IG-88 Posted February 9, 2021 Author Share #26 Posted February 9, 2021 39 minutes ago, Vaifranz said: Maybe my problem is the backplane. that would be my guess too, reconnection errors usually originate from cable problems, the value refers directly to s.m.a.r.t. "UDMA_CRC_Error_Count" also did a run of creating a raid5 with the 2nd jmb585 controller and 3617, no reconnection errors 2 hours ago, Vaifranz said: [ 3365.801343] ata14: SATA link up 3.0 Gbps (SStatus 123 SControl 300) thats odd, the "normal" link speed is 6.0 Gbps Quote Link to comment Share on other sites More sharing options...
IG-88 Posted February 11, 2021 Author Share #27 Posted February 11, 2021 On 2/7/2021 at 2:02 AM, Vaifranz said: Hi, I am testing the JMB585 card in a ds3617 system but I am having several problems and I don't understand why. Fresh installation, DSM 6.2.3 update 3, during the installation phase everything is ok, during the RAID realization phase (both SHR and RAID5 or 6) I find several disconnections of HDD. Tried it on two motherboards, AsRock Z97 extreme9 and AsRock C226WS +, but the problems are the same, HDD WD RED. ok, id did some more tests, nothing in case of reconnections (that points to interface/cable/backlane/connectors) there where still zero but i did see something "unusual" in the dmesg log but only for WD disks (had two 500GB disks one 2.5" the other 3.5") nothing like that with HGST, Samsung, Seagate or a Crucuial SSD MX300 [ 98.256360] md: md2: current auto_remap = 0 [ 98.256363] md: requested-resync of RAID array md2 [ 98.256366] md: minimum _guaranteed_ speed: 10000 KB/sec/disk. [ 98.256366] md: using maximum available idle IO bandwidth (but not more than 600000 KB/sec) for requested-resync. [ 98.256370] md: using 128k window, over a total of 483564544k. [ 184.817938] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 184.825608] ata5.00: failed command: READ FPDMA QUEUED [ 184.830757] ata5.00: cmd 60/00:00:00:8a:cf/02:00:00:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.845546] ata5.00: status: { DRDY } [ 184.849222] ata5.00: failed command: READ FPDMA QUEUED [ 184.854373] ata5.00: cmd 60/00:08:00:8c:cf/02:00:00:00:00/40 tag 1 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.869165] ata5.00: status: { DRDY } [ 184.872839] ata5.00: failed command: READ FPDMA QUEUED [ 184.877994] ata5.00: cmd 60/00:10:00:8e:cf/02:00:00:00:00/40 tag 2 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.892784] ata5.00: status: { DRDY } ... [ 185.559602] ata5: hard resetting link [ 186.018820] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 186.022265] ata5.00: configured for UDMA/100 [ 186.022286] ata5.00: device reported invalid CHS sector 0 [ 186.022331] ata5: EH complete [ 311.788536] ata5.00: exception Emask 0x0 SAct 0x7ffe0003 SErr 0x0 action 0x6 frozen [ 311.796228] ata5.00: failed command: READ FPDMA QUEUED [ 311.801372] ata5.00: cmd 60/e0:00:88:3a:8e/00:00:01:00:00/40 tag 0 ncq 114688 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 311.816151] ata5.00: status: { DRDY } ... [ 312.171072] ata5.00: status: { DRDY } [ 312.174841] ata5: hard resetting link [ 312.634480] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 312.637992] ata5.00: configured for UDMA/100 [ 312.638002] ata5.00: device reported invalid CHS sector 0 [ 312.638034] ata5: EH complete [ 572.892855] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 572.900523] ata5.00: failed command: READ FPDMA QUEUED [ 572.905680] ata5.00: cmd 60/00:00:78:0a:ec/02:00:03:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 572.920462] ata5.00: status: { DRDY } ... [ 573.630587] ata5.00: status: { DRDY } [ 573.634262] ata5: hard resetting link [ 574.093716] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 574.096662] ata5.00: configured for UDMA/100 [ 574.096688] ata5.00: device reported invalid CHS sector 0 [ 574.096732] ata5: EH complete [ 668.887853] ata5.00: NCQ disabled due to excessive errors [ 668.887857] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 668.895522] ata5.00: failed command: READ FPDMA QUEUED [ 668.900667] ata5.00: cmd 60/00:00:98:67:53/02:00:04:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 668.915449] ata5.00: status: { DRDY } ... [ 669.601057] ata5.00: status: { DRDY } [ 669.604730] ata5.00: failed command: READ FPDMA QUEUED [ 669.609879] ata5.00: cmd 