moondefer Posted August 12 Share #1 Posted August 12 (edited) Hi, i need help please Motherboard : ASROCK E3C224D4I-14S with SAS LSI 2308 in IT mode Cpu : intel core i5 4690k 4th generation Memory : 32 Gb ECC I'm actually with DSM 6.2.3.25426 Update 3 with DS3617xs I've decided to made test for update to DSM 7.2-64570 with DS3622xs+ I haved try loader TCRP and ARPL with only 1 disk for tested before try a real update with my 8 raid disks But i don't understand why all the time i have this error when i must install DSM pat file if i do a usb stick with 7.1-42661 loader or a 7.2-64570 loader : Curiously if i do a usb stick loader with 7.0.1-42218 with TCRP or ARLP i don't have this error when i choose my DSM pat file 7.0.1-42218 ? So i try this : loader with 7.0.1 then update to 7.1.1-42962 with DSM link work then i haved try to update again with DSM link to 7.2-64570 but error (21) and update is impossible loader with 7.0.1 then update directly with DSM link 7.2-64570 work and on the loader i found line with infos than zimage was changed to 7.0.1 to 7.2 but in DSM i'm with update 3 and in the loader infos i'm with update 2 ??? After this i try the last test on my real raid with the usb stick used just before who successfully updated zimage to 7.2. But again after migrate options proposed (ds3617xs to ds3622xs+ so it's good) i choose the DSM pat file and again i have this damned error message. Pat file i haved , found on synology center : 7.0.1-42218 : DSM_DS3622xs+_42218.pat OK only with this 7.1-42661 : DSM_DS3622xs+_42661.pat KO 7.1-42661 (with Update 1) : DSM_DS3622xs+_42661.pat KO (this include update 1) 7.2-64570 (with Update 1) : DSM_DS3622xs+_64570.pat KO (this include update 1) On synology download center i can't find 7.1-42661 without update 1 and 7.2-64570 without update 1 Only post i haved found on this it's with someone with a CPU 3th generation but it's not my case, i found info on my intel core i5 4690k and apparently it's a 4th generation and have FMA3 instruction. Thanks for your help I add infos from the manual motherboard on the CPU possible options (it's not my cpu in the picture) because i seen some post where people deactivated somme C options Edited August 12 by moondefer Quote Link to comment Share on other sites More sharing options...
dj_nsk Posted August 13 Share #2 Posted August 13 Try Arc loader (see the link in my sign below). download image (last version) and write it to USB-flash boot, select DSM platform (DS3615?) and version (7.2), answer "OK" to all other asks find DSM by Synology Assistant after rebooting, agree to migrate/restore with the settings saved, confirm the automatic installation .pat via the internet Enjoy Quote Link to comment Share on other sites More sharing options...
alepodj Posted August 13 Share #3 Posted August 13 Hello Having the same issue, even with the Arc Loader method above(much better by the way) for both 3622/3617, v7.2, yes to all, .pat via internet or manually... everything smooth until fails at 55%. System: HP DL380 Gen10, 10 drives detected, connected to the onboard sata controller Quote Link to comment Share on other sites More sharing options...
dj_nsk Posted August 13 Share #4 Posted August 13 34 минуты назад, alepodj сказал: Hello Having the same issue, even with the Arc Loader method above(much better by the way) for both 3622/3617, v7.2, yes to all, .pat via internet or manually... everything smooth until fails at 55%. System: HP DL380 Gen10, 10 drives detected, connected to the onboard sata controller Is Your current DSM version 7? or 6? at the beginning you need upgrade to 7.0 at least (it changes system partitions) Quote Link to comment Share on other sites More sharing options...
