TinyCore RedPill Loader (TCRP)

[Peter Suh]

8 hours ago, Hackaro1 said:

 

@Peter Suh

SOLVED !!!! I report here so that it can be helpful for others. 

 

The problem raised when my USB stick failed, at that point my NAS was on DSM 7.2 64570 Update 1. When I rebuilt the loader with your TCRP loader (v. 1.0.1.0) and built it with old v. 64570 I didn’t know you was referring to the Update 3 version of that DSM. That probably causes the problem of Storage Manger that was not opening. 

Infact I was able to replicate the problem with a new, fresh disk. I built a new USB stick from scratch and I went for DSM 64570 in your TCRP builder, and even with a new HDD and new install the Storage Manger didn’t open up. 

Now I have upgraded everything to DSM 7.2.1 Update 3 (69057-3) and on my RAID array everything is running smoothly as before! So thanks! 😊

 

Just 1 more question for you: my NAS have two 1 Gb ports, both of them with original Syno MAC addresses, and both QC and DDNS services are working like a charm; as I don’t want to mess up with those services but I’d like to add a 10Gb eth card to speed up things over my home network, I’m thinking to buy an original Synology card, like Synology E10G18-T1, in order to have an “official” MAC address to deal with. The problem may be that DS920+ doesn’t support any ethernet card expansion so do you think it can work? Will I need to rebuild the USB stick? 

 

XPENOLOGY is a program that hacks the limitations of the specifications of the genuine product.

What you need to consider for the Ethernet card part is devices other than the genuine support list.
There just isn't enough verified feedback data for the Ethernet modules (recompiled and manufactured by Redpill) used by each chipset.
Sometimes, if the Ethernet module provided by Redpill does not work or is unstable,

it may be removed and replaced with the original module originally provided by Synology (called a vanilla module).

So far, the modules below have been such cases.
In the case of DS920+ Gemini Lake, there was no such case.
https://github.com/PeterSuh-Q3/arpl-modules/tree/main/vanilla/geminilake-4.4.302

In the case of DS3622xs+ Broadwell NK, the ixgbe and Mellanox modules are replaced with Vanillo modules and passed.
https://github.com/PeterSuh-Q3/arpl-modules/tree/main/vanilla/broadwellnk-4.4.302

In the case of AQC107, it seems necessary to check in detail whether the module used is aqc111.ko or whether the Vanillo module works well enough.

 

------------------------------------------------------------------------------------------------

Also, here is additional information about the detailed chipset of Synology's genuine Ethernet.

E10G17-F2 : Mellanox ConnectX-3 Pro CX312B
E10G18-T1 : Aquantia AQC107-B1
E10G18-T2: Marvell QL41102 (MAC) + BCM84886 (PHY)
E10M20-T1: Aquantia AQC107 (built-in PLX PCIe switch and 2EA M.2 slots)
E10G21-F2: Intel X710
E25G21-F2: Intel XXV710 (estimated)
E10G22-T1-Mini : Aquantia AQC107
E10G30-T2: Broadcom BCM57416 (estimated)
E25G30-F2: Broadcom BCM57414 (estimated)

Model name - If it starts with T, it is 10GBASE-T (RJ45), if it starts with F, it is SFP+ or SFP28, and the last number is the number of ports.

E10G18-T1, E10M20-T1, and E10G22-T1-Mini also support 2.5G/5G standard IEEE 802.bz.

[Prutzer]

I am running the great loader of Peter Suh on a I3 7100 without any problems. It runs fine. V 1.0.1.0.

But I would like to run it with a more power efficiënt mainbord: Gigabyte GA-N3150N-D3V.

 

I build a new loader and everything went well. (DVA1622)

Starting takes a long time. It is waiting a long time before counting down. The system is pingable.

After it starts counting down I got this screen and after that the screen is going blank and the system is not pingable anymore.

 

Is this system to slow? Is the Asrock N100M supported?
Am I doing something wrong?

 

 

The mainbord GA-N3150N-D3V is a bit a weird mainbord. I have more issues booting from USB with board. Some USB stick will work, some won't.

 

 

 

 

Edited December 23, 2023 by Prutzer

[Prutzer]

Log:

 

 

[siri_uk]

 
Please upload a screen capture showing 4 windows before the loader build is completed.
 

Solved now I wrote the image with win32diskimager instead of balenaetcher, works fine now, thank you for all your hard work

Sent from my SM-S918B using Tapatalk

[Peter Suh]

33 minutes ago, siri_uk said:

Solved now I wrote the image with win32diskimager instead of balenaetcher, works fine now, thank you for all your hard work

Sent from my SM-S918B using Tapatalk
 

 

The output of balenaEtcher behaved strangely. I experienced it too.