60/00:f0:98:65:53/02:00:04:00:00/40 tag 30 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 669.624748] ata5.00: status: { DRDY } [ 669.628425] ata5: hard resetting link [ 670.087717] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 670.090796] ata5.00: configured for UDMA/100 [ 670.090814] ata5.00: device reported invalid CHS sector 0 [ 670.090859] ata5: EH complete [ 6108.391162] md: md2: requested-resync done. [ 6108.646861] md: md2: current auto_remap = 0 i could shift the problem between ports by changing the port so its specific for WD disks it was on both kernels 3617 and 918+ (3.10.105 and 4.4.59) as the error points to NCQ and i found references on the internet a tried to "fix" it by disabling NCQ for the kernel i added "libata.force=noncq" to the kernel parameters in grub.cfg, rebooted and did the same procedure as before (with 918+) and i did not see the errors (there will be entry's about not using ncq for every disks, so its good to see that the kernel parameter is used as intended) in theory it might be possible to just disable ncq for some disks that are really WD but that would need intervention later if anything is changed on the disks in general there was not problem with the raid's i build even with the ncq errors and btrfs had nothing to complain i'd suggest to use this when having WD disks in the system i'm only using HGST and Seagate on the system with the jmb585 so it was not visible before on my main nas Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 11, 2021 Share #28 Posted February 11, 2021 Interesting, I have all WD disks! Except the ones I use for tests. Does the 3615 kernel also have this "flaw"? Could you be more precise and tell me where to insert the command "libata.force = noncq", in the file grub.cfg? I would also like to do some tests on the systems I have. Thank you. Quote Link to comment Share on other sites More sharing options...
IG-88 Posted February 11, 2021 Author Share #29 Posted February 11, 2021 4 hours ago, Vaifranz said: Does the 3615 kernel also have this "flaw"? same kernel version as 3617, no reason to assume a change in central ahci code would not be in both 4 hours ago, Vaifranz said: Could you be more precise and tell me where to insert the command "libata.force = noncq", in the file grub.cfg? just add it at the end of the line as additional parameter to the others set common_args_918='syno_hdd_powerup_seq=1 HddHotplug=0 syno_hw_version=DS918+ vender_format_version=2 console=ttyS0,115200n8 withefi elevator=elevator quiet syno_hdd_detect=0 syno_port_thaw=1' a space separates entry's and the line is "closed" with the ' for 3617 its "set common_args_3617=" Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 11, 2021 Share #30 Posted February 11, 2021 2 hours ago, IG-88 said: 7 hours ago, Vaifranz said: Does the 3615 kernel also have this "flaw"? same kernel version as 3617, no reason to assume a change in central ahci code would not be in both What I thought, I'm not very well versed in computer systems, just a little passionate. 2 hours ago, IG-88 said: 7 hours ago, Vaifranz said: Could you be more precise and tell me where to insert the command "libata.force = noncq", in the file grub.cfg? just add it at the end of the line as additional parameter to the others set common_args_918='syno_hdd_powerup_seq=1 HddHotplug=0 syno_hw_version=DS918+ vender_format_version=2 console=ttyS0,115200n8 withefi elevator=elevator quiet syno_hdd_detect=0 syno_port_thaw=1' a space separates entry's and the line is "closed" with the ' for 3617 its "set common_args_3617=" Thanks for the tip, I'll try to follow it ... in the meantime I did a test: run 3617, DSM 6.2.3 RAID 5 on AsRock Z97 extreme 9 without "libata.force = noncq" Maxtor STM3320820AS on Z97 (built-in mobo) Maxtor 6L250S0 on Z97 (built-in mobo) WDC WD6400AAKS-22A7B0 on JMB585 (SATA card) WDC WD4000KS-00MNB0 on JMB585 (SATA card) WDC WD3200AAJS-00Z0A0 on JMB585 (SATA card) WDC WD3200AAKS-00YGA0 on JMB585 (SATA card) Seagate ST3320620AS on ASMEDIA 1061 (built-in mobo) RAID completed and no disconnections or warnings except this: [ 5023.397585] perf interrupt took too long (2643 > 5000), lowering kernel.perf_event_max_sample_rate to 50000 I don't know what to think, except that my problem is the backplane Quote Link to comment Share on other sites More sharing options...