moondefer Posted August 13 Author Share #5 Posted August 13 (edited) Thanks for the reply dj_nsk The ARC loader seems to be most advandced. But with the same result When the loader start i have one red line but perhaps it's normal ? "user requested editing..." What i try : I put a formated 6 Tb disk and the usb stick with fresh ARC loader I choose DS3622xs+ then 7.2 version ... And again when i put the pat file, with DSM link or with my pat file on computer same result, 55% then error said image is corrupted. i have an xpenology since 5.2 loader With 5.2 loader install was very simple after editing config file With 6.X loader, sataportmap parameters need to be edit all the time we add a new disk Example with 6.X loader, i started with 6 diks, my sataportmap was 42 then 43 with 7 disk then 44 with 8 disks On the motherboard their are 2 hosts : SATA/SAS Storage motherboard : PCH Built-in Storage- Intel® C224 : 4 x SATA3 6.0 Gb/s (from mini SAS connector), 2 x SATA2 3.0 Gb/s, Support RAID 0, 1, 5, 10 and Intel® Rapid Storage Additional SAS Controller- LSI 2308: 8 x SAS2 6Gbps ( from 2x mini SAS 8087 connector) I must said than LSI SAS controller was in IT mode (unraid )and not in IR mode As i use only 4 port on C224 (because last 2 was only 3 gb/s) then i want use 8 port on LSI 2308 So as i said before sataportmap work good with 6.x loader with 4x parameters I found this strange with 7.X loader (with only 1 blank disk for test) : With TCRP loader : I ve seen loader haved driver for intel c220 then it find 1 disk (test disk connected so normal) and 6 ports (Yes it's right but i don't want to use lhe last 2) And it find LSI SAS and it download SCSI driver for it, said nothing about port then it write sataportmap = 6 and DiskIdxMap=00 With ARPL loader it write sataportmap = 6 and DiskIdxMap=00ff and i read sataportmap infos in loader it said 6 green port for intel and 7 green port for LSI With ARC loader it write sataportmap = 1 and DiskIdxMap=00 So i have a different result with 3 different loader I don't know if this is the reason why i have the error with pat file corrupted at 55% ? I can't find a solution alone, i need help for a solution so thanks to all for help link to the motherboard : https://www.asrockrack.com/general/productdetail.asp?Model=E3C224D4I-14S#Specifications link to the motherboard manual : https://download.asrock.com/Manual/E3C224D4I-14S.pdf link to the LSI 2308 manual : https://www.supermicro.com/manuals/other/LSI_HostRAID_2308.pdf Thanks for the help Edited August 13 by moondefer Quote Link to comment Share on other sites More sharing options...
alepodj Posted August 13 Share #6 Posted August 13 24 minutes ago, dj_nsk said: Is Your current DSM version 7? or 6? at the beginning you need upgrade to 7.0 at least (it changes system partitions) That is the one different from OP, im doing a clean install, no previous versions of DSM, but same error at the end(55%), figure i would post here Quote Link to comment Share on other sites More sharing options...
moondefer Posted August 13 Author Share #7 Posted August 13 (edited) OK With your previous reply alepodj and dj_nsk i understand now than if i want update my xpenology DS3617xs in 6.2.3 with 8 raid disks to DS3622xs+ 7.2 i must before start to update it to 7.0.1 first, but if i do this what i must do after for update it to 7.2 ? Yes i see than with 7.x their are more partition than 6.x. But i have the same problems than alepodj with my test with only 1 blank disk for test before a try for the true update And this problem afraid me to try a true update with my 8 raid disks because it's unsuccefull with only a try to install a fresh 7.2 on a formated disk. When i haved made test with only 1 formated disk, 7.0.1 install allways work but not 7.1 or 7.2, and after if i try to update to 7.1 or 7.2 with link or my pat file from computer i haved allways the same error at 55% corrupted file Edited August 13 by moondefer Quote Link to comment Share on other sites More sharing options...
alepodj Posted August 13 Share #8 Posted August 13 @moondefer How are u getting 7.0.1? With Arc i can only see 7.2/7.1 options. Quote Link to comment Share on other sites More sharing options...
alepodj Posted August 13 Share #9 Posted August 13 Ok somehow it went thru now, i dont know if the amount of times i tried, but the only thing i changed was booting from a different USB device type. I was using a USB to M.2 Adapter and went directly to a regular USB drive, but i swear i thought i tried that 1st lol Quote Link to comment Share on other sites More sharing options...
