Jump to content
XPEnology Community

marigo

Member
  • Posts

    49
  • Joined

  • Last visited

Posts posted by marigo

  1. On 8/30/2022 at 9:19 PM, dvuv said:

    Hello,

     

    I'm little puzzled and looking for a hint from community.

    I have 2 Intel NUCs of different generations. Tried both automatic redpill and tinycore.

    1) based on i3-7100u CPU. Installed 1622 baremetal, everything (AI tasks) works fine, no issues. Decided to install through proxmox to not waste entire NUC for a single task, specified host CPU, and PCI passthrough for GPU (HD 620). Result – AI processes constantly crash and tasks do not work. One small detail if it matters – when enabling Local Screen, or right before shut down, attached monitor goes garbage. All other time I see Syno 1622 logo.

     

    2) I pulled HDD from this NUC and installed it as-is (proxmox setup) into 8-gen with i3-8109U CPU (HD 655 gpu). Booted without making any changes, and boom, everything works! So at this point i'm certain that I followed all steps for setting up passthrough in proxmox according to instructions and it works.

     

    I have no idea why 1622 works on baremetal, doesn't work in proxmox, but works in proxmox on a next-gen CPU. Logic says to me that it should work in the first case, but it doesn't. I'm out of clues on where to look or what else to try, any suggestions?

     

    Great to see that you have it working in Proxmox. I had also selected the "host" CPU, but AI also not working. But now I red that I have to use PCI (GPU) passthrough.

    I am on a Intel i5 12th gen, so it should work as soon I have the passthrough enabled. (will do that asap)

     

    On thing I noticed that the CPU is fluctuating and I see some spikes sometimes to 15-20%. Also the used RAM is also increasing in time...

     
  2. On 6/22/2022 at 7:40 AM, marigo said:

    Hi Flyride,

     

    Proxmox is running on an Asrock J1900 motherboard and the output from "satamap" is:

     

    tc@box:~$ ./rploader.sh satamap
    Machine is VIRTUAL Hypervisor=KVM
    
    Found "00:07.0 Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA Controller [AHCI mode] (rev 02)"
    Detected 6 ports/3 drives. Mapping SATABOOT drive after maxdisks
    WARNING: Other drives are connected that will not be accessible!
    
    Computed settings:
    SataPortMap=1
    DiskIdxMap=10

     

     

    For somehow the user_config.json file is adjusted to:

     

    "SataPortMap": "58",
     "DiskIdxMap": "0A00"

     

     

    I have two drives attached in Proxmox. Maybe one drive is not recognized because it is 10GB. (and the other drive is 100GB)

    As I have red there had to be one drive over 21GB....Maybe the 10GB is too small...

     

    EDIT: Changed the small drive to 22GB. Now it's working!

    I have tried to reproduce the error, but it is (was) all working with loader version v0.4.6

    If I execute the "./rploader.sh satamap now" it will detect the controllers and write the correct values to the user config:

     

    tc@box:~$ ./rploader.sh satamap now
    Machine is VIRTUAL Hypervisor=KVM
    HBA: 00:07.0 Disks : 6
    HBA: 00:01.1 Disks : 2
    SataPortMap=62
    DiskIdxMap=0002
    Should i update the user_config.json with these values ? [Yy/Nn]

     

    When I do the update and fullupgrade the "satamap" is not working on proxmox and I have to manually adjust the values

     

    Just to let you know...

     

  3. 16 hours ago, flyride said:

    Yes, something is wrong.  You'll need to post more information about your hardware (motherboard, any disk controllers) and the output from satamap if you'd like some advice. Be sure that you "update" and "fullupgrade" prior to running satamap

     

     

    Hi Flyride,

     

    Proxmox is running on an Asrock J1900 motherboard and the output from "satamap" is:

     

    tc@box:~$ ./rploader.sh satamap
    Machine is VIRTUAL Hypervisor=KVM
    
    Found "00:07.0 Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA Controller [AHCI mode] (rev 02)"
    Detected 6 ports/3 drives. Mapping SATABOOT drive after maxdisks
    WARNING: Other drives are connected that will not be accessible!
    
