TinyCore RedPill Loader (TCRP)

[-iliya-]

10 часов назад, rayy сказал:

Can you please tell me which user_config file do you edit?  I would like to know its location on the loader USB.  Thanks.

I edit the file when creating the bootloader and yes every time I recreate it

[pocopico]

1 hour ago, -iliya- said:

I edit the file when creating the bootloader and yes every time I recreate it

 

Last session user_config.json is always on the third partition of the loader under lastsession folder. Last extensions used are also listed there.

[Kaneske]

Are there any update reportings, since 7.1.1 Update 1 is out?

 

My Rig is an ESXi VM with 2x 2008 IT-Mode HBA and XGBE Intel 10Gbe NIC

Actually I’m on 0.9.0.2 TCRP DEV, DSM 7.1.1-42962

[pocopico]

4 hours ago, Kaneske said:

Are there any update reportings, since 7.1.1 Update 1 is out?

 

My Rig is an ESXi VM with 2x 2008 IT-Mode HBA and XGBE Intel 10Gbe NIC

Actually I’m on 0.9.0.2 TCRP DEV, DSM 7.1.1-42962

 

Yes currently i'm testing the TCRP friend. Its a buildroot solution like Jumkey/fabio does to patch at boot. I'm almost done but i need to create a function so that all people that have already their systems without tcrp friend can bring that over to their image. It should be soon ready *

[pocopico]

19 hours ago, pocopico said:

 

Yes currently i'm testing the TCRP friend. Its a buildroot solution like Jumkey/fabio does to patch at boot. I'm almost done but i need to create a function so that all people that have already their systems without tcrp friend can bring that over to their image. It should be soon ready *

 

I need a tester for TCRP friend. Ideally one with a running system that needs to upgrade to TCRP Friend. If anyone is interested PM me. 

[phone guy]

2 hours ago, pocopico said:

 

I need a tester for TCRP friend. Ideally one with a running system that needs to upgrade to TCRP Friend. If anyone is interested PM me. 

I tried to PM you, says pocopico can not receive messages ?

I have a baremetal nas now running tcrp 08.xx dsm7.1u4 ds918

and a proxmox server, running 8 drives on lsi hba card, tcrp 08.xx dsm7.1u4

 

PM me if I can help.

[pocopico]

11 minutes ago, phone guy said:

I tried to PM you, says pocopico can not receive messages ?

I have a baremetal nas now running tcrp 08.xx dsm7.1u4 ds918

and a proxmox server, running 8 drives on lsi hba card, tcrp 08.xx dsm7.1u4

 

PM me if I can help.

 

Yes, my mailbox is full  , i'll keep you posted 

[Itay1778]

Is there an ETA when v9 is out? Or the possibility to update properly from 7.1-42661 Update 4 to 7.1.1-42962 update 1?
 On DS1918+ and DS3622xs+

[dimakv2014]

12 часа назад, pocopico сказал:

 

I need a tester for TCRP friend. Ideally one with a running system that needs to upgrade to TCRP Friend. If anyone is interested PM me. 

I have spare baremetal with N2807 running DS920+ DSM7.1u1 which doesn't have FMA3 and still runs it, if you will tell what to do, can do the testing. Moreover, If remote access needed, I can provide it.

Edited September 23, 2022 by dimakv2014

[lord_kimbou]

A rapid question, I need a new Hba card, 9500-16i it's supported by mpt3sas driver or only 9300/9400?? That's all!! Thanks!!

[jcgechevarria]

On 5/12/2022 at 2:12 PM, Peter Suh said:

 

I searched all the way through that topic.
The model mentioned in this topic is the GA-H87M-HD3.
It is different from the model we have.
Just because the models are similar, it cannot be concluded that the same phenomenon will occur.

I has that issue too, on three Gigabyte motherboards, H87 and H97 chipsets, solved changing  "time setting" to manual on DSM Regional settings.

[Dreadnought]

On 9/23/2022 at 4:17 PM, pocopico said:

 

Yes, my mailbox is full  , i'll keep you posted 

@pocopico

For me TCRP friend (7.0.1) is working great but I have to delete line 2744 in the rploader.sh ('cd /home/tc/redpill-load')

I was also to Update to 7.0.1 U5 --> 7.1.0 -> 7.1.1 without any problems!

 

I found one bug:

I have to execute ./rploader.sh bringfriend twice otherwise the TinyCore Friend boot entry won't be created.

Edited September 25, 2022 by Dreadnought
Bug report

[Kaneske]

@pocopico I can test too if you like.

