Jump to content
XPEnology Community

Automated RedPill Loader (ARPL)


fbelavenuto

Recommended Posts

Hi all!
First of all, I want to thank you for a wonderful tool!
Configuration: ASrock 4105, 16Gb RAM, Marvel PIC-e to 2 SATA, 2 HDD 1 Gb, 2 HDD 10Tb, 2 SSD 256 Gb. All pairs of HDDs in RAID 1. Basic working configuration with ~7 Tb of data.
Made a migration from the old self-made bootloader DSM 7.0 918+ to the latest version of DSM 7.1 920+ (new alpha 7). Everything went smoothly! The order of my actions:
1. Disconnected one disk from each pair (just in case);
2. Booted from a new flash drive, compiled the bootloader with standard parameters (real s / n);
3. Booted into the bootloader, DSM applied the migration to 920+;
4. Passed the migration as in real Syno, two reboots, all in automatic mode;
5. After all the checks, I connected the remaining disks and restored the RAID (it took less than 1 day).

DSM works quickly and without problems. RAID recovery very fast. Previously, recovery took up to 3 days.
There is only one problem, it is VMM. After the update, something happened with performance of virtual machines. Win10 works as a lagy video, it is not possible to work through RDP. A similar situation with Windows Server 22. It slows down terribly. More than 5 seconds may elapse between pressing the interface button and the response! Previously, I even sometimes forgot that I was working on a virtual machine via RDP. Virtual machines run with 4 VCPUs and 8 Gb RAM. Home Assistant OS takes almost 10 minutes to load.
Someone faced similar? Any guesses what could happen? Only the flash drive (Trancend 32Gb USB 3) changed in the configuration...

Edited by Drones
Link to comment
Share on other sites

Il y a 2 heures, SanKen a dit :

How to install addons?

https://github.com/fbelavenuto/arpl-modules/blob/main/geminilake-4.4.180/button.ko 

This is probatly working on PWRButtom.

Hi,

This module and many others are already integrated into the system by arpl.

First update from menu.sh  "update arpl" and "Update addons",  then check that addon "acpid" is selected before generate the build loader.

And It must works

Link to comment
Share on other sites

Hi Guys,

 

I have an issue as follows

Device Baremetal thinkcentre m910 i7, 32gb ram. 

Installing from scratch as just building and testing the nas as of yet.

TCRP has worked with DSM7.0.1

 

So ARPL 0.4 Alpha 7 burned to flashdrive

boot and configure loader DS918+, DSM 7.1.1. 42962 and standard options, no addons 

boot, i can connect to the NAS install using the 42962 pat file from Syno site.

it installs reboots and first steps name, admin ... ok

i have 1 ssd in the nas created a volume.After a reboot of nas i am just at the welcome screen to install the nas again.

Any idea?

 

Edit:

did install again to see what happens. let the 10 minutes to pass and then Error "something went wrong" Connect again button

then welcome .. install 

So i am in install loop

 

Thank you very much for help

 

Edited by norberto_
Link to comment
Share on other sites

9 hours ago, Drones said:

Hi all!
First of all, I want to thank you for a wonderful tool!
Configuration: ASrock 4105, 16Gb RAM, Marvel PIC-e to 2 SATA, 2 HDD 1 Gb, 2 HDD 10Tb, 2 SSD 256 Gb. All pairs of HDDs in RAID 1. Basic working configuration with ~7 Tb of data.
Made a migration from the old self-made bootloader DSM 7.0 918+ to the latest version of DSM 7.1 920+ (new alpha 7). Everything went smoothly! The order of my actions:
1. Disconnected one disk from each pair (just in case);
2. Booted from a new flash drive, compiled the bootloader with standard parameters (real s / n);
3. Booted into the bootloader, DSM applied the migration to 920+;
4. Passed the migration as in real Syno, two reboots, all in automatic mode;
5. After all the checks, I connected the remaining disks and restored the RAID (it took less than 1 day).