    Computed settings:
    SataPortMap=1
    DiskIdxMap=10

     

     

    For somehow the user_config.json file is adjusted to:

     

    "SataPortMap": "58",
     "DiskIdxMap": "0A00"

     

     

    I have two drives attached in Proxmox. Maybe one drive is not recognized because it is 10GB. (and the other drive is 100GB)

    As I have red there had to be one drive over 21GB....Maybe the 10GB is too small...

     

    EDIT: Changed the small drive to 22GB. Now it's working!

  4. I am using tinycore-redpill.v0.4.6.img to install DSM7.1 on Proxmox KVM.

    For storage I have attached 2 SATA drives.

     

    Everytime I use "./rploader.sh mapsata" I get incorrect output:

     

    "SataPortMap": "1",
    "DiskIdxMap": "10",

     

    When Build for "Appololake 918+" There are no drives detected. If I changed to an already working VM (Sataportmap: 62, DiskIdxMap:0002) Then only one drive is working.

     

    Am I doing something wrong?

  5. - Outcome of the update: SUCCESSFUL

    - DSM version prior update: DSM 6.2.3-25426 (JUN's loader 1.04b)

    - Loader version and model: RedPill  TinyCore V.04.6 DS918+ DSM-7.1.0-42661

    - Using custom extra.lzma: NO

    - Installation type: Proxmox KVM

    - Additional comments:

    add https://github.com/pocopico/redpill-load/raw/master/redpill-misc/rpext-index.json

    add https://github.com/pocopico/redpill-load/raw/develop/redpill-acpid/rpext-index.json

    add https://raw.githubusercontent.com/pocopico/redpill-load/master/redpill-virtio/rpext-index.json

  6. Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update:  DSM 7.0.1-42218 Update 2

    - Loader version and model: RedPill  Tinycore 0.4.6

    - Installation type: Proxmox KVM

    - Additional comments: Reboot required by the update, update additional apps

  7. 40 minutes ago, Invaliduser said:

     

    Are you generating a new mac address for the 7.0.1 build?  If so, you will need to assign that mac address to the VM's network device.

     

    Hi Invaliduser, I have used a new MAC address, so I will try that. (could it be that simple) thx for quick reply.

    EDIT: This works to update the MAC address of the VM with the one used in "user_config.json" (no workaround needed) THX!!

     

    In the meanwhile I discovered a workaround, to detach the "old" disk and add a new one. After boot the diskstation is visible and I can install DSM7.

    After install I shutdown the machine and added the datadisk again, so together with the new disk and booted. Now the diskstation sees the datadisk again and I can migrate to DSM7. After migration I deleted the new disk en rebooted. Everything seems te work. :)

  8. I have tried to update my VM on Proxmox from 6.2.3 to 7.0.1 (apollolake-7.0.1-42218) but no IP is given from DHCP.

    My setup is to boot from Juns loader 1.04B and have one bootdisk on SATA0 and a datadisk on SATA1:

     

    • vm-xxx-disk-0.raw <- bootdisk
    • vm-xxx-disk1.qcow2 <- datadisk

    I deleted the bootdisk (detach and remove drive) and added a new (raw) bootdisk with 2GB capacity. Changed the NIC from "E1000" to "Virtio", loaded Tinycore loader on this disk and booted. Everything is looking fine, Update the user_config.json and backup this configuration.

     

    After the reboot (booted in grub menu from SATA) there is no connectivity and no IP address has been given out by DHCP.

    If I stop this VM and detach my datadisk (which was previous om dsm6.2.3) and boot again, I get an IP address from DHCP and find.synology.com finds this "Diskstation".

    But there are no disks found to install DSM.

     

    Am I missing something here? I thought it could be the "satamap" so I applied "./rploader satamap now" and this will find the disks in KVM and added to user_config.json

    But that didn't do the trick.

     

    When I create a new VM with Tinycore loader, I can add a second disk with no problem and install DSM7 normally. So greenfield works, but upgrade not yet.

