August 28, 201213 yr This is my third attempt to upgrade my parity disk from 2TB to 3TB. I have done the following: 1- Precleared new drive - 3 times - it passed each time. 2- Stopped array 3- Assigned "no device" to Parity slot 4- Used "shutdown" button in GUI to shutdown server 5- Removed old parity drive from server 6- Inserted new Parity drive in server 7- Restarted server 8- Assigned new drive to Parity 9- Restarted array Did I follow the correct procedures? This is the first time I have attempted a parity upgrade and, so far, I have failed. Below are screenshots of the parity sync in progress and the disk properties. Should this upgrade work? I know, maybe it will, maybe it won't!
August 28, 201213 yr You've actually performed a few extra steps, but they really do not matter in this case. Preclear drive Stop array replace old parity drive with new Assign new drive as parity Start the array. The extra step you performed was in un-assigning the original parity disk. (it did no harm, but could have been skipped.)
August 28, 201213 yr Author I read the guide in this thread http://lime-technology.com/forum/index.php?topic=2280.15 concerning parity upgrades. According to that, the steps were OK; however, I noticed something that concerns me. The new parity disk never showed as a blue ball (new disk not in array) before I added it to the array and began the parity sync. It has always showed as an orange ball (invalid data) both before being added and during the parity sync. Should I be concerned about this?
August 28, 201213 yr I read the guide in this thread http://lime-technology.com/forum/index.php?topic=2280.15 concerning parity upgrades. According to that, the steps were OK; however, I noticed something that concerns me. The new parity disk never showed as a blue ball (new disk not in array) before I added it to the array and began the parity sync. It has always showed as an orange ball (invalid data) both before being added and during the parity sync. Should I be concerned about this? no need for concern.
August 29, 201213 yr Author Not surprisingly, my parity upgrade again failed. This time with 15 write errors (last time it was only two). Here is a screenshot of the results: And here is the syslog (I was preclearing another disk so the first part of the log is the preclear results): Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 194 Temperature_Celsius 0x0002 161 161 000 Old_age Always - 34 (Min/Max 7/47) Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 196 Reallocated_Event_Count 0x0032 100 100 000 Old_age Always - 0 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 197 Current_Pending_Sector 0x0022 100 100 000 Old_age Always - 0 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 198 Offline_Uncorrectable 0x0008 100 100 000 Old_age Offline - 0 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 199 UDMA_CRC_Error_Count 0x000a 200 200 000 Old_age Always - 0 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 223 Load_Retry_Count 0x000a 100 100 000 Old_age Always - 0 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: SMART Error Log Version: 1 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: No Errors Logged Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: SMART Self-test log structure revision number 1 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: No self-tests have been logged. [To run self-tests, use: smartctl -t] Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: SMART Selective self-test log data structure revision number 1 Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 1 0 0 Not_testing Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 2 0 0 Not_testing Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 3 0 0 Not_testing Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 4 0 0 Not_testing Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: 5 0 0 Not_testing Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Selective self-test flags (0x0): Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: After scanning selected spans, do NOT read-scan remainder of disk. Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: If Selective self-test is pending on power-up, resume after 0 minute delay. Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: == Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: ============================================================================ Aug 28 21:46:35 MediaNAS preclear_disk-diff[26944]: Aug 28 22:17:01 MediaNAS crond[1155]: failed parsing crontab for user root: cron="" Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967296, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967304, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967312, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967320, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967328, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967336, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967344, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967352, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967360, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967368, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967376, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967384, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967392, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967400, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967408, limit=4294967295 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967288/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967296/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967304/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: md_do_sync: got signal, exit... Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967312/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967320/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967328/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967336/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967344/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967352/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967360/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967368/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967376/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967384/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967392/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: disk0 write error Aug 28 22:23:16 MediaNAS kernel: handle_stripe write error: 4294967400/0, count: 1 Aug 28 22:23:16 MediaNAS kernel: md: recovery thread sync completion status: -4 Aug 28 22:23:16 MediaNAS kernel: md: recovery thread woken up ... Aug 28 22:23:16 MediaNAS kernel: md: recovery thread has nothing to resync Aug 28 22:53:11 MediaNAS kernel: mdcmd (22): spindown 2 Aug 28 22:53:12 MediaNAS kernel: mdcmd (23): spindown 3 Aug 28 23:02:14 MediaNAS sshd[31273]: Accepted password for root from 192.168.0.208 port 63836 ssh2 Aug 28 23:02:14 MediaNAS sshd[31297]: lastlog_filetype: Couldn't stat /var/log/lastlog: No such file or directory Aug 28 23:02:14 MediaNAS sshd[31297]: lastlog_openseek: /var/log/lastlog is not a file or directory! Aug 28 23:02:14 MediaNAS sshd[31297]: lastlog_filetype: Couldn't stat /var/log/lastlog: No such file or directory Aug 28 23:02:14 MediaNAS sshd[31297]: lastlog_openseek: /var/log/lastlog is not a file or directory! Aug 28 23:02:29 MediaNAS sshd[31273]: syslogin_perform_logout: logout() returned an error Once again, I am seeing "attempt to access beyond end of device" errors and then handle stripe write errors. This is always what causes the parity upgrade failure. What is going on? What can I check for the cause of this problem? I have tried with the disk connected to the MB SATA ports and to a port on the PCIe SATA controller. I only got two write errors when connected to the PCIe controller. This latest failure was when connected to a built-in MB SATA port. The MB is the Biostar TH61-ITX (UEFI BIOS) and the PCIe SATA controller is the SYBA SY-PEX40048 which uses the Marvell 88SX7042 chipset.