DSM works quickly and without problems. RAID recovery very fast. Previously, recovery took up to 3 days.
There is only one problem, it is VMM. After the update, something happened with the performance of virtual machines. Win10 works as a lagy video, it is not possible to work through RDP. A similar situation with Windows Server 22. It slows down terribly. More than 5 seconds may elapse between pressing the interface button and the response! Previously, I even sometimes forgot that I was working on a virtual machine via RDP. Virtual machines run with 4 VCPUs and 8 Gb RAM. Home Assistant OS takes almost 10 minutes to load.
Someone faced similar? Any guesses what could happen? Only the flash drive (Trancend 32Gb USB 3) changed in the configuration...

I have like 10 Vms with DSM on my ESXi and 1 BareMetal, I still like 3 vms with DSM 6.2.x with JUN loader, a few with pocopico with DSM 7.0.1 and a few with ARPL with the last 7.1.1-42962. The VMs that have the network driver VMXNET3 (10GB VMware virtual network card) work perfectly with MTU9000 (all my network infrastructure is configured with MTU9000). With the DSM 7.X version using the INTEL E1000E or INTEL E1000 cannot change the MTU from MTU15000 the option shows only from 1100 to 1500. 

 

On the VMs with JUN loader with the DSM 6.2.X the MTU9000 works perfectly with any virtual network card (VMXNET3, E1000 or E1000E).

On BareMetal I have the Intel NIC E1000E driver and cannot use the MTU9000. I guess the DSM 7 has some problems with the intel old card drivers, or the problem is on the side of the loaders in version DSM7.

 

What is the network card on your ASROCK? Remember RDP is a network protocol. 

 

PS: I updated 1 VM with DSM 6.2.x and migrated the DSM changing the JUN for the ARPL and because I use the keep settings, the MTU9000 from DSM6 still remain on the DSM7 using the INTEL virtual network card, but if click to change the MTU the option for MTU9000 is gone. I suspect that the DSM7 has some problems with the old INTEL cards.

 

Is there an option to inject the intel E1000E addon manual?? How is the syntax of the command line for adding addons? I copied the URL of the ADDON but it says that is wrong. I cannot find on the readme or any help file how to inject an addon manually on the menu.sh

Link to comment
Share on other sites

On 7/13/2022 at 11:31 AM, pigr8 said:

 

in ESXi you cannot use a virtual usb, you have to choose a virtual hdd (satadom implementation) or a real usb in passthough, i have to check if it's a viable solution.

Was going trough this thread since I didn't know about this flavor of loader and had to give info for this.

 

In ESXi 7 it is actually possible to have virtual USB in VM and it is possible to boot DSM from it. This is how I run DS918 on it, but there are some limitation like not being able to work with any image which has 3.x kernel.

So here is what works for me:

  1. Add USB 3 controller to VM
  2. UEFI Bios has to be used
    Edit Settings => VM Options => Boot Options
    Frimware: EFI
    Enable UEFI secure boot: uncheck
  3. Edit Settings => VM Options => Advanced in the option Configuration Parameters: Edit Configuration and then add
    usb_xhci:3.present = "TRUE"
    usb_xhci:3.deviceType = "disk"
    usb_xhci:3.fileName = "/path/to/tinycore-redpill-uefi.vmdk"
    usb_xhci:3.readonly = "FALSE"
  4. You need to use UEFI loader immage and correct vid pid for ESXi has to be set in loader:
    vid=0x0e0f
    pid=0x0005
  5. Don't forget to set correct mac and serial, generate with loader serial and mac and then set those in VM config

You also need to convert loader image from raw to vmdk you can use StarWind V2V for this or simply create file tinycore-redpill-uefi.vmdk with following content:

# Disk DescriptorFile
version=1
CID=0d1377b4
parentCID=ffffffff
createType="vmfs"

# Extent description
RW 2097152 VMFS "tinycore-redpill-uefi.img" 0

# The Disk Data Base
#DDB

ddb.adapterType = "lsilogic"
ddb.geometry.cylinders = "131"
ddb.geometry.heads = "255"
ddb.geometry.sectors = "63"
ddb.uuid = "13 9e c3 cf c3 cf 86 0b-a1 90 b9 72 04 b0 91 7e"
ddb.virtualHWVersion = "4"

and put loader image in the same folder as this vmdk, loader has to be named tinycore-redpill-uefi.img

 

When you set it up like this you can use USB boot for DS918.