PM me then…

 

Any news about 7.1.1-1 so far with TCRP?

My Promox Virtual (Disk too) Setup did clean online install to 7.1.1-1 flawless (ARPL)…but I’m afraid to get my TCRP main XPEN with the HBAs on ESXi updated since there’s much Data on it.

Edited September 25, 2022 by Kaneske

[likeme]

Compiling "geminilake-7.1.1-42962" with V0.9.2.3, "CMOS Checksum bad" which has existed since history https://github.com/fbelavenuto/arpl/issues/174 problem still exists. The latest ARPL has been solved, I hope pocopico also work hard to solve this historical problem. Thank you.

[pocopico]

5 hours ago, likeme said:

Compiling "geminilake-7.1.1-42962" with V0.9.2.3, "CMOS Checksum bad" which has existed since history https://github.com/fbelavenuto/arpl/issues/174 problem still exists. The latest ARPL has been solved, I hope pocopico also work hard to solve this historical problem. Thank you.

 

Yes issue will be solved with the change of the kernel patching methodology provided by @fbelavenuto 

[Polanskiman]

@pocopico I am not sure if this has been asked before but there are 171 pages so I'll ask again.

 

I need to understand these commands and when and why they need to be run. I see in the DSM Updates Reporting thread that some people run some of these commands and some other people omit some of them yet they seem to be able to update DSM regardless.  So when is it necessary/compulsory to run each of these commands ?

I have obviously read @flyride tutorial and gone through your Github repo, but other than the quick instructions during install I don't see thorough explanation on using these commands. Not that it requires a novel but I'd interested to understand the rational.

 

The reason I ask is also because yesterday I upgraded from DSM 6.1 to 7.1. While doing so I followed Flyride's tutorial and he states that both the update and fullupgrade commands are to be run. The update command asked if I wanted to update to 0.9.2.5, however isn't that still in development? I also didn't run the fullupgrade yet I was able to proceed and install DSM 7.1.

 

Anyhow, you get the point. If you could shed some light on this it would be appreciated and apologies if you already explained all of it. Perhaps there is already some resource explaining all this?

 

./rploader.sh update

 

./rploader.sh fullupgrade

 

./rploader.sh postupdate

 

./rploader.sh backup

 

./rploader.sh backuploader

 

 

Thank you.

[pocopico]

On 9/28/2022 at 7:32 PM, Polanskiman said:

@pocopico I am not sure if this has been asked before but there are 171 pages so I'll ask again.

 

I need to understand these commands and when and why they need to be run. I see in the DSM Updates Reporting thread that some people run some of these commands and some other people omit some of them yet they seem to be able to update DSM regardless.  So when is it necessary/compulsory to run each of these commands ?

I have obviously read @flyride tutorial and gone through your Github repo, but other than the quick instructions during install I don't see thorough explanation on using these commands. Not that it requires a novel but I'd interested to understand the rational.

 

The reason I ask is also because yesterday I upgraded from DSM 6.1 to 7.1. While doing so I followed Flyride's tutorial and he states that both the update and fullupgrade commands are to be run. The update command asked if I wanted to update to 0.9.2.5, however isn't that still in development? I also didn't run the fullupgrade yet I was able to proceed and install DSM 7.1.

 

Anyhow, you get the point. If you could shed some light on this it would be appreciated and apologies if you already explained all of it. Perhaps there is already some resource explaining all this?

 

./rploader.sh update

 

./rploader.sh fullupgrade

 

./rploader.sh postupdate

 

./rploader.sh backup

 

./rploader.sh backuploader

 

 

Thank you.

Hi @Polanskiman, thanks for your very valid comments.

 

The previously develop repo moved to main and the previously main repo moved to stable. There are a lot of flags to set and i plan to simplify this a bit in the near future.

 

A more detailed info can be found if you run : 

 

./rploader.sh help 

 

Basically if you start from a previously installed system there are many things you can do to upgrade to latest.

 

A. Recently, i introduced the TCRP Friend which is a buildroot system, that assists with an on the fly patching after an upgade. So the postupdate process should not be required after bringing the TCRP Friend over to your loader.

 

To bring over the TCRP Friend you can run

 

./rploader.sh update
./rploader.sh fullupdate
./rploader.sh bringfriend

 

Then the default grub boot selection will be TCRP friend that tries to solve as many drawbacks we have as we can. 

 

Thats still under development but i have tested all platforms and they should have no issues.

 

B. Perform the upgade using the DSM GUI and then perform the postupdate command to update the ramdisk with the latest files.

