can be combined with the parity data to reconstruct the missing data. In general, a parity process is designed to detect a single bit change across a given set of bits, by setting the value of an added bit such that a summation across that set of bits is forced to a known value.
The added bit is known as a parity bit. In unRAID, the parity bits are stored on a parity drive independent of the data drives. This parity bit works across the set of bits in the same relative bit position on each drive. So the 57th bit of the parity drive is the parity bit for the 57th bit of all of the data drives.
A parity check of the 57th bit position therefore adds up all of the 57th bits of every drive including the parity drive, and checking to see total is an EVEN number. Why an EVEN number? When a drive is added to a parity-protected unRAID array, it is first cleared by writing zeroes to all bits of all sectors of that drive. A zero value will not impact parity. Once the new drive is full of zeros, unRAID can silently slip it into the array without needing to rebuild parity.
Several disks can be added simultaneously in this way. At these times, all of the disks including parity are read to reconstruct the data to be written to the target disk.
As the sum of the bits is always even, unRAID can reconstruct any ONE missing piece of data the parity or a data diskas long as the other pieces are correct. As parity is so important for data reconstruction, ensure that parity is correct by running parity checks periodically. If you physically remove a single disk from your array you will still be able to read and write to it. In the same way, the parity disk is updated as needed when writing to the simulated disk.
If you now install a replacement disk, it can be completely rebuilt from the re-constructed contents based on parity and all the other data drives. In a parity syncthe system reads all the data disks and writes the computed parity to the parity disk.
In a parity checkthe system reads all the data disks and the parity disk, comparing computed parity with stored parity. This operation has a flag:. Only the first parity check errors are reported.
The messages in the system log are generated for each sector address where a parity mis-match occurs. So if you tried a parity-check on an array that doesn't have valid parity, the system log would quickly become massive; hence the limit to messages. A 'block' is bytes. Parity disk being valid means that there is a parity disk present, and sometime in the past a parity sync completed without error or being cancelled.
If all array disks are valid except for one of them, it means that the one that's not valid can be reconstructed using the data from all the other ones. Of course, a user can always deem parity invalid by un-assigning it. Once parity has been calculated there should only be 2 ways that there should be parity check errors:. Using parity to protect your data inevitably has an impact on performance when you are writing data to the array.
Any time a data disk is written to, the parity disk needs to be updated as well.By opentoeOctober 22, in General Support. Anyone build a server with new hardware lately, using hard drives and no cache drive? I'm curious as to what top write speed you are getting on a large single file copy. Like an ISO file or something.
Been trying to help a friend remotely and he's telling me on how slow his write speeds are. I asked him to send me his logs, but haven't gotten them yet. I have watched this in the past and my experience is similar to yours.
When that cache is full then the speed drops back to the speed of the physical disk writes. I haven't activated my parity drive yet and I am maxing out my Lan connection and getting pretty much MB consistently copying stuff over - and that so far is a couple TB into a 4 TB copy for this folder.
I recently upgraded parity and one data drive to 4TB on a server that otherwise had all 2 TB drives. I copied over a TB of data to the drive Change the "1" to "0" or reboot to turn off turbo writes.
What turbo writes does is spin up all the disk to calculate parity during writes. Not really energy efficient. Anyway, I only turn on turbo writes when transferring large amounts of data. Normal writes only spin up the disk being wrote to and the parity disk. Never knew that was even an option. I'll try it myself, then offer the suggestion to my friend too.
I don't know if he keeps his running all the time though. I remember this option from V5. Pretty nice if moving large files.We hope to see it grow in coverage and detail, beyond the current limited set of questions and answers, and the common use of links to answers rather than a full answer.
The process of adding questions and answers involves first adding a question with a link to a relevant forum thread, then hoping for someone to take the time to add a full writeup. It is nearly entirely written by users of unRAID. It is not an official document written by Lime-technology LLC. With apologies, this page is under construction, and the included info is still rather meager and dated, but with time and your help, it will improve.
Once a disk is disabled, its contents may be invalid, because there may have been uncompleted writes. All further write requests will be similarly simulated. It is very important to determine WHY the disk was disabled, because very often it is not the fault of the drive.
More often than not, we find that there is an interface issue to the drive, such as a bad SATA cable very common or bad or loose SATA cable connectors, or power issues such as bad or loose power cable or its connectors, bad power cable splitter, too many drives on the same 12v power rail, or bad power supply.
Some times, it is just a disk controller that went offline, making it impossible to access the drive.