For reference this is how user config looks like:

{
  "extra_cmdline": {
    "vid": "0x0e0f",
    "pid": "0x0005",
    "sn": "<SERIALNR>",
    "mac1": "<mac1>",
    "mac2": "<mac2>"
  },
  "synoinfo": {
	  "netif_num" : "2"
  },
  "ramdisk_copy": {}
}

 

On Proxmox virtual usb boot can be done by following these steps:

  1. Upload tinycore-redpill-uefi.img to proxmox in folder /var/lib/vz/images/
  2. Set your VM bios to OVMF (UEFI)
  3. Add following to vm config (ie /etc/pve/qemu-server/100.conf)
    args: -device 'qemu-xhci,addr=0x18' -drive 'id=my_usb_disk,file=/var/lib/vz/images/ tinycore-redpill-uefi.img,if=none,format=raw' -device 'usb-storage,id=my_usb_disk,drive=my_usb_disk,bootindex=5'
  4. Correct vid pid has to be set for proxmox
    vid: 0x46f4
    pid: 0x0001
  5. Don't forget to set correct mac and serial, generate with loader serial and mac and then set those in VM config

For reference VM config (100.conf) would look something like this:

args: -device 'qemu-xhci,addr=0x18' -drive 'id=my_usb_disk,file=/var/lib/vz/images/tinycore-redpill-uefi.img,if=none,format=raw' -device 'usb-storage,id=my_usb_disk,drive=my_usb_disk,bootindex=5'
bios: ovmf
boot: order=sata0;net0
cores: 2
memory: 2048
name: test
net0: e1000=00:12:34:56:78:9A,bridge=vmbr0,firewall=1
numa: 0
ostype: l26
sata0: local-lvm:vm-100-disk-0,size=8G
scsihw: virtio-scsi-pci
smbios1: uuid=c17d8446-9f13-4af4-96ea-6bcaed77e5ea
sockets: 1
vmgenid: d89fa586-cb9b-475e-87ac-873d2babb8ba

 

  • Like 1
Link to comment
Share on other sites

On 9/14/2022 at 7:19 AM, SanKen said:

How to install addons?

https://github.com/fbelavenuto/arpl-modules/blob/main/geminilake-4.4.180/button.ko 

This is probatly working on PWRButtom.

 

23 hours ago, buggy25200 said:

Hi,

This module and many others are already integrated into the system by arpl.

First update from menu.sh  "update arpl" and "Update addons",  then check that addon "acpid" is selected before generate the build loader.

And It must works

As @buggy25200 said, this module already exists in the loader, it is not necessary to add anything. On all machines I could test it worked.

I'll investigate further to try to find out why it doesn't work on yours.

Link to comment
Share on other sites

16 hours ago, dferara said:

Is there an option to inject the intel E1000E addon manual?? How is the syntax of the command line for adding addons? I copied the URL of the ADDON but it says that is wrong. I cannot find on the readme or any help file how to inject an addon manually on the menu.sh

I can compile a new version of the driver for you to test. I still haven't created documentation on how to generate an addon!

https://github.com/fbelavenuto/arpl-modules/tree/main/src/e1000e

What DSM model are you using?

  • Like 1
Link to comment
Share on other sites

1 час назад, fbelavenuto сказал:

Too bad! Let's investigate the reason for this problem, maybe something with some driver!

 

Thanks for your reply!
I tried different options for choosing a driver for hard drives and network cards, but the situation does not change. I tried to start a fresh installation, but the "slide show" is already at the stage of installing windows.
I tried another (exactly working) bootable USB flash drive.
It was also noticed that in my virtual windows 10 the "System interrupts" process heavily loads the processor. 

Then I began to think that something might be wrong with my SSDs. I ran a speed test and a full SMART. No problems were noticed.
I will also try to boot today without a ZigBee USB stick based on the CC2531 (in syno: CC2531 ZigBee - Texas Instruments, Inc.). It is used for smart home and forwarded to Home Assistant OS.