IG-88 Posted February 11, 2021 Author Share #31 Posted February 11, 2021 3 minutes ago, Vaifranz said: I don't know what to think, except that my problem is the backplane what backplabe is it? some old part that does not support 6Gbps of SATA3? Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 11, 2021 Share #32 Posted February 11, 2021 8 minutes ago, IG-88 said: what backplabe is it? some old part that does not support 6Gbps of SATA3? I don't know exactly, I should take it apart, but it doesn't seem very old to me, the strange thing is that it is now running, not with JMB585 but with HBA, Dell 200 in IT mode (9211), it works fine, but it happens that every 4 / 5 days freezes, remains on but not accessible. Quote Link to comment Share on other sites More sharing options...
IG-88 Posted February 11, 2021 Author Share #33 Posted February 11, 2021 58 minutes ago, Vaifranz said: I don't know exactly, I should take it apart, but it doesn't seem very old to me, the strange thing is that it is now running, not with JMB585 but with HBA, Dell 200 in IT mode (9211), it works fine, but it happens that every 4 / 5 days freezes, remains on but not accessible. maybe try another nas distribution like omv or freenas to see if its working stable Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 12, 2021 Share #34 Posted February 12, 2021 3 hours ago, IG-88 said: maybe try another nas distribution like omv or freenas to see if its working stable Yes, a great idea, I can also try the system that is installed there without a backplane, so only mobos, disks and HBA cards. I will update you, in the meantime thanks for your time. Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 15, 2021 Share #35 Posted February 15, 2021 (edited) I did some tests, my system works fine without the 24 bay case backplane, I will test it with other systems like OMV or Unraid or FreeNAS to understand if the problem is hardware or software. Here I took a photo of the only code present. I searched the net but found nothing about it, the case has this code CSE-S46524. I hope it will be useful to someone. Edited February 15, 2021 by Vaifranz Quote Link to comment Share on other sites More sharing options...