August 29, 201213 yr I would have guessed you could never get more than one write error, since the disk will be disabled once the first write error occurs. I'm a bit confused. As far as seeking past the end of the device... Something looks odd... This is the error: Aug 28 22:23:16 MediaNAS kernel: attempt to access beyond end of device Aug 28 22:23:16 MediaNAS kernel: sde1: rw=1, want=4294967376, limit=4294967295 Now, 4294967295 blocks * 512 = 2TB, not 3TB. It is as if the partition on the 3TB disk was created with an MBR style partition, and not with a GPT partition. That would make /dev/sde1 as 2TB (max) with un-predictable results. Or, it could be that your disk controllers are unable to handle disks greater than 2.2TB (and why they are saying limit=4294967295) What do you see when you type: gdisk -l /dev/sde what do you get when you type preclear_disk.sh -v
August 29, 201213 yr Author What do you see when you type: gdisk -l /dev/sde I get the following from this command: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) Should I upgrade to unRAID 5 RC6 (I am on v5 beta 14 now) and a newer Linux kernel? what do you get when you type preclear_disk.sh -v The preclear script version number is 1.13 A recent review on Newegg concerning my motherboard indicates that despite having a UEFI BIOS, the MB *will not* support 3TB disks. He changed to an Asrock MB and had no problems with 3TB disks. I am beginning to think the MB is the problem. Even though it says nothing in the BIOS release notes about UEFI/3TB disks, there have been two BIOS versions released for my MB beyond the one loaded on my board. It appears all they addressed were new socket 1155 CPUs (Ivy Bridge). I have contacted Biostar about the MB and 3TB disk support. I doubt I will hear anything. They have never answered my prior inquiries.
August 29, 201213 yr What do you see when you type: gdisk -l /dev/sde I get the following from this command: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) That is not helping, as libstdc++.so seems to be missing,and that might prevent a proper partition from being created. are you running or loading any add-ons that might be overwriting the shared library? If not, then the rc6-test2 version of unRAID is what you want to try... It has the newest kernel, newest drivers, and the additional fixes for NFS. http://download.lime-technology.com/download/unRAID%20Server%205.0-rc6-r8168-test2%20AiO.zip (disable any plugins or add-ons that might have added added libraries.) What does fdisk -lu /dev/sde show? (It should complain the disk is > 2.2TB and you should use gdisk instead) What does dd if=/dev/sde count=195 | od -c -A d | sed 30q show If your motherboard chipset does not support 3TB drives, and cannot be upgraded, and the external SATA disk controller does not support them either, then I see a new MB in your future.