    Hope someone can point me in the right direction.

  9. - Outcome of the update:  SUCCESSFUL

    - DSM version prior update: DSM 6.2.3-25426 Update 2

    - Loader version and model: Jun's Loader v1.04b DS918+

    - Using custom extra.lzma: No

    - Installation type: Proxmox 6.2-6

    - Additional comment: Intel E1000, FixSynoboot.sh, reboot

  10. - Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update: DSM 6.2.3-25426

    - Loader version and model: JUN'S LOADER v1.04b - DS918+

    - Using custom extra.lzma: NO

    - Installation type: Proxmox 6.2.6 VM

    - Additional comments: Used the FixSynoboot.sh to update

  11. - Outcome of the update:  SUCCESSFUL

    - DSM version prior update: DSM 6.2.2 update 4

    - Loader version and model: JUN'S LOADER v1.04b - DS918+

    - Using custom extra.lzma: NO

    - Installation type: VM - Proxmox KVM 6.1-5

    - Additional comments: auto install from DSMIntel NIC E1000, reboot required

  12. - Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update:  DSM 6.2.2-24922 Update 2

    - Loader version and model: Jun v1.04b - DS918+

    - Using custom extra.lzma: NO

    - Installation type: VM - Proxmox 5.4.5 KVM (used standard Intel E1000 NIC with 1.04b loader; no need to change to E1000E)

    - Additional Comment: Reboot required 

  13.  Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update: DSM 6.2.1-23824 update 2

    - Loader version and model: Jun 1.04b - DS918+

    - Using custom extra.lzma: NO

    - Installation type: BAREMETAL - Asrock J1900-ITX

    - Additional comments: NONE

  14. - Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update: DSM 6.2.1-23824 Update 1

    - Loader version and model: JUN'S LOADER v1.04b - DS918

    - Using custom extra.lzma: NO

    - Installation type: BAREMETAL - Asrock J1900ITX

    - Additional comments: none

  15. Great news!

     

    Didn't  know about this power button package; thank you.

    My J1900 is running just fine for about five days.

    Now I see the difference between a virtual and a barebone installation. It is running much smoother now.

     

    With surveillance station package and 5 IP camera's it's using between 4 - 8 % of CPU power.

    I am monitoring it with Grafana, telegraf and Influxdb.

     

     

    xpenology_grafana_1.jpg

    • Like 1
  16. The v1.04 was made to override the problems with the PCI devices. I thought I had red that somewhere here.

     

    Yes, I am running xpe on a J1900ITX and it's working. (so far so good)

    I was used to install xpe in a KVM on Proxmox, but this is much faster than I thought it would be.

     

    I guess you have something to do. ;)

    • Like 1
  17. Thanks for your reply.

     

    I have figured it out. My installed SSD had several (linux) partitions. When a new DSM has to be installed DSM don't likes a harddrive with partitions.

    So I downloaded a live linux distribution, made a bootable USB en deleted all the "old" partitions so I had a clean disk.

     

    The boot process with V1.04b loader works now fine and latest DSM is installed.

     

    • Like 1
  18. Hi all,

     

    I have a spare J1900 PC to try a baremetal installation of Xpenology. I have followed the tutorial and I can find the diskstation in my LAN. I use the DS918+ loader  1.04b.

    When I want to install DSM I get a warning that "disk 1" will be formated. When I click "ok"  the installation will  begin but end soon with the message: "unable to format the disk, error 35".

     

    I have one Sata SSD on the first SATA3 port and is recognized in the BIOS.

     

    I updated the grub.cfg with the VID/PID of the USB device and changed the MAC and SN of the installation.

     

    Am I missing something here?

     

     

  19. Hi all,

     

    I make use of the Surveillance station package on xpenology. After some random time the CPU usage will double the utilization. It goes from 20% to 40%.

    The only thing I can do about it, is to restart the xpenology and it goes back to normal.

     

    xpenology is running on KVM in Proxmox 5.2.8.

     

    Anyone else with this issue and knows a solution?

×
×
  • Create New...