Vaifranz Posted February 15, 2021 Share #36 Posted February 15, 2021 (edited) [ 98.256360] md: md2: current auto_remap = 0 [ 98.256363] md: requested-resync of RAID array md2 [ 98.256366] md: minimum _guaranteed_ speed: 10000 KB/sec/disk. [ 98.256366] md: using maximum available idle IO bandwidth (but not more than 600000 KB/sec) for requested-resync. [ 98.256370] md: using 128k window, over a total of 483564544k. [ 184.817938] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 184.825608] ata5.00: failed command: READ FPDMA QUEUED [ 184.830757] ata5.00: cmd 60/00:00:00:8a:cf/02:00:00:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.845546] ata5.00: status: { DRDY } [ 184.849222] ata5.00: failed command: READ FPDMA QUEUED [ 184.854373] ata5.00: cmd 60/00:08:00:8c:cf/02:00:00:00:00/40 tag 1 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.869165] ata5.00: status: { DRDY } [ 184.872839] ata5.00: failed command: READ FPDMA QUEUED [ 184.877994] ata5.00: cmd 60/00:10:00:8e:cf/02:00:00:00:00/40 tag 2 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 184.892784] ata5.00: status: { DRDY } ... [ 185.559602] ata5: hard resetting link [ 186.018820] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 186.022265] ata5.00: configured for UDMA/100 [ 186.022286] ata5.00: device reported invalid CHS sector 0 [ 186.022331] ata5: EH complete [ 311.788536] ata5.00: exception Emask 0x0 SAct 0x7ffe0003 SErr 0x0 action 0x6 frozen [ 311.796228] ata5.00: failed command: READ FPDMA QUEUED [ 311.801372] ata5.00: cmd 60/e0:00:88:3a:8e/00:00:01:00:00/40 tag 0 ncq 114688 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 311.816151] ata5.00: status: { DRDY } ... [ 312.171072] ata5.00: status: { DRDY } [ 312.174841] ata5: hard resetting link [ 312.634480] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 312.637992] ata5.00: configured for UDMA/100 [ 312.638002] ata5.00: device reported invalid CHS sector 0 [ 312.638034] ata5: EH complete [ 572.892855] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 572.900523] ata5.00: failed command: READ FPDMA QUEUED [ 572.905680] ata5.00: cmd 60/00:00:78:0a:ec/02:00:03:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 572.920462] ata5.00: status: { DRDY } ... [ 573.630587] ata5.00: status: { DRDY } [ 573.634262] ata5: hard resetting link [ 574.093716] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 574.096662] ata5.00: configured for UDMA/100 [ 574.096688] ata5.00: device reported invalid CHS sector 0 [ 574.096732] ata5: EH complete [ 668.887853] ata5.00: NCQ disabled due to excessive errors [ 668.887857] ata5.00: exception Emask 0x0 SAct 0x7fffffff SErr 0x0 action 0x6 frozen [ 668.895522] ata5.00: failed command: READ FPDMA QUEUED [ 668.900667] ata5.00: cmd 60/00:00:98:67:53/02:00:04:00:00/40 tag 0 ncq 262144 in res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 668.915449] ata5.00: status: { DRDY } ... [ 669.601057] ata5.00: status: { DRDY } [ 669.604730] ata5.00: failed command: READ FPDMA QUEUED [ 669.609879] ata5.00: cmd 60/00:f0:98:65:53/02:00:04:00:00/40 tag 30 ncq 262144 in res 40/00:00:e0:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) [ 669.624748] ata5.00: status: { DRDY } [ 669.628425] ata5: hard resetting link [ 670.087717] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 670.090796] ata5.00: configured for UDMA/100 [ 670.090814] ata5.00: device reported invalid CHS sector 0 [ 670.090859] ata5: EH complete [ 6108.391162] md: md2: requested-resync done. [ 6108.646861] md: md2: current auto_remap = 0 I had the same problem with this HDD model: WDC WD60EFAX-68SHWN0, connected to JMB585. So it's really a problem with Western Digital drives especially the 6Tb WD Red model. I haven't tried the “libata.force = noncq” modification yet. What difference does the system have with this change? Edited February 15, 2021 by Vaifranz Quote Link to comment Share on other sites More sharing options...
IG-88 Posted February 15, 2021 Author Share #37 Posted February 15, 2021 6 hours ago, Vaifranz said: I haven't tried the “libata.force = noncq” modification yet. What difference does the system have with this change? no NCQ (sata native command queuing) https://en.wikipedia.org/wiki/Native_Command_Queuing also try to disable VT-d in bios, might have effects too Quote Link to comment Share on other sites More sharing options...
masters Posted July 17, 2021 Share #38 Posted July 17, 2021 (edited) Hi, these ones seem interesting: https://www.sybausa.com/index.php?route=product/product&path=64_181_85&product_id=1087 (Quad ASM1064 connected to a ASM2806 PCI-e Bridge) https://aliexpress.com/item/1005002813788487.html (ASMedia ASM1166 + JMicron JMB5xx) https://aliexpress.com/item/1005002767617226.html (ASM1166 12-24 ports, even they described them as ASM1064) Any thoughts? Edited July 17, 2021 by masters Quote Link to comment Share on other sites More sharing options...
