Jump to content
XPEnology Community

DSM 6.2 Loader


jun

Recommended Posts

Hi,

I have recently updated my USB boot to 1.03b and accidentally instead of migrate, installed and now my admin password is reset. I was foolish not to enable ssh so the method I found would not work. is there anyway to reset via the boot or any other ways which I am not aware of? my last resort is to buy a new HDD and set up a new environment then migrate my two existing HDD once the new system is up, is this the only option left? your assistance is greatly appreciated.

 

Thanks in advance.

Link to comment
Share on other sites

Just reporting some information in the hopes that others find it useful.

 

System:

ASRock Z490M-ITXac

Intel Core i3 10100 s1200

4x12TB Toshiba HDDs (CMR & 5 year warranty)

16GB Corsair DDR4 Venagance 2666 RAM

Noctua Cooler NH-LPi low profile cooler

Fractal Design Node 304 code (space for 6 HDDs)

Samsung EVO 950 512GB SSD (proxmox, nas boot and nas ssd cache)

 

Software:

Proxmox 6.2-1 OS for VM's and Docker Containers

HostAPD for a Wireless access point using the built-in WiFi card on the motherboard.

 

Config:

PCIe Passthrough for the HDDs to the NAS VM.  48TB RAID-5.

2x 32GB Virtual SSDs for NAS Cache

1x 16GB Virtual SSD for NAS apps.

 

vm-100-xpenology1-q35-pcei-1hd-2ssd.thumb.JPG.f2e2480fcb0d96c6a8b69f30455750ed.JPG

 

VM Config

args: -device 'nec-usb-xhci,id=usb-ctl-synoboot,addr=0x18' -drive 'id=usb-drv-synoboot,file=/var/lib/vz/template/iso/synoboot_ds918_1.04-vm-xpenology1.img,if=none,format=raw' -device 'usb-storage,id=usb-stor-synoboot,bootindex=1,removable=off,drive=usb-drv-synoboot'
balloon: 0
bios: ovmf
bootdisk: sata0
cores: 2
efidisk0: local-lvm:vm-100-disk-1,size=4M
hostpci0: 00:17,pcie=1
ide2: none,media=cdrom
machine: q35
memory: 2048
name: xpenology1
net0: e1000=D2:F2:B8:6B:D0:40,bridge=vmbr0,firewall=0
numa: 0
ostype: l26
sata0: local-lvm:vm-100-disk-2,size=16G,ssd=1
sata1: local-lvm:vm-100-disk-3,size=32G,ssd=1
sata2: local-lvm:vm-100-disk-0,size=32G,ssd=1
scsihw: lsi
serial0: socket
smbios1: uuid=3964d1cc-e350-4fb3-b133-6903f5f05d7e
sockets: 1
vmgenid: c7f7222e-5bc7-4928-b6d1-6e0cd94f24d2

 

A personal note to Synology marketing - If you hadn't dropped Wireless AP support from the new DSM/NAS systems I'd have upgraded my existing DS415+ to a DS920+  That and the C2000 CPU-of-death issue.

 

What didn't work:

* Single SATA VM backed by a ZFS pool - It worked, i.e. you could create a simple volume in the running NAS on the ZFS pool, but proxmox only showed 22TB available capacity for a RAIDZ1 ZFS pool instead of the expected 30.64GB (48TB-12TB=36-overhead=33.6TB (30.64TiB).  The pool is correct, (when using zpool list and zfs list) but when you add an SATA disk to the VM it shows the wrong capacity. - https://wintelguy.com/zfs-calc.pl