The first step then is to examine both the syslog to see what kind of errors are being reported, and the SMART report for the drive, to check its status and errors. Please see the Troubleshooting page for more information and instructions. For advanced users, see the The Analysis of Drive Issues page for examples of drive and interface error messages. If you determine that the drive is bad, then it needs to be replaced, see "How do I replace a hard disk?
If the drive was fine, and you either found and corrected an interface issue, or could not find anything wrong, then the disabled drive still needs to be re-enabled.
There are 2 ways to do it, and the choice depends on whether you have made changes to this drive after it was disabled. It was taken out of service and disabled because a "write" to it failed. It is guaranteed to not have the correct current contents. If you do not care about losing anything recently written to the drive, or can live with a file that was partially written with the subsequent "write" failed, or file-system corruption because the file structures were not written properly, then read the Trust My Array procedure, and see if your array qualifies for it.
If so, then the Trust My Array procedure is perfect for quickly returning your array to all green. Let the ensuing parity check complete. You have been warned.By wrenchmonkeyJune 10, in General Support. I've been running Unraid for almost a year now.
I recently noticed since upgrading to 6. Includes the cache.
unraid write speeds
To be fair, I did move the drives to a new case; and am now running a different board with dual processors. I've tried turning on Reconstruct write, and it had literally zero effect. And it shouldn't matter for writes to the cache drive anyway. I've looked everywhere I know to look to find anything it would consider a disk 29, and can't find anything. I assume disk 0 is one of the parity drives? I googled the error and from the posts I've found, it's related to a bug with Unraid not being able to spin down SAS drives?
Not sure if that's accurate, but it makes sense since I do have SAS drives in my machine. I spent the past few days going through, cleaning things up, making sure only the data that I want on the cache drive is on there, and that there are no files stuck in limbo trying to transfer out of the cache due to insufficient space on a share.
Desperate for any ideas at this point. Thanks in advance for any tips or ideas you guys might have! Do you have some examples of other writes? The screenshot you posted is doing disk to disk inside the array, so those speeds shown are perfectly in line with what I would expect.
That seems horrendous even for spindle drives to me. But yeah, even when transferring from data to cache, or over network to cache, the speeds are pretty bad.
Here's a Krusader screenshot from a test I just did moving to the cache drive. It peaked at 70MBPS for literally 2 seconds, before dropping down to 5MBPS and then crawling back up to this and staying right around after that. You are providing too little information for anyone to able to figure out the issue.
How are the drives connected? Many MB have both, old ones in particular. And I have to guess since there's not any info. You should also fix the spamming in your syslog, if it's the "SAS drive spin down bug" causing that like you said you should disable spin down for the parity drives if they are unable to spin down anyway.
And disk 29 is the parity2 drive FYI. Cache-to-cache is actually significantly better. I fixed the log spamming issue, I think. No change on transfer speeds though. All SATA connectors on this board are 3. It also has an onboard SAS controller, which I've connected with breakout cables. I have a breakout board for it as well, but currently am just plugged directly into the SAS port for most of the drives. It's an older server board, but it's still a pretty capable board.By johnnyfiveMarch 20, in General Support.
My old NAS was basically a windows computer with share drives. I understand writes will be slower due to parity and such.
NCQ is also enabled. I will add a cache disk in the future, hopefully that will give better write speeds as well.
My main question is how to I increase my read speeds? An additional question is I did not use preclear when I set up unraid. Will that be an issue? Also if you know other things I can do to increase write speeds, that would be helpful other than adding a cache drive. There are those who have gone to 10Gb Ethernet to up speeds but that is a significant hardware investment and can take a bit of filddling to get your network set up correctly to exploit such hardware.
Your just too fast. I would like fast read speeds if I'm transferring files and such off the drive for various purposes. I'm going to test going to windows 7 on my unraid set-up and trying a transfer to see if it's hardware or a software issue.
My tests for this speed was when I changed my Unraid box to a windows 7 OS. I unplugged my usb drive unraid and put in another HDD with windows 7 in it to see if it was a hardware issue and it seems those would be my 'maximum speeds' with this system build.
Just to be sure, youre not transfering files from usb? Just unraid runs on usb? Connectet to hdds via sata 3? Try connectiong all drives to your mobo, i heared somewhere that normal controllers could make trouble.
Designing an unRAID server
Normal you want a HBA card cheaper and you can connect more drives. Unless the NIC doesnt work well in Linux. I had a parity check going. Thanks nuhll. MTU values have to match on a given network segment. If you change one NIC from bytes you have to change them all, and to the same value. You can post now and register later. If you have an account, sign in now to post with your account. Note: Your post will require moderator approval before it will be visible.