Edited December 26, 2023 by Peter Suh

[gericb]

2 hours ago, siri_uk said:

Solved now I wrote the image with win32diskimager instead of balenaetcher, works fine now, thank you for all your hard work

Sent from my SM-S918B using Tapatalk
 

Personally, I would not even use win32diskimager.  It's only v1.0.0, no updates since 1/2023 might as well be a deprecated product.  Rufus on the other hand, I've never had a problem with regarding the various loaders here, or others like Windows, Linux, ISO's etc.  It's updated regular, always being improved, enhanced.  BalenaEtcher was basically filling that void for macOS of a general USB creation tool, but it's more of a side product to the rest of what the company offers, though at least they are constantly trying to improve it with new releases, it's often has some quirky issues.

 

https://rufus.ie/en/

[siri_uk]

3 hours ago, gericb said:

Personally, I would not even use win32diskimager.  It's only v1.0.0, no updates since 1/2023 might as well be a deprecated product.  Rufus on the other hand, I've never had a problem with regarding the various loaders here, or others like Windows, Linux, ISO's etc.  It's updated regular, always being improved, enhanced.  BalenaEtcher was basically filling that void for macOS of a general USB creation tool, but it's more of a side product to the rest of what the company offers, though at least they are constantly trying to improve it with new releases, it's often has some quirky issues.

 

https://rufus.ie/en/

Thanks for the information I'll move onto rufus

[Prutzer]

On 12/23/2023 at 12:38 PM, Prutzer said:

I am running the great loader of Peter Suh on a I3 7100 without any problems. It runs fine. V 1.0.1.0.

But I would like to run it with a more power efficiënt mainbord: Gigabyte GA-N3150N-D3V.

 

I build a new loader and everything went well. (DVA1622)

Starting takes a long time. It is waiting a long time before counting down. The system is pingable.

After it starts counting down I got this screen and after that the screen is going blank and the system is not pingable anymore.

On 12/23/2023 at 12:38 PM, Prutzer said:

 

 

Is there anyone who knows why I can't boot TCRP with this mainbord? It was running the older loaders for years.

[DomZ]

@Peter Suh what's the process to change loader synology model, I was in DS923+ and would like to move on SA6400 but I failed

I did this :

• Reboot to TCRP
• Choose SA6400 model into model list
• Rebuild loader
• Reboot
• Wait a few minutes
• and try to find the synology with web assistant and synology assistant, Also tried with the link behind the QR code, but synology is never found

 

Did I miss something ?

Thanks

 

EDIT:

Switch back to DS923+, and everything is back ... then I supose my process is correct but SA6400 synology model doesn't works with my configuration ...

Edited December 30, 2023 by DomZ

[Peter Suh]

3 hours ago, DomZ said:

@Peter Suh what's the process to change loader synology model, I was in DS923+ and would like to move on SA6400 but I failed

I did this :

• Reboot to TCRP
• Choose SA6400 model into model list
• Rebuild loader
• Reboot
• Wait a few minutes
• and try to find the synology with web assistant and synology assistant, Also tried with the link behind the QR code, but synology is never found

 

Did I miss something ?

Thanks

 

EDIT:

Switch back to DS923+, and everything is back ... then I supose my process is correct but SA6400 synology model doesn't works with my configuration ...

 

Please tell us all your detailed hardware specifications.
The NIC part is especially important.

[DomZ]

Il y a 6 heures, Peter Suh a dit :

 

Please tell us all your detailed hardware specifications.
The NIC part is especially important.

 

First, this an esxi 8.0.1 VM with 32 go ang 8 cores.

Server is HP ML350 Gen110 with :

• 2x Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz (10 cores / 20 threads each)
• 64 go of RAM DDR4
• NIC is HPE Ethernet 1gb 4 port 369i
• RAID  Controller : HPE Smart Array P816i-a SR Gen10

 

Thanks

 

EDIT : NIC seems to be : Intel X722 series Modular LAN on motherboard (mLOM) network adapters

 

Edited December 31, 2023 by DomZ

[Peter Suh]

 
First, this an esxi 8.0.1 VM with 32 go ang 8 cores.
Server is HP ML350 Gen110 with :

• 2x Intel® Xeon® Silver 4114 CPU @ 2.20GHz (10 cores / 20 threads each)
• 64 go of RAM DDR4
• NIC is HPE Ethernet 1gb 4 port 369i
• RAID  Controller : HPE Smart Array P816i-a SR Gen10

 
Thanks
 
EDIT : NIC seems to be : Intel X722 series Modular LAN on motherboard (mLOM) network adapters
 

SA6400 is not yet stable in the sata dom ata loader method used in VM. It is highly likely that a kernel panic occurred. There are cases where the panic goes through with a few reboots of the loader, but this is random. The recommended method for VMs is to recognize the loader like a USB. Proxmox has a way to do this, but it is not easy to do so with esxi. I think you could try booting the VM by passing through the loader on the actual USB stick.