August 29, 201213 yr Author What do you see when you type: gdisk -l /dev/sde I get the following from this command: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) That is not helping, as libstdc++.so seems to be missing,and that might prevent a proper partition from being created. are you running or loading any add-ons that might be overwriting the shared library? If not, then the rc6-test2 version of unRAID is what you want to try... It has the newest kernel, newest drivers, and the additional fixes for NFS. http://download.lime-technology.com/download/unRAID%20Server%205.0-rc6-r8168-test2%20AiO.zip (disable any plugins or add-ons that might have added added libraries.) What does fdisk -lu /dev/sde show? (It should complain the disk is > 2.2TB and you should use gdisk instead) Here's the output of the fdisk command: Disk /dev/sde: 3000.6 GB, 3000592982016 bytes 255 heads, 63 sectors/track, 364801 cylinders, total 5860533168 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x00000000 Device Boot Start End Blocks Id System /dev/sde1 1 4294967295 2147483647+ 0 Empty Partition 1 does not end on cylinder boundary What does dd if=/dev/sde count=195 | od -c -A d | sed 30q show Here's the ouput of the dd command: 0000000 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000448 002 \0 \0 377 377 377 001 \0 \0 \0 377 377 377 377 \0 \0 0000464 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000496 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 U 252 0000512 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0066048 326 214 352 \n 7 206 030 034 s 205 i 002 022 \0 \0 \0 0066064 \0 \0 \0 \0 \0 \0 \0 \0 004 \0 \0 T 343 * 025 0066080 204 003 \0 \0 036 \0 \0 \0 \0 \0 \0 \0 \0 020 314 003 0066096 212 002 002 \0 R e I s E r 2 F s \0 \0 \0 0066112 003 \0 \0 \0 005 \0 326 025 002 \0 \0 \0 372 % \0 \0 0066128 001 \0 \0 \0 = 326 306 225 l 356 F 033 217 313 5 360 0066144 016 023 251 t \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 0066160 \0 \0 \0 \0 $ \0 036 \0 021 310 352 N \0 N 355 \0 0066176 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0066240 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 001 \0 \0 \0 0066256 374 017 \0 \0 w 016 \0 \0 y 016 \0 \0 ' 016 \0 \0 0066272 # 016 \0 \0 = 016 \0 \0 > 016 \0 \0 g 002 \0 \0 0066288 K 020 \0 \0 277 / \0 \0 & 2 \0 \0 b 2 \0 \0 0066304 b H \0 \0 Y D \0 \0 274 \ \0 \0 274 \ \0 \0 0066320 360 \ \0 \0 377 \ \0 \0 4 _ \0 \0 \r _ \0 \0 0066336 022 _ \0 \0 036 _ \0 \0 030 _ \0 \0 j _ \0 \0 0066352 006 _ \0 \0 \v _ \0 \0 035 _ \0 \0 037 _ \0 \0 0066368 252 u \0 \0 256 u \0 \0 357 v \0 \0 020 v \0 \0 0066384 252 w \0 \0 252 w \0 \0 246 w \0 \0 246 w \0 \0 0066400 223 x \0 \0 254 x \0 \0 217 x \0 \0 211 x \0 \0 0066416 362 x \0 \0 362 x \0 \0 K x \0 \0 E x \0 \0 As to the matter of add-ons, I have Simple Features and Crashplan. The others are just utility packages (unmenu, openssl, powerdown, pciutils, etc.)
August 29, 201213 yr I've got a feeling that since the GPT partition could not be created properly you have a 2TB partition. (and writing past its end gives you the error) I see the normal "protective MBR partition table" starting on sector 1. To see how the disk is actually partitioned, you would need a working gdisk command.
August 29, 201213 yr Author I've got a feeling that since the GPT partition could not be created properly you have a 2TB partition. (and writing past its end gives you the error) I see the normal "protective MBR partition table" starting on sector 1. To see how the disk is actually partitioned, you would need a working gdisk command. I just updated the flash drive to v5 RC6-r8168-test2 by copying bzimage and bzroot over the existing configuration. I got the same error about the libraries: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) I will try a "clean" RC6 configuration with no add-ons and see if that makes a difference with gdisk.
August 29, 201213 yr I've got a feeling that since the GPT partition could not be created properly you have a 2TB partition. (and writing past its end gives you the error) I see the normal "protective MBR partition table" starting on sector 1. To see how the disk is actually partitioned, you would need a working gdisk command. I just updated the flash drive to v5 RC6-r8168-test2 by copying bzimage and bzroot over the existing configuration. I got the same error about the libraries: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) I will try a "clean" RC6 configuration with no add-ons and see if that makes a difference with gdisk. One of your add-ons is clobbering it. I am running test2 here, with no plugins, and have no issues with gdisk.
August 29, 201213 yr Author What do you see when you type: gdisk -l /dev/sde I get the following from this command: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) That is not helping, as libstdc++.so seems to be missing,and that might prevent a proper partition from being created. /usr/lib/libstdc++.so.6 exists on the flash drive as does /usr/lib/libstdc++.so.6.0.13 and /usr/lib/libstdc++.so.6.0.9 Is GLIBCXC_3.4.11 the "missing" piece. If so, where should I look for that?