 

root@proxmox:~# zfs list
NAME                 USED  AVAIL     REFER  MOUNTPOINT
tank                30.7T  33.8M      140K  /tank
tank/vm-100-disk-0  30.7T  30.6T     69.6G  -
root@proxmox:~# zpool list
NAME   SIZE  ALLOC   FREE  CKPOINT  EXPANDSZ   FRAG    CAP  DEDUP    HEALTH  ALTROOT
tank  43.6T  95.9G  43.5T        -         -     0%     0%  1.00x    ONLINE  -

root@proxmox:/tank# zfs list -o space,compressratio,recordsize,volblocksize -r -t all
NAME                AVAIL   USED  USEDSNAP  USEDDS  USEDREFRESERV  USEDCHILD  RATIO  RECSIZE  VOLBLOCK
tank                33.8M  30.7T        0B    140K             0B      30.7T  1.11x     128K         -
tank/vm-100-disk-0  30.6T  30.7T        0B   69.6G          30.6T         0B  1.11x        -        8K

 

* The DSM3615/DSM3617 synoboot 1.03 images.  They boot in the VM, there is some disk activity but no output on the VM console or VM serial console.  The DS419+ boots fine.

* Changing the VM from i440 to q35 re-orders the HDD ID's inside the NAS and kills the RAID-5 array.  I was inititally using PCI passthough, two of the 4 drives were no-longer detected.  The solution appears to be to re-upload the synboot image to the VM, delete all the SATA VM attached disks and re-create them all...

* Trying to re-install the VM was a 6 hour headache, I was getting Error Code 13 on re-install onto fresh disks, deleting an re-uploading the synoboot image fixed that - as detailed here.

 

 

 

 

 

Link to comment
Share on other sites

  • 3 weeks later...

Hi people

I have a problem with my new Hardware, all 3 Version of the Juns boot loaders 1.04b(dsm918+) and 1.03b(DS3615/7xs)  stops loading on "Screen will stop updating shortly" - Continue", after this message the keyword stops responding the only working button is press and hold power button or the reset button. I try all 3 versions of Juns Bootloader.

 

Her my HW:

Mainboard: https://www.gigabyte.com/Motherboard/Z490I-AORUS-ULTRA-rev-1x#kf

CPU: Intel Core i9-10900 (LGA 1200, 2.80GHz, 10-Core)

RAM: Corsair Vengeance LPX Black (2x, 32GB, DDR4-3200, DIMM 288)

NvMe: 2 x Samsung 970 EVO Plus (2000GB, M.2 2280)

HDD: 3 x Seagate IronWolf Pro(16Tb)

USB-Stick(Bootloader): Sandisk Ultra Fit 16Gb - SDCZ430-016G-G46

 

Some one can help?

 

Regards

Edited by XDOM
Link to comment
Share on other sites

18 hours ago, XDOM said:

I have a problem with my new Hardware, all 3 Version of the Juns boot loaders 1.04b(dsm918+) and 1.03b(DS3615/7xs)  stops loading on "Screen will stop updating shortly" - Continue", after this message the keyword stops responding the only working button is press and hold power button or the reset button. I try all 3 versions of Juns Bootloader.

 

Her my HW:

Mainboard: https://www.gigabyte.com/Motherboard/Z490I-AORUS-ULTRA-rev-1x#kf

 

1 hour ago, luchuma said:

you need to define nic chipset Intel® 2.5 GbE and find if its supported by the dsm native drivers.

 

intel's new driver "igc" for the 2.5GbE nic's is only available in kernel 5.x and intel does not intent to provide a driver for older kernels (found a response from a intel developer about that, search the forum for details, i already wrote about that)

it would need a specifically backpored driver for kernel 3.10.105 and 4.4.59 (i found a redhat backported driver but as they have there own way's apart from the vanilla kenrnel the source did not compile with our kernel and i'm not able to fix that as i can't code)

 

if it has to be 2.5GbE then realtek or a 10G nic capable of 5G/2.5G

Link to comment
Share on other sites

Hi everyone

The NAS is Up and Running after the installation of a supported Network card "Intel X550-T2 (PCI Express 3.0)", thank you a lot.