Deneon Posted October 16, 2021 Share #39 Posted October 16, 2021 Hello. What do you think of such a scheme? Скрытый текст Board from aliexpress, 4 chips JMB585, in total 20 sata ports. The board can be purchased separately: Скрытый текст And modules with JMB585 for the board: Скрытый текст Will Jun's Loader v1.03b DS3617xs dsm 6.2.3 start without problems? And is it possible to buy a board and 1 module first, and then add modules as needed? Motherboard Gigabyte GA-H61M-S1, cpu i3 2120, in this configuration, pci-e operates in 2.0 mode. Speed is not important, it is important to increase the volume of hard drives, ssd is not planned. Quote Link to comment Share on other sites More sharing options...
IG-88 Posted October 16, 2021 Author Share #40 Posted October 16, 2021 (edited) On 10/16/2021 at 5:50 PM, Deneon said: Hello. What do you think of such a scheme? Board from aliexpress, 4 chips JMB585, in total 20 sata ports. The board can be purchased separately: And modules with JMB585 for the board: you already might have recognized that there is no pcie bridge chip, so you would need a board/bios supporting pcie bifurication (4x4x4x4x) https://peine-braun.net/shop/index.php?route=information/information&information_id=7 i also mentioned that kind of m.2 jmb585 here https://xpenology.com/forum/topic/35882-new-sataahci-cards-with-more-then-4-ports-and-no-sata-multiplexer/?do=findComment&comment=192836 beside not working properly its also problematic for mechanical reasons as the pcb is very thin and you might break the board when plugining in sata while the m.2 board is in place and the forces from 5x sata cable can be to much too (it would at least need some added cable support to reduce the force/weight on the m.2 board) (i might try the forum search, i wrote one or two times about that m.2 jmb585, still have it but not using it, in the end used a m.2 "extension cable" with a pcie slot on the other side and a standard pcie card with a jmb585) Quote Will Jun's Loader v1.03b DS3617xs dsm 6.2.3 start without problems? And is it possible to buy a board and 1 module first, and then add modules as needed? as its just ahci there is no problem with drivers, its part of synologys original kernel, you can add cards later on but keep in mind dsm as we use it has a 24 drive limit and also jun's 918+ loader will only support 16 drives ootb (can manually modded for more), loader for 3615/17 comes with a 12 drive default, but can also be modded for more (usually people mod the soynoinfo.conf of the installed dsm, modding the patch of the loader is also possible but more complicated as it needs to mod jun's patch and repacking the extra.lzma) Quote Motherboard Gigabyte GA-H61M-S1, cpu i3 2120, in this configuration, pci-e operates in 2.0 mode. Speed is not important, it is important to increase the volume of hard drives, ssd is not planned. imho bifurication would need a newer hardware and even there its not that much supported maybe a pcie 8x sas card is a option? if performance is not so important (like only using 1Gbit network) one of the controlles mentioned 1 post above might be a option, but they usually split all the ports into two pcie lanes as the (cheap) pcie bridg chip only supports two pci lanes on the host, and thats 2x500MB/s for 10-16 ports ... the 3rd link is kind of strange, these 20/24 port cards are there with a 16x pcie and 4x pcie connector, that might indicate a simple pcie bridge chip with just two lanes, but will not be more then 4 lanes and with pcie 2.0 ists not going to be much, without knowing what chips are used its not suggested to buy a card like this Edited December 28, 2021 by IG-88 Quote Link to comment Share on other sites More sharing options...