August 30, 201213 yr What do you see when you type: gdisk -l /dev/sde I get the following from this command: gdisk: /usr/lib/libstdc++.so.6: version 'GLIBCXC_3.4.11' not found (required by gdisk) That is not helping, as libstdc++.so seems to be missing,and that might prevent a proper partition from being created. /usr/lib/libstdc++.so.6 exists on the flash drive as does /usr/lib/libstdc++.so.6.0.13 and /usr/lib/libstdc++.so.6.0.9 Is GLIBCXC_3.4.11 the "missing" piece. If so, where should I look for that? you should have no shared library files on the flash drive. I show this as far as dependencies (notice no mention of GLIBCXC_3.4.11) root@Tower2:~# ldd /sbin/gdisk linux-gate.so.1 => (0xb7748000) libicuio.so.44 => /usr/lib/libicuio.so.44 (0xb7738000) libicuuc.so.44 => /usr/lib/libicuuc.so.44 (0xb7602000) libuuid.so.1 => /lib/libuuid.so.1 (0xb75fe000) libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0xb750e000) libm.so.6 => /lib/libm.so.6 (0xb74e7000) libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0xb74ca000) libc.so.6 => /lib/libc.so.6 (0xb7367000) libpthread.so.0 => /lib/libpthread.so.0 (0xb734e000) libicudata.so.44 => /usr/lib/libicudata.so.44 (0xb6510000) libicui18n.so.44 => /usr/lib/libicui18n.so.44 (0xb633a000) libdl.so.2 => /lib/libdl.so.2 (0xb6335000) /lib/ld-linux.so.2 (0xb7749000)
August 31, 201213 yr Author After disabling a bunch of add-ons (I also managed to somehow kill unraid-web), the gdisk command worked. It reported that the disk had an MBR and changed it to a GPT. Even though GPT: 4K aligned (factory-erased) was reported as the format type in the web interface, the disk clearly had an MBR instead. After that, the parity sync to the 3TB drive worked. Come to think of it, the precelar script did indicate it was creating MBR and I just glossed over that at the time. Should it say GPT during preclear or is MBR just the generic term used for whatever partition table it is creating? Now, I can move on to replacing a 750GB array disk with a 3TB as well and getting unraid-web running again. Thanks, Joe.
September 1, 201213 yr Author My 3TB WD EZRX also had an MBR from the preclear instead of a GPT. Thanks to Joe's advice earlier in this thread, I was smart enough to use gdisk on it before attempting to replace a smaller drive. The array drive upgrade came off without a hitch. Now I just need to add two more 3 TB drives to the array and I'll be done with disk upgrades for the foreseeable future. I still don't know why the 3TB drives precleared with MBRs created instead of GPTs, but, at least I know enough now to check them before adding to the array and I now know I can fix them with gdisk.
September 2, 201213 yr My 3TB WD EZRX also had an MBR from the preclear instead of a GPT. Thanks to Joe's advice earlier in this thread, I was smart enough to use gdisk on it before attempting to replace a smaller drive. The array drive upgrade came off without a hitch. Now I just need to add two more 3 TB drives to the array and I'll be done with disk upgrades for the foreseeable future. I still don't know why the 3TB drives precleared with MBRs created instead of GPTs, but, at least I know enough now to check them before adding to the array and I now know I can fix them with gdisk. You should not ever need to do anything manually with the gpart command. The preclear process puts a protective MBR on all disks over 2.2TB (and a special MBR in place on those <= 2.2TB). It is to convince all older utilities and processes the disk is fully utilized. It is EXPECTED by unRAID. If you use gpart on the disk after it is precleared, it is NO LONGER PRECLEARED!!!!!!!!. unRAID will clear it again if you use it to expand your array and your array will be off-line while it does. In your case, it probably did not hurt as you are upgrading/replacing an existing disk, as long as you have a working gpart command for unRAID to use, as unRAID will create the GPT partition on the new disk as part of the expansion (or, at least, it should ) The protective MBR and lack of a GPT partition is part of how it recognizes the disk has been pre-cleared. Joe L.
September 2, 201213 yr Author You should not ever need to do anything manually with the gpart command. The preclear process puts a protective MBR on all disks over 2.2TB (and a special MBR in place on those <= 2.2TB). It is to convince all older utilities and processes the disk is fully utilized. It is EXPECTED by unRAID. If you use gpart on the disk after it is precleared, it is NO LONGER PRECLEARED!!!!!!!!. unRAID will clear it again if you use it to expand your array and your array will be off-line while it does. In your case, it probably did not hurt as you are upgrading/replacing an existing disk, as long as you have a working gpart command for unRAID to use, as unRAID will create the GPT partition on the new disk as part of the expansion (or, at least, it should ) The protective MBR and lack of a GPT partition is part of how it recognizes the disk has been pre-cleared. Joe L. OK good to know that I should not need to run gdisk after a preclear. As you said, since I was upgrading an existing disk, it made no difference and the array was still accessible while the disk was being upgraded from 750GB to the precleared 3TB. Since my next step is adding two new 3TB disks to the array, I understand I should just preclear them and add them without running gdisk manually since UNRAID can do what it needs to do with the partitions now that gdisk/gpart works properly and the shared libraries are not getting clobbered by and add-on. Thanks again for your help.
September 3, 201213 yr Unfortunately, I had the same issues when adding 3TB parity disk to array after preclear. However, I disabled all add-on's EXCEPT clean powerdown and unMENU and was able to get 3TB parity online afterward. Wondering if I will receive errors during parity check after re-enabling add-ons as I did when rebuilding parity? SO, I plan on roughing the waters and giving parity check a whirl while add-ons are active. Thanks for this thread!
Archived
This topic is now archived and is closed to further replies.