Only 75 emoji are allowed. Display as a link instead. Clear editor.
Upload or insert images from URL. Product Pricing Community Blog. Sign In. Search In. General Support.An unRAID server is made up of a standard set of parts. This page discusses some of the general issues that need to be considered when designing your server, and how they influence the parts that you choose.
The Hardware Compatibility page should be read in conjunction with this page. It lists brands and models that have been reported by users to work with unRAID builds, and includes recommended builds, with links to online stores in several countries. Security of data is paramount with any filestorage system, including unRAID. Although unRAID provides for a parity disk to protect against data loss or other malfunction, it is best to reduce risk further by using reputable brands, and products that have been tested by other users.
The place where the server will be stored influences purchase and design decisions, both aesthetically and practically. Full tower and mid tower cases have plentiful space for hard drives, but are bulky and have a large footprint; if the server is to be in plain view, a smaller form factor or a more expensive and better looking case may be more acceptable.
For a standard unRAID server, focused on reading and writing files and calculating parity data, high-specification parts are not required. If the server will be used for CPU-intensive tasks, such as torrents, video encoding or virtual machines, higher specification CPUs with dual cores will be advantageous. If the server will be used for tasks such as running torrents or virtual machines, at least 4 GBs of RAM is recommended. Using 1x2GB allows more room for future expansion up to 4 or 8 GB.
Check that the memory matches the motherboard's DIMM slots.
Just look for a motherboard with plenty of SATA ports. Also, a separate graphics card is not required by unRAID. Only purchase one if your motherboard does not have onboard video i. They usually end in either a 60G or an 80G.
There is a balance between the use of fewer, large capacity drives to reduce the amount of power required by the system, with an increased risk of large-scale data loss if two drives fail at the same time. Green or LP low power drives — or rpm — are often recommended both for parity and data because they are cooler, quieter, and use less power than higher speed drives.
Improved performance can be seen if both the parity drive and the data drives are of a faster rotational speed: having a rpm parity drive with all rpm data drives will not be of benefit. There is a compromise — having a rpm parity drive and one or more rpm data drives the drives that will be used most-used as well as some rpm data drives. In this case, writes to the rpm data drives will be faster, and writes to the rpm data drives will be slower.
Using a rpm cache drive will improve the server's perceived write speed as all writes will initially be going to the faster cache drive. A cache drive will only help with perceived write speeds, not read speeds. However, a cache drive only defers the inevitable slow write into the parity protected array until later in the night when the server is otherwise idle. Temporally spacing out purchases of drives will decrease the chances of getting a bunch of drives from the same potentially bad batch.
Mixing and matching drive makes, models, and retailers can help as well. To increase the number of drives that can be held in a case, 4-in-3 or 5-in-3 stacker modules can be used to fit 4 or 5 hard drives into the space of three 5.
These modules typically come in two forms:.
Airflow and temperature regulation are extremely important in maintaining the health of the drives. Monitor the drive temperatures to ensure that they remain in optimal ranges. Drive temperatures between 30 and 39 C are ideal, between 40 and 45 C are acceptable. Drive temperatures over 45 C are in the dangerous range. Many hard drives will shut themselves down if they get above 55 C.
Good quality SATA cables are recommended; cheap cables have been known to cause problems. Locking SATA cables are also recommended, as non-locking cables are more easily dislodged due to drive vibration and, more commonly, user clumsiness.By VexorgJuly 16, in General Support. Back to it being slow. Had the machine off for a couple days, waiting on some parts to come in.2 Gaming Rigs, 1 Tower - Virtualized Gaming Build Log
I re-checked the cache settings and --get says they're all set to enabled. However when I did force it for sdf and sdg and refreshed the log I see an entry each for sdf and sdg saying it's now enabled.
You can post now and register later. If you have an account, sign in now to post with your account. Note: Your post will require moderator approval before it will be visible. Restore formatting. Only 75 emoji are allowed. Display as a link instead. Clear editor. Upload or insert images from URL. Product Pricing Community Blog. Sign In. Search In. General Support.
Reply to this topic Start new topic. Recommended Posts. Posted October 23, edited. Hello, just signed up to say thanks and share my config. Share this post Link to post. Posted October 26, edited. Did the WCE thing on all drives, no effect. Will try mounting an NTFS disk in a bit, do a test transfer, and see if things change.