tattico83 Posted November 18, 2021 Share #41 Posted November 18, 2021 Hi everyone, I don't usually write on the forum because the guides are very comprehensive, but now I'm asking for help with a problem I've had for some time. I have an Asrock N3150 motherboard with DSM6.1.5 and connected a WD RED 2TB and another 1TB hdd. Everything worked fine until I tried to add two hdd's with this expansion card https://it.aliexpress.com/item/4000205684859.html?spm=a2g0s.9042311.0.0.27424c4dssOWnY The problem arises when I connect two additional hdds to the card. If I don't connect them it doesn't give problems, if I connect them it doesn't start DSM, so I tried to connect them once DSM started, everything worked perfectly, but once restarted, it gave the same problems, but now if I connect them when DSM is started, does not recognize them. I would also like to update DSM, what is the latest upgradeable version? Is there a quick method? Can you help me? Quote Link to comment Share on other sites More sharing options...
IG-88 Posted November 18, 2021 Author Share #42 Posted November 18, 2021 5 hours ago, tattico83 said: The problem arises when I connect two additional hdds to the card. If I don't connect them it doesn't give problems, if I connect them it doesn't start DSM, so I tried to connect them once DSM started, everything worked perfectly, but once restarted, it gave the same problems, but now if I connect them when DSM is started, does not recognize them. i'd suggest to throw it away, might be a faulty one or e specific problem with the board, in theory it should work (as its ahci) but it the bios or build quality is a issue you can't do much (maybe try some bios settings that are pcie or controller related) i use 88se92xx (and jmb585/582) in my systems and had 88se92xx cards in many boards and not seen a problem like that, try a cheap 9215 (pcie 1x) or a asm chip based pcie 1x card (asm is not that praised but i had one of these on a board for years and it never did anything wrong) (ASM1061 or the newer ASM1064 comes to mind as a pcie 1x solution, there can also be asm1062 on a pcie 1x card, that would leave one pcie lane of the chip unused but would not be any problem when using it as a pcie 1x card) 5 hours ago, tattico83 said: I would also like to update DSM, what is the latest upgradeable version? Is there a quick method? with juns loaders its 6.2.3 update3 the N3150 cpu should also worj with 918+ image (1.04b loader) if you remeber what you did for 6.1.x with jun's older laoder, its pretty much the same with loader 1.03b/1.04b (usb vid/pid in grub.conf is the important thing) https://xpenology.com/forum/topic/13333-tutorialreference-6x-loaders-and-platforms/ https://xpenology.com/forum/topic/29872-tutorial-mount-boot-stick-partitions-in-windows-edit-grubcfg-add-extralzma/ https://xpenology.com/forum/topic/12422-xpenology-tool-for-windows-x64/ https://xpenology.com/forum/topic/7973-tutorial-installmigrate-dsm-52-to-61x-juns-loader/ if there are driver issues then maybe add a extra.lzma with more/newer drivers) https://xpenology.com/forum/topic/28321-driver-extension-jun-103b104b-for-dsm623-for-918-3615xs-3617xs/ Quote Link to comment Share on other sites More sharing options...
tattico83 Posted December 6, 2021 Share #43 Posted December 6, 2021 (edited) Ok, grazie mille per l'aiuto @IG-88, farò dei test e te lo dirò. Edited December 6, 2021 by tattico83 Quote Link to comment Share on other sites More sharing options...
Jacopo Bassan Posted December 21, 2021 Share #44 Posted December 21, 2021 @IG-88 thanks for the deep dive into this tests, I came here too because I was looking for a pcie to sata card, do you have any more specific recommendations for a recent 6xSATA3 one? I've got a SilverStone DS380 case with 8bays and with an ASRock H87M-ITX I can only add up to 4 disks, and since I currently have 3 WD Red and gonna soon need one more (and probably a sata SSD for cache too) I was looking around on what could work without hassle. I've seen the SilverStone ECS06 but not much updated info with DSM. Thanks in advance if you can help! Quote Link to comment Share on other sites More sharing options...
