rickym Posted July 13, 2019 Share #1 Posted July 13, 2019 I am setting up another Xpenology box with 3 4tb drives. Is it possible to setup the box without doing the parity check? Even with 4 2 tb drives on my other box it took forever. 2nd question is if at some time I have to replace a hard drive will it have to do the whole parity check again. Will be setup as SHR 2 like my other box. Quote Link to comment Share on other sites More sharing options...
merve04 Posted July 14, 2019 Share #2 Posted July 14, 2019 when you setup a new volume, it won’t do a parity check as there’s no parity to check. It can prompt you to do a sector check which can be skipped. If you need to replace HDDs, it’s one at a time and it will do a parity check on each replacement. Quote Link to comment Share on other sites More sharing options...
rickym Posted July 14, 2019 Author Share #3 Posted July 14, 2019 I just set up this new server with 4hard drives And it Automatically started doing the parity check and I didn’t see any option to not do that unless I’m not understanding what you’re saying. This is a brand new Xpenology box Quote Link to comment Share on other sites More sharing options...
jensmander Posted July 14, 2019 Share #4 Posted July 14, 2019 Every time a raid with parity (raid5, 6 or shr-x) is created, expanded or repaired, DSM will perform the parity check. It’s possible to skip this by rebooting the system, but not advisable. Other raid setups like mirroring (raid 1 or 10), stripes (raid 0) or jbod don‘t need this. 1 Quote Link to comment Share on other sites More sharing options...
rickym Posted July 14, 2019 Author Share #5 Posted July 14, 2019 (edited) Since I have setup as SHR2 I guess I could have setup as raid 10 instead to avoid this check. I don't plan on adding more disks or larger disks so is there really any difference between shr2 and raid 10. I have read some about it this morning but still a little confused about the difference with my setup. I am going to also do a backup and thinking hyperbackup looks like an easy way to backup to usb. I want shr2 or raid 10 due to the fact you could lose 2 drives and still be able to recover. This is just for home use and mainly I just like to learn more and I enjoy computers. Edited July 14, 2019 by rickym Quote Link to comment Share on other sites More sharing options...
jensmander Posted July 14, 2019 Share #6 Posted July 14, 2019 I always prefer „classic“ raid modes over shr. Shr is designed for the user with no to little knowledge and supports easy raid setups with drives of different size. And (imho) classic raids are more easy to recover and accessible with other *nix systems. Raid 10 is a combination of 2x stripes and has a theoretical redundancy of 2 faulty drives. If you have 4x2TB then the system will create two raid 0 and mirror these two raid 0. The result is (technical correct spoken) raid 1+0. That’s the reason why you will loose 50% of your total capacity. Due to it’s nature raid 1+0 is the fastest raid mode you can set up (read & write speeds) with redundancy. As written above in theory two drives can fail when using 4 drives. Raid 5 uses one drive for parity. That results in the biggest available volume size (beside raid 0 or jbod which provide no redundancy at all). Raid 6 uses two drives for parity and is slower and you‘ll lose more available capacity. There are other raid modes like 50, 60, 5E, 6E, etc. 50 & 60 are mixed modes of striping and parity. All these modes usually need special hardware raid controllers. 1 Quote Link to comment Share on other sites More sharing options...
rickym Posted July 14, 2019 Author Share #7 Posted July 14, 2019 (edited) Thanks so much. Still learning and that is exactly what I wanted to know. Raid 10 it is. I know it's over kill as I am also going to back my nas to usb. Edited July 14, 2019 by rickym Quote Link to comment Share on other sites More sharing options...
rickym Posted July 14, 2019 Author Share #8 Posted July 14, 2019 I have one move question. My nas is a 4 bay and don't anticipate ever needing more space an than the 4 drives in there but if I setup as raid 10 and at sometime want to have larger drives would I have to backup whole server and then add all 4 larger drives and reset them up in raid 10 array? Hope that makes sense. Quote Link to comment Share on other sites More sharing options...
jensmander Posted July 14, 2019 Share #9 Posted July 14, 2019 Yep. Classic Raid always takes the smallest drive as „reference“ (what makes sense). If you have 1x1TB and 3x4TB it results in 2TB in case of raid 10. So if you want to expand you always have to backup your data, recreate a new raid and restore the data. It‘s not possible to expand it step by step with larger drives. 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.