Edited by Drones
Link to comment
Share on other sites

4 hours ago, fbelavenuto said:

I don't know what the problem is! Can you try installing an older version like 42218?

Dear fbelavenuto,

 

tried alpha7 full from scratch again set to DS918+, 42218

built loader then booted

the drive in the metal is formatted fully to start that from cratch too.

install screen --> used the 42218 pat from synology site

rebooted loader noticed zImage andRamdisk change so adjusted.

but then it goes again to same install screen as first time.

If i check device status it shows not installed.

the i let it to download latest from synology --> loader recognised so changed builid to 42962

then it still boots to install screen

 

edit:

tested tcrp 0.9 dev version -- it installs the NAS and works as expected -- but i would like to have ARPL :-)

even did nvme patch manually that is also still working in 42962

 

Edited by norberto_
Link to comment
Share on other sites

39 minutes ago, Hostilian said:

Sorry, what are you saying? The tools only work in Docker?

 

I've got DSM 7 working perfectly fine in ESXi 7 (and previously in VMware Workstation Pro).

I'm just wondering about the tools.

 

I know it works i used to have dsm7 on esxi too.. but openvmtools don't work on dsm7 because you can't have root access to shutdown or reboot, the solution is a workaround using openvmtools in a docker container that uses local ssh to make it work.

  • Like 1
Link to comment
Share on other sites

7 часов назад, Drones сказал:

 

Thanks for your reply!
I tried different options for choosing a driver for hard drives and network cards, but the situation does not change. I tried to start a fresh installation, but the "slide show" is already at the stage of installing windows.
I tried another (exactly working) bootable USB flash drive.
It was also noticed that in my virtual windows 10 the "System interrupts" process heavily loads the processor. 

Then I began to think that something might be wrong with my SSDs. I ran a speed test and a full SMART. No problems were noticed.
I will also try to boot today without a ZigBee USB stick based on the CC2531 (in syno: CC2531 ZigBee - Texas Instruments, Inc.). It is used for smart home and forwarded to Home Assistant OS.

Good evening!
I tried booting without CC2531. Nothing changed.
I also tried migrating to 918+. The migration was successful, but the situation has not changed.
I don't know what else to try (

Link to comment
Share on other sites

@fbelavenuto One question, I update to arpl v0.4 alpha 8 from update menu while compile github and crash the boot. I have download older v0.4 alpha 7 to recovery. Is it possible to update from update menu when not is possible to download from github the new version?

 

ARPL v0.4 alpha 8 works well. I am testing now with Proxmox.

Captura de pantalla 2022-09-16 001413.jpg

Edited by apriliars3
Link to comment
Share on other sites

2 hours ago, Drones said:

Good evening!
I tried booting without CC2531. Nothing changed.
I also tried migrating to 918+. The migration was successful, but the situation has not changed.
I don't know what else to try (

Maybe try another loader?  Have you tried with TCRP v0.9 (develop)?

Edited by renegadeBE
Link to comment
Share on other sites

12 hours ago, fbelavenuto said:

I can compile a new version of the driver for you to test. I still haven't created documentation on how to generate an addon!

https://github.com/fbelavenuto/arpl-modules/tree/main/src/e1000e

What DSM model are you using?

 

Hi fbelavenuto!!

 

Here is the list of a few that I have, I have more on VM, but because are repeated configuration I didn't include here on the pictures.

 

You can see that the E1000 and E1000E that i HAVE with DSM 6.2.X JUN loader with the DSM 6 works fine, I forgot to include the picture of the DSM 6 with E1000E. 

The peculiar thing is the VM with DSM 6.2X with E1000E migrated to DSM 7 keep the settings of MTU9000 but if I would try to change, on the options only show 1500.

I have a lot of RAM on my ESXi and a lot of CPU power so I can try to test any special driver you can give me.

I don't know how to include the ADDON, when i click on download from a external addon and put the link you copy on the post it gives me this error.

 

image.png.ac21751a887cddb12cd7b51301f4e691.png

 

List of my VMs and Baremetal:

 

image.thumb.png.d0ae03e1b99fbe30c8a5a417c4cd365e.png

  • Like 1
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...