February 3, 201115 yr Hey guys. I have unraid 4.7 installed. I did alot of reading before i set it up, but like most projects i wound up with a couple things i would have done differently if i had it to do over again. All 5 of 2TB disks are the same exact seagate model (bought from different locations). The thing i realized last night in reading this forum is that if a disk registers bad i have no way to telling what disk3 is in the case. So my plan was to one at a time shutdown and unplug a single sata cable from a single drive. When unraids starts it will detect a failure and i will see which drive failed. FYI i just did a verify on parity check and these disks, also just did a preclear just fine less then a week ago. So im willing to risk being unprotected for a few minutes several times. My worry is that when i shut down and plug back in the disk unraid will have marked the disk bad and it wont let it recover properly. Is this a valid concern?
February 3, 201115 yr Yes - a valid concern. This is not a good approach. You could use the trust my array procedure after each unplugging, but this is not a good solution. Give some details as to how your disks are installed in the case. Are they in drive cages? Screwed directly into the case? The easiest way to track your drives (IMO) is to record their serial numbers (I actually just record he last few characters) and which slot I have put it in. I wrote the myMain plugin for unmenu which has a way to record this information. Would pulling the disks and looking at the label on top be possible? Many users label the back of their disks with the serial number to make it easy to locate in case of need.
February 3, 201115 yr Hey guys. I have unraid 4.7 installed. I did alot of reading before i set it up, but like most projects i wound up with a couple things i would have done differently if i had it to do over again. All 5 of 2TB disks are the same exact seagate model (bought from different locations). The thing i realized last night in reading this forum is that if a disk registers bad i have no way to telling what disk3 is in the case. So my plan was to one at a time shutdown and unplug a single sata cable from a single drive. When unraids starts it will detect a failure and i will see which drive failed. FYI i just did a verify on parity check and these disks, also just did a preclear just fine less then a week ago. So im willing to risk being unprotected for a few minutes several times. My worry is that when i shut down and plug back in the disk unraid will have marked the disk bad and it wont let it recover properly. Is this a valid concern? You are correct, once marked as bad it will want to re-construct to a different drive. I would Stop the array Copy the "config" folder on the flash drive elsewhere. (It has a record of the state of all the disks) Then, perform your experiments. (have you considered just looking at the serial number of the disks? They would identify them) When done identifying the drives, with the array stopped copy the original config/super.dat file back from where you saved it to config/super.dat Then reboot the machine. It should reboot, see the original files and all disks intact. (assuming you got all the cables back in place)
February 3, 201115 yr One step further - these Seagates (especially if the 2TB LP models) may have the "bad" firmware. You can search here on how to perform a firmware upgrade to save yourself some grief in the future. One disk at a time.
February 3, 201115 yr Author Wow this might have been a classic "over think". I did save a screenshot of my devices page in a safe spot, and getting to the drives isnt any issue, so if i has a future problem i can carefully pull drives until i find the label with the serial number that went bad on it. Im at work, so dont have access to the drives but it makes sense that the serial coded into the drive will be written on the top of the drive. FYI sounds like i got my answer but to answer the question i have this case..no extra drive enclosures. http://www.newegg.com/Product/ImageGallery.aspx?CurImage=11-119-216-TS&ISList=11-119-216-Z01%2c11-119-216-Z02%2c11-119-216-Z03%2c11-119-216-Z04%2c11-119-216-Z05%2c11-119-216-Z06%2c11-119-216-Z07%2c11-119-216-Z08%2c11-119-216-Z09%2c11-119-216-Z10&S7ImageFlag=1&Item=N82E16811119216&Depa=0&WaterMark=1&Description=COOLER%20MASTER%20RC-692-KKN2%20CM690%20II%20Advanced%20Black%20Steel%20ATX%20Mid%20Tower%20Computer%20Case Thank you guys!
February 3, 201115 yr Author OMG..why would they keep selling drives that are in the recall serial range?!?!? Amazon and newegg both.. SHAME ON YOU!! Both of the places i bought from sold me the drives in need of a firmware fix to not die.. = P If anyone reads this thread later, see here http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?Tab=search&Hilite=&Module=selfservice&TargetLanguage=selfservice&DocId=213915&NewLang=en
Archived
This topic is now archived and is closed to further replies.