Sent from my iPhone using Tapatalk

[Peter Suh]

5 hours ago, DomZ said:

 

First, this an esxi 8.0.1 VM with 32 go ang 8 cores.

Server is HP ML350 Gen110 with :

• 2x Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz (10 cores / 20 threads each)
• 64 go of RAM DDR4
• NIC is HPE Ethernet 1gb 4 port 369i
• RAID  Controller : HPE Smart Array P816i-a SR Gen10

 

Thanks

 

EDIT : NIC seems to be : Intel X722 series Modular LAN on motherboard (mLOM) network adapters

 

 

https://www.starwindsoftware.com/blog/usb-support-in-vmware-vsphere

[DomZ]

Il y a 4 heures, Peter Suh a dit :

 

SA6400 is not yet stable in the sata dom ata loader method used in VM. It is highly likely that a kernel panic occurred. There are cases where the panic goes through with a few reboots of the loader, but this is random. The recommended method for VMs is to recognize the loader like a USB. Proxmox has a way to do this, but it is not easy to do so with esxi. I think you could try booting the VM by passing through the loader on the actual USB stick.


Sent from my iPhone using Tapatalk

 

 

Thanks @Peter Suh, I'm planning to switch from ESXI to Proxmox, then I will retry and may be ask you more details about this

About ESXI to Proxmox migration, I suppose there is nothing special to do after vmdk backup and image convertion ? 

Edited December 31, 2023 by DomZ

[Peter Suh]

9 minutes ago, DomZ said:

 

Thanks @Peter Suh, I'm planning to switch from ESXI to Proxmox, then I will retry and may be ask you more details about this

About ESXI to Proxmox migration, I suppose there is nothing special to do after vmdk backup and image convertion ? 

 

Remember this...

 

proxmox uses img files.

Please note only the information to be retrieved from the existing vmdk (SN / MAC ADDR) and enter it again.

 

PVE installation steps
1. Install Synology SA6400
This article uses a simulated USB disk to boot Synology.

Special configuration required for a virtual machine that emulates a USB flash drive:
1. Select the model as q35
2. Enable SeaBIOS, or select UEFI and enter the BIOS to turn off secure boot.
3. Cancel all Boot Orders under Options
4. Add the following line in /etc/pve/qemu-server/virtual machine id.conf (note to modify the image path):

 

args: -device 'nec-usb-xhci,id=usb-bus0,multifunction=on' -drive 'file=/path/to/tcrp.img,media=disk,format=raw,if=none,id=drive -disk-bootloader' -device 'usb-storage,bus=usb-bus0.0,port=1,drive=drive-disk-bootloader,id=usb-disk-bootloader,bootindex=999,removable=on'

Edited December 31, 2023 by Peter Suh

[DomZ]

il y a 2 minutes, Peter Suh a dit :

 

Remember this...

 

PVE installation steps
1. Install Synology SA6400
This article uses a simulated USB disk to boot Synology.

Special configuration required for a virtual machine that emulates a USB flash drive:
1. Select the model as q35
2. Enable SeaBIOS, or select UEFI and enter the BIOS to turn off secure boot.
3. Cancel all Boot Orders under Options
4. Add the following line in /etc/pve/qemu-server/virtual machine id.conf (note to modify the image path):

 

args: -device 'nec-usb-xhci,id=usb-bus0,multifunction=on' -drive 'file=/path/to/tcrp.img,media=disk,format=raw,if=none,id=drive -disk-bootloader' -device 'usb-storage,bus=usb-bus0.0,port=1,drive=drive-disk-bootloader,id=usb-disk-bootloader,bootindex=999,removable=on'

 

RIght now, my xpenology VM has been installed using a vmdk image that was already set, and loader is embed in the VM image (I suppose).

I have used this youtube toturial :

 

 

What you propose is to redo the image from scratch right ?

[Peter Suh]

3 minutes ago, DomZ said:

What you propose is to redo the image from scratch right ?