I have some questions now.

image.thumb.png.c017316c76fa935fc948e94bff5160c3.png

Whats about the CPU Cores showed in the GUI and ssh terminal? I assume that the 8 Cores bellow are 2(threads) x 4 Physical Cores showed in the DSM Info tab. Are the other 6 Physical Cores unused?

grep -i --color "model name"  /proc/cpuinfo

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz

model name : Intel(R) Core(TM) i9-10900 CPU @ 2.80GHz


Is the support of SRV-IOV for the Network Card "Intel X550-T2 (PCI Express 3.0)" not available in Xpenology?

image.thumb.png.91fe4d4e317c1ccddd799e069506b84a.png

Whats about Transcoding Features of the CPU Intel Core i9-10900? Need some drivers, another module or extra/extra2.lzma? i found something about this here: 

 

Regards

 

Link to comment
Share on other sites

3 hours ago, XDOM said:

Is the support of SRV-IOV for the Network Card "Intel X550-T2 (PCI Express 3.0)" not available in Xpenology?

in guess yes if you use my extra.lzma because it contains ixgbevf.ko and this should support it

 

3 hours ago, XDOM said:

Whats about Transcoding Features of the CPU Intel Core i9-10900? Need some drivers, another module or extra/extra2.lzma? i found something about this here: 

thats the right spot and i tries to explain there what to do

if you look up your cpu

https://ark.intel.com/content/www/us/en/ark/products/199328/intel-core-i9-10900-processor-20m-cache-up-to-5-20-ghz.html

you can see its gpu device id, 0x9BC5

so you would need to copy the medded i915,ko file to your system to (hopefully) make it work (there was not that much feedback)

 

there is also something else

https://xpenology.com/forum/topic/13333-tutorialreference-6x-loaders-and-platforms/

about cpu core/HT support

3615/918+  - 8 cores (incl. "ht cores")

3617  - 16 cores (incl. "ht cores")

 

so your 10core (20 with HT) is even to heavy for 3617 (that does not have i915 support for transcoding) with HT active (HT off might cost you the power of one real cpu core? as ht adds about 10-15% performace per core)

biggest possible and newest cpu is not best choice here, when using 918+ and HT off you will loose even two real cpu cores as it only supports 8 cores (might only make sense when modded i915 driver is working for this cpu)

 

Link to comment
Share on other sites

Hi @IG-88

Than you a lot for your answers, at begin y buy the new HW to use it with ESXi and Xpenology as VM togheter with other VMs but after install ESXi i see that i can't use  the Raid/SATA Controller in pic passthrough mode. Then i decide to install Xpenology on the bare metal ;-) that will be the reason of the oversizing of the HW.

 

What do you think should be the best solution to get the most power of the Hardware above?

 

Regards

Link to comment
Share on other sites

Sumary of your explanation about CPU Power

 

You wrote that HT On increase the Power of a cpu to 10-15% x core but in my case this means that i will have (115% CPU Power / 20) * 8(because i can only support 8 Cores with the 918+ Right? Than i will only have 46% of the CPU Power

 

If i disable HT i lose the 10-15% increase of the CPU Power but los only 2 Cores of cause from 918+ > (100% CPU Power / 10(Cores) x 8)= 80% CPU Power used 

 

Right?

Link to comment
Share on other sites

Disable HT you are using 8 real cores.  Enable HT you are using 4 real cores + four hypertheads.  8 real cores beat 4C+4T any day.

 

Change to DS3617xs and disable HT and you will use all 10 cores.  Enable HT and you will use 8 real cores plus 8 hyperthreads.  I would still pick the real cores over hyperthreads in this case.

Link to comment
Share on other sites

1 hour ago, XDOM said:

the Raid/SATA Controller in pic passthrough mode

maybe not the onbard as its used for booting esxi but if you add a jmb585 with 5 ports you could hand over that to  the vm or you use rdm mapping of disks attached to the onboard controller, i guess both can be combined if needed - but after lookiing what board you choose .... yikes, mini-ITX is extremly limiting the options with just one pcie slot and that shitty 2.5G intel nic that has no linux driver below kernel 5.0 and no vmware driver ... btw. if you bought for vmware esxi, how did you plan to use the onboard nic?