IG-88 Posted December 28, 2021 Author Share #45 Posted December 28, 2021 (edited) On 12/21/2021 at 11:22 AM, Jacopo Bassan said: do you have any more specific recommendations for a recent 6xSATA3 one? with 6 there is not much choice asm1166 based have 6 ports (with 2 pcie lanes supporting gen3) 8 port controllers (without multiplier) https://xpenology.com/forum/topic/19854-sata-controllers-not-recognized/?tab=comments#comment-122709 and also https://www.startech.com/en-de/cards-adapters/8p6g-pcie-sata-card last featuring 4 x asm1061 two port chips with one pcie lane gen2 each and a pcie bridge chip (PI7C9X2G608GP) bringing one lane of 4 to each asm1061, its like having four asm1061 cards in one slot - still only pcie 2.0 but at least it uses 4 lanes (the two cards from the forum link above only use 2 lanes for 8 ports) Quote I've got a SilverStone DS380 case 8x 3.3" and 4x2.5" Quote with 8bays and with an ASRock H87M-ITX I can only add up to 4 disks, and since I currently have 3 WD Red and gonna soon need one more (and probably a sata SSD for cache too) if you are only using 1Gbit network (112 MByte/s) you might not see any difference and when using "a" sata ssd it will be only read cache and that would be even more useless in a home environment, so if you really want ssd cache plan two disks but i would not spent money on that, more things can go wrong and no gain i use 10G network, 12x 4TB, no ssd cache and have 500-700 MB/s good enough for me Quote I was looking around on what could work without hassle. I've seen the SilverStone ECS06 but not much updated info with DSM. the ecs06 is just a asm1166 card and there are other brands my asm1166 test card from last year had one issue as it made dsm see 32 sata ports but as long as its the 2nd card of two (onboard + pcie sata) its not problem and even with more cards there might be ways to deal with it (DiskIdxMap and SataPortMap) asm1166 card seems to be the best way there are also lsi sas cards that use 8 pcie 2.0 lanes and offer 8 sata ports but with 918+ i'm not confident using them (3617 would be ok as the drivers come with dsm) Edited December 29, 2021 by IG-88 Quote Link to comment Share on other sites More sharing options...
Jacopo Bassan Posted January 1, 2022 Share #46 Posted January 1, 2022 On 12/29/2021 at 12:34 AM, IG-88 said: if you are only using 1Gbit network (112 MByte/s) you might not see any difference and when using "a" sata ssd it will be only read cache and that would be even more useless in a home environment, so if you really want ssd cache plan two disks but i would not spent money on that, more things can go wrong and no gain i use 10G network, 12x 4TB, no ssd cache and have 500-700 MB/s good enough for me Yeah I would have done just a read-only cache, I thought it would have helped with accessing smaller files as in my use case. Quote the ecs06 is just a asm1166 card and there are other brands my asm1166 test card from last year had one issue as it made dsm see 32 sata ports but as long as its the 2nd card of two (onboard + pcie sata) its not problem and even with more cards there might be ways to deal with it (DiskIdxMap and SataPortMap) asm1166 card seems to be the best way there are also lsi sas cards that use 8 pcie 2.0 lanes and offer 8 sata ports but with 918+ i'm not confident using them (3617 would be ok as the drivers come with dsm) Looking around I found this one Pcie 4x to 6 SATA which seems nice, I'm not in an hurry to get it but I like to do my research in advance, mostly I want something that can support full sata 3 speeds on all 6 HDDs, I think this should do, right? Thanks again. Quote Link to comment Share on other sites More sharing options...