moondefer Posted August 13 Author Share #10 Posted August 13 (edited) Yes you are true ARC loader have only 7.1 and 7.2, only TCRP and ARPL have 7.0 and 7.1 and 7.2. So this is the question for migrating 6.2 to 7.2, we must use 2 loader for the task if we want ARC loader for the final ? i don't know. I need really a scenario for do that... I think i have noticed the partition modification when i made my test with 1 drive, after a succefully update with 7.0.1 i noticed a message in DSM who ask me to restart the DSM after the first start because the HDD database was modified and need a restart. I understand now why when i try directly to update to 7.1 this can't work. And why i have a success when i used ARPL to update to 7.0.1 then i haved restart for database HDD as asked, then i clic on DSM link update in DSM for 7.2 update and then i have succes to update to 7.2 only this time and the loader put a message on the screen as zimage was updated 7.0 to 7.2. So the loader have detected the update and update the loader automatically !!! if this is that it's a super loader. Edited August 13 by moondefer Quote Link to comment Share on other sites More sharing options...
moondefer Posted August 14 Author Share #11 Posted August 14 (edited) Today after some test with a test drive i decided to try a real upgrade from 6.2 to 7.2 1) have build a ARPL loader for update to 7.0.1 Update OK I must repair some packet and i made an config file save Then i clic on the DSM link for update to 7.2-64570 Update 3 The loader write to the screen connected to the nas than zimage was changed from 7.0.1 to 7.2 I don't nkow why but after this i have only 4 icon on the desktop, before that i have all my icon I don't understand why but my group and volume 1 was here but i can't see all my DATA ? So i go to Stockage and i see than my green disk was set to 1-4 then 7-10 Because the loader find my 1 host (intel c224) with 6 port and 4 disk, but i use only the 4 first port (sata3 6 GB/s) and not other (the last 2 port was only sata2 3 GB/s) But the loader have put satamap value to 6, he count the last 2 port than i don't use The ARPL loader is a pain for change satamap value or do command line, i think TCRP is most better for do all we want I made another stick loader with ARC I modify satamap value to 4 At first connection i have a recoverable status and choose it but finish with error I restart then in loader i choose restore from DSM (i tested this with success before when tested) Then i could start DSM correctly with all updated and my disk was to port 1-8 correctly But again, group and volume 1 was here but can't see all my data How i can retrieve the data ? their are here but not visible ? we can see 8.7 To free on volume1 With SSH i can see this : /$ cat /proc/mdstat Personalities : [raid1] [raid6] [raid5] [raid4] [raidF1] md2 : active raid5 sda3[11] sde3[10] sdf3[9] sdg3[12] sdh3[4] sdd3[8] sdc3[2] sd b3[7] 40989903808 blocks super 1.2 level 5, 64k chunk, algorithm 2 [8/8] [UUUUUU UU] md1 : active raid1 sda2[0] sde2[7] sdf2[6] sdg2[5] sdh2[4] sdd2[3] sdc2[2] sdb2[ 1] 2097088 blocks [12/8] [UUUUUUUU____] md0 : active raid1 sda1[0] sde1[7] sdf1[6] sdg1[5] sdh1[4] sdd1[3] sdc1[2] sdb1[ 1] 2490176 blocks [12/8] [UUUUUUUU____] unused devices: <none> mdadm --detail /dev/md2 /dev/md2: Version : 1.2 Creation Time : Tue Nov 13 20:04:35 2018 Raid Level : raid5 Array Size : 40989903808 (39091.02 GiB 41973.66 GB) Used Dev Size : 5855700544 (5584.43 GiB 5996.24 GB) Raid Devices : 8 Total Devices : 8 Persistence : Superblock is persistent Update Time : Tue Aug 15 01:19:15 2023 State : clean Active Devices : 8 Working Devices : 8 Failed Devices : 0 Spare Devices : 0 Layout : left-symmetric Chunk Size : 64K Name : DiskStation2:2 (local to host DiskStation2) UUID : f5cb8545:3f29b3e3:70e01c5b:283ef78f Events : 154623 Number Major Minor RaidDevice State 11 8 3 0 active sync /dev/sda3 7 8 19 1 active sync /dev/sdb3 2 8 35 2 active sync /dev/sdc3 8 8 51 3 active sync /dev/sdd3 4 8 115 4 active sync /dev/sdh3 12 8 99 5 active sync /dev/sdg3 9 8 83 6 active sync /dev/sdf3 10 8 67 7 active sync /dev/sde3 Someone can help me please ? parhaps with mount, unmount, perhaps reassemble raid ? If i delete volume 1 and recreate one data can apear again ? Perhaps the disk order have changed ? i have this one time with 6.2 loader at a moment disk order have changed but all work correctly Edited August 14 by moondefer Quote Link to comment Share on other sites More sharing options...