 

1 hour ago, XDOM said:

Than i will only have 46% of the CPU Power

no, 10 cores that add 15% of a real core each make 1.5 cores so your 100% would be 11.5 cores, when switching down to 8 its ~70% of what would be possible

not sure whats, but it depends on the software at best its 25% i guess so it could be a 2.5 cores equivalent more with HT , making it 12.5 vs 8 and thats 64%

 

maybe you can sell the board (cheaper then the cpu) and get a mATX or ATX with more pcie slots? that way you would be able to use your cpu with esxi as you can use a added ahci controller and even a 10G nic later, there should be a 16x slot a 4x slot and at least one 1x slot (all by lanes available, thats what i bought)

Link to comment
Share on other sites

Hi @IG-88

For this setup the principal decision criteria was small and quiet, that's the reason because i buy a mITX. I already have a Laboratory setup of 3 Server in a ESXi Cluster. This Machine is only for Home und it should have the most performance be quiet, small, powerful but with hi energy efficiency(low consumption).

 

I have a question about your explanation about the Cores.

  1. Wen HT is enabled the OS(in my case DSM 6.2.3) see 20 core but support and can control only 8 o the 20 cores?
  2. wen HT is disables the os see only the 10 physical cores and only 8 of that 10 are usable? 

 

Hi @flyride

I already try to use the DS3617xs but it dosent boot :-(

 

Thank you for your help

Edited by XDOM
Link to comment
Share on other sites

2 hours ago, XDOM said:
  • Wen HT is enabled the OS(in my case DSM 6.2.3) see 20 core but support and can control only 8 o the 20 cores?
  • wen HT is disables the os see only the 10 physical cores and only 8 of that 10 are usable? 

yes to both and if you take the 20 version you will end up with 4 real cores and 4 from HT and in irl performance the HT cores are about 15-25% of a real core

so if you go baremetal with 918+ then you should disable HT to get the most out of it

flyride did comment on this above

3617 would give you the most cpu power but you loose hardware transcoding of the intel cpu (qsv)

 

not sure of that is a option for you but proxmox seems to use kernel 5.x and should have support for the 2.5GbE intel nic, maybe thats a possible replacement for esxi?

 

2 hours ago, XDOM said:

I already try to use the DS3617xs but it dosent boot :-(

 

 

loader 1.03b does not support uefi, you need csm to be active in bios and you also need to be booting from the non uefi usb device (you might see the legacy usb device only after booting once with csm active, was the case on my gigabyte board and it took me a while to realize)

 

https://xpenology.com/forum/topic/13333-tutorialreference-6x-loaders-and-platforms/

Link to comment
Share on other sites

3 hours ago, qq124718842 said:

Hello, can you tell me if DSM supports Intel turbo? There are no commands to view

 

its a question to be answered by synology?

we are bound to whats usable with the kernel that comes from synology, we might add some modules but it depends on the kernel config whats possible or working

examples here are hyper-v or amd support, not in the kernel nothing we can do about it without having a replacement kernel without braking dsm or its protections

 

i can't say anything about that as i dont know how linux handles this

you can check about the cpu config in the kernel config synology uses and if you still think its possible to use turbo you can commet on that

we have bromolow (3615), broadwell (3617) and apollolake (918+)

https://sourceforge.net/projects/dsgpl/files/toolkit/DSM6.2/

ds.%type of platform%-6.2.dev

\usr\local\x86_64-pc-linux-gnu\x86_64-pc-linux-gnu\sys-root\usr\lib\modules\DSM-6.2\build\.config

 

Link to comment
Share on other sites

On 11/13/2020 at 12:50 PM, IG-88 said:

in guess yes if you use my extra.lzma because it contains ixgbevf.ko and this should support it

 

Hi @IG-88

I copy the "extra.lzma/extra2.lzma for loader 1.04b ds918+ DSM 6.2.3 v0.13.3" to my boot Loader image but the SR-IOV are still not available :-(  

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...