IG-88 Posted January 1, 2022 Author Share #47 Posted January 1, 2022 42 minutes ago, Jacopo Bassan said: Looking around I found this one Pcie 4x to 6 SATA which seems nice, I'm not in an hurry to get it but I like to do my research in advance, mostly I want something that can support full sata 3 speeds on all 6 HDDs, I think this should do, right? Thanks again. there are different cards under this link, if its asm1166, the chip only has 6 sata ports, anything more will be by a sata multiplexer and thats not supported/used by dsm so it will always be 6 ports usable in dsm its important to read the specs of a chip and the asm1166 will only use two pcie3.0 lanes making it max. ~2000MB/s for 6 ports and that below sata3 link speed https://www.asmedia.com.tw/product/45aYq54sP8Qh7WH8/58dYQ8bxZ4UR9wG5 the pcie connector is no indication on the lanes used some manufacturer like to make it look impressive (but 4x is normal in that case as the pcie spec is only defined for 1x 4x 8x 16x connectors) any ssd's used should be an the onboard chipset sata to have optimal performance under all conditions if its just one or two ssd's and four normal hdd's it should also be fine on the asm1166 but its not going to be optimal with a all ssd's approach Quote Link to comment Share on other sites More sharing options...
Jacopo Bassan Posted January 2, 2022 Share #48 Posted January 2, 2022 Thanks for the clear explanation, now I’ve a more clear view in mind, and I really don’t want to give up performance specially for parity check, I don’t want to think about just regular use which I well know is bottlenecked by network speed. I’m currently still on DS3615xs with DSM 6.1.7 and I’ll probably update to DS3617xs so maybe LSI SAS isn’t bad for my use case, I read is supported as is right? I don’t need hardware encoding or NVME cache so I think I won’t miss DS918+. Quote Link to comment Share on other sites More sharing options...
IG-88 Posted January 3, 2022 Author Share #49 Posted January 3, 2022 22 hours ago, Jacopo Bassan said: I’m currently still on DS3615xs with DSM 6.1.7 and I’ll probably update to DS3617xs so maybe LSI SAS isn’t bad for my use case, I read is supported as is right? both come with native driver s from synology, 3617 has newer drivers then 3615 imho main problem performance wise is the hdd cache, you can't enable write back mode with lsi sas and if the controller is setting it by default to write trough then it will matter, i did not made complete comparisons but it looked like a ahci based setup with write back enabled had more performance the a lsi sas (10G nic needed to see any difference, if its just 1-4 ports of 1G nic i dont think you will see any difference) 22 hours ago, Jacopo Bassan said: I don’t need hardware encoding or NVME cache so I think I won’t miss DS918+. for real use cases you would need two of them to have write cache too, only read cache with just one is usually not that good in a home use scenario (pointless imho, a 16 or 32 GB RAM might have more impact then a single ssd - sata or nvme) 10G nic mandatory to see any reasonable performance gain also i prefer jmb585 over asm1166 (jmb585 made it into my live system and asm1166 is now only a spare for testing) in theory lsi sas should be ok with 3617 (would be the combination i would use if i want to use lsi sas - but i have not used the lsi sas for more then testing drivers lately - so take it with a grain of salt) if you find some one that has done more testing and has reliable results then go with that Quote Link to comment Share on other sites More sharing options...
Jacopo Bassan Posted January 7, 2022 Share #50 Posted January 7, 2022 I can’t thank you enough, with this plenty of options coming out of this wasn’t that straightforward. In conclusion as I’m adding 6TB drives one at a time when I need them, I’ll just don’t worry too much at this time on what I’ll use as soon as I’ll need the 5th one in the setup and fill up the sata ports in my motherboard. But if I had to get something today I’ll say yeah I’ll go with JMB585 5xSata and call it a day. Not planning to use SSD for caching anymore nor main storage so no need for crazy speeds, the JMB585 would be more than enough (I’ll not use the sata 5 port) for 8 max HDDs in my setup without bottlenecking the parity checks or normal use, and no driver worries etc, plus is cheaper than other solutions. In the meanwhile I’ll for sure upgrade to 16GB RAM (2x8) and that one more 6TB drive I need. If I’ll need more (faster network etc) I’ll have to rethink the entire setup and I don’t think that will happen anytime soon since I’m happy with my setup since late 2014, as I have no problem with both file server/backup use or media streaming. By that time maybe there will be newer options too. Thanks again for the patience 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.