moondefer Posted August 15 Author Share #12 Posted August 15 (edited) OK Confirmed than i have the same thing that happen to me when i update to 6.1 to 6.2 many year before 4 Disk on the LSI 2308 inverse their order ? Again i have inverted the 4 cable to the 4 disk for have the good order (1 sas to 4 sata cable) Now it seems the order is good now # mdadm --detail /dev/md2 /dev/md2: Version : 1.2 Creation Time : Tue Nov 13 20:04:35 2018 Raid Level : raid5 Array Size : 40989903808 (39091.02 GiB 41973.66 GB) Used Dev Size : 5855700544 (5584.43 GiB 5996.24 GB) Raid Devices : 8 Total Devices : 8 Persistence : Superblock is persistent Update Time : Tue Aug 15 02:41:35 2023 State : clean Active Devices : 8 Working Devices : 8 Failed Devices : 0 Spare Devices : 0 Layout : left-symmetric Chunk Size : 64K Name : DiskStation2:2 (local to host DiskStation2) UUID : f5cb8545:3f29b3e3:70e01c5b:283ef78f Events : 154625 Number Major Minor RaidDevice State 11 8 3 0 active sync /dev/sda3 7 8 19 1 active sync /dev/sdb3 2 8 35 2 active sync /dev/sdc3 8 8 51 3 active sync /dev/sdd3 4 8 67 4 active sync /dev/sde3 12 8 83 5 active sync /dev/sdf3 9 8 99 6 active sync /dev/sdg3 10 8 115 7 active sync /dev/sdh3 But i don't see my data again The raid md2 seems to inchanged cat /proc/mdstat Personalities : [raid1] [raid6] [raid5] [raid4] [raidF1] md2 : active raid5 sda3[11] sdh3[10] sdg3[9] sdf3[12] sde3[4] sdd3[8] sdc3[2] sdb3[7] 40989903808 blocks super 1.2 level 5, 64k chunk, algorithm 2 [8/8] [UUUUUUUU] md1 : active raid1 sda2[0] sdh2[7] sdg2[6] sdf2[5] sde2[4] sdd2[3] sdc2[2] sdb2[1] 2097088 blocks [12/8] [UUUUUUUU____] md0 : active raid1 sda1[0] sdh1[7] sdg1[6] sdf1[5] sde1[4] sdd1[3] sdc1[2] sdb1[1] 2490176 blocks [12/8] [UUUUUUUU____] Oh my god, it's because i have deactivated my admin user as they said... very stupid grrrrrrrrrr, ok i go and search how replace my old admin by the new user. Edited August 15 by moondefer Quote Link to comment Share on other sites More sharing options...
bobby1701 Posted September 16 Share #13 Posted September 16 Hi Moondefer and all, I have very similar situation to you, that is a 4 disk SHR array. I have been reading and re-reading your notes and plan to do what you did, test with a new disk before I upgrade my real system. Just to explain, I had: DS3617xs which was updated by mistake to DSM 6.2.4-25556 and after that no longer could connect to the DSM. Having been advised to upgrade to DS3622xs+, I want to go to DSM 7.2-64570 Update 3 Can you please let me know of the key points to take care of when updating my real system? By the way, I am planning to use DSM 7.2-64570 Update 3. Any issues with that? Many thanks 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.