 

Yes, that's right.

[DomZ]

il y a 11 minutes, Peter Suh a dit :

 

Yes, that's right.

 

Got it, I will create another xpenology VM to validate the process, thanks.

[gericb]

7 hours ago, DomZ said:

 

Thanks @Peter Suh, I'm planning to switch from ESXI to Proxmox, then I will retry and may be ask you more details about this

About ESXI to Proxmox migration, I suppose there is nothing special to do after vmdk backup and image convertion ? 

So is the general notion that ESXI is more stable, reliable, robust, functional, easier setup with lessor issues than Proxmox?

[Peter Suh]

21 hours ago, DomZ said:

 

Got it, I will create another xpenology VM to validate the process, thanks.

 

There was a typo.

 

not

id=drive -disk-bootloader

 

this is correct.
id=drive-disk-bootloader

 

They should not be spaced apart.

 

args: -device 'nec-usb-xhci,id=usb-bus0,multifunction=on' -drive 'file=/var/lib/vz/template/iso/tcrp.img,media=disk,format=raw,if=none,id=drive-disk-bootloader' -device 'usb-storage,bus=usb-bus0.0,port=1,drive=drive-disk-bootloader,id=usb-disk-bootloader,bootindex=999,removable=on'

Edited January 1 by Peter Suh

[Dittel]

Hi colleagues!

I repeat my post from another topic shortly: I had normal working DS3622xs+ version 7.1.1 based on proxmox (discussed here about a half of year ago). It was not really used so I tried to make some experiments like system update. Unfortunately my hands runs more quickly than hands  so I crashed the system and can't return to normal state. So I tried to re-build it from scratch and obtained non-working system: it is repeatedly informed me "the system is restorable" ("we see you move disks from another system here so the system should be restored so please push the button") but after reboot everything remains in the same state.

I spent a lot of hours to understand what I can do but I can restore the DSM just after re-build it to completely new system DS918+ (instead of DS3622xs+ one). I could even start to DSM version 7.2.0 here. It is the goal I can't obtain with DS3622xs+.

So - is it possible some mistakes are present in the loader?

PS.

one problem is still now present - loader DS3622xsp-7.2.0-64570 is not built due to "not found". DS918+ is built OK with this version number.

PS2

Same symptoms are present with another colleagues used baremetal - so it is not the problem of proxmox hypervisor.

[Tibag]

Hi all!

 

I am currently running DSM 7.1.1-42962 Update 5 with 0.9.4.6 as a DS3622xs+ via ESXI. 

 

I first through about manually upgrading with the package for DSM 7.2-64561, then doing the others, but I can't find it on http://download.synology.com/download/DSM/criticalupdate/update_pack/.

 

What's your recommended path to DSM 7.2.1 69057-Update 3? 

Edited January 9 by Tibag

[gericb]

On 12/14/2023 at 9:03 AM, Peter Suh said:

 

Torvalds' script is mentioned in the article linked above by @vbz14216
In the end, it was made to work with r8125.

https://forums.freebsd.org/threads/patching-igc-driver-to-support-killer-e3100g.86080/#post-577569

https://github.com/torvalds/linux/blob/master/drivers/net/ethernet/realtek/r8169_main.c#L165

This script is also applied to my mshell module repo.
However, there is no such part in kernel 4.
It is applied to kernel 5, which only uses SA6400.

https://github.com/PeterSuh-Q3/arpl-modules/blob/main/src/5.x/drivers/net/ethernet/realtek/r8169_main.c#L177

It seems that the already distributed SA6400 supports KillerLAN E3100G.
Want to be the first to try the SA6400 model?

Could you please forget about running on Intel i225 like Windows OS and look forward to Killer LAN running on r8125?

So far, so good! I can report that SA6400 with your current Loader w/Friend, it was effortless and largely flawless on the test XPS 8950.  The only thing I could report at this time, was the screen resolution when booting into Friend, was lower than optimal, so I needed to resize and arrange the 4 windows a bit, but that's trivial.  It recognized all 4 SSD installed, as well at the 2 x 256GB NVMe SSD installed for possible R/W cache options testing.  Created a RAID F1 smoothly.  Singular network connection, recognized as predicted and performing well thus far. 

[waspsoton]

@Peter Suh i am trying do a build but i keep getting a checksum error.  any ideas?

[SuperJack]

Here's the problem. Changed flash drives, rebuilt the bootloader several times, but the problem persists. After the command: ./boot.sh booting goes normally. I will be grateful for any help.

Edited January 24 by SuperJack