This as i said will soon be obsolete as we now have the TCRP Friend, but there might be cases that we will need it in the future so i plan to leave this command available.

[Peter Suh]

@pocopico

I continue to test the operation of the NVMe cache in the built-in M.2 slot today.
In the previously reported ASUS IOT H310i-IM-A R2.0 Intel MOBO, the TCRP friend did not work with NVMe,

so this time only NVMe was removed and then the TCRP friend was activated again.

As expected, the TCRP friend boots normally.

 

Another test is ASUS TUF B450M-PLUS GAMING Ryzen MOBO,
With NVMe cache installed, the TCRP friend wouldn't boot and it was possible with TCRP.
Even after clearing the NVMe cache in the same way, TCRP friend won't boot.

I wanted to find another cause, but since it is baremetal, there is no way to monitor the log without a serial port.
This MOBO provides a COM1 port, but as is the case with all MOBOs these days, it does not come with a serial port bracket.

The serial port bracket I bought today seems to be delivered in 2 days. At that time, I will monitor the log of this MOBO once more.

 

Activating the NVMe cache of the built-in M.2 in the TCRP friend seems to be the issue.

I am doing my best to resolve this issue.
I think you said that you can only test in a VM environment and not on baremetal.
So, if there is any way I can help further, please feel free to ask.

[pocopico]

4 minutes ago, Peter Suh said:

@pocopico

I continue to test the operation of the NVMe cache in the built-in M.2 slot today.
In the previously reported ASUS IOT H310i-IM-A R2.0 Intel MOBO, the TCRP friend did not work with NVMe,

so this time only NVMe was removed and then the TCRP friend was activated again.

As expected, the TCRP friend boots normally.

 

Another test is ASUS TUF B450M-PLUS GAMING Ryzen MOBO,
With NVMe cache installed, the TCRP friend wouldn't boot and it was possible with TCRP.
Even after clearing the NVMe cache in the same way, TCRP friend won't boot.

I wanted to find another cause, but since it is baremetal, there is no way to monitor the log without a serial port.
This MOBO provides a COM1 port, but as is the case with all MOBOs these days, it does not come with a serial port bracket.

The serial port bracket I bought today seems to be delivered in 2 days. At that time, I will monitor the log of this MOBO once more.

 

Activating the NVMe cache of the built-in M.2 in the TCRP friend seems to be the issue.

I am doing my best to resolve this issue.
I think you said that you can only test in a VM environment and not on baremetal.
So, if there is any way I can help further, please feel free to ask.

 

I have several baremetal systems but most of them are on "production" . I dont have an NVMe baremetal option available but i can test NVMe on VMware as well. 

 

[Peter Suh]

58 minutes ago, pocopico said:

 

I have several baremetal systems but most of them are on "production" . I dont have an NVMe baremetal option available but i can test NVMe on VMware as well. 

 

 

Virtualbox also supports PCIE type NVMe.
So I also tried to enable NVMe on the VM.
As you can see, in the same way as in DS3622xs+ baremetal.

I modified /etc.defaults/extensionPorts but it didn't work.
I don't think I've ever heard any reports of successful NVMe caching in VMs.
Once it can be activated from DSM, it can be extended to the test of TCRP friends, but it seems impossible at the moment.
I think we should focus on baremetal.

 

[cferra]

28 minutes ago, Peter Suh said:

 

Virtualbox also supports PCIE type NVMe.
So I also tried to enable NVMe on the VM.
As you can see, in the same way as in DS3622xs+ baremetal.

I modified /etc.defaults/extensionPorts but it didn't work.
I don't think I've ever heard any reports of successful NVMe caching in VMs.
Once it can be activated from DSM, it can be extended to the test of TCRP friends, but it seems impossible at the moment.
I think we should focus on baremetal.

 

Yeah it works on Baremetal. I’ve got that going now. 

[pocopico]

 

On 9/28/2022 at 9:36 PM, Peter Suh said:

I modified /etc.defaults/extensionPorts but it didn't work.
I don't think I've ever heard any reports of successful NVMe caching in VMs.
Once it can be activated from DSM, it can be extended to the test of TCRP friend

 

I've tested on VMware and NVMe is possible with TCRP Friend having no issues.

 

1. /etc/extensionPorts and /etc.defaults/extensionPorts should be edited

 

 dmesg |grep nvme
[    1.964684] nvme nvme0: pci function 0000:0b:00.0

cat /etc/extensionPorts
[pci]
pci1="0000:0b:00.0"
pci2="0000:00:11.0"

cat /etc.defaults/extensionPorts
[pci]
pci1="0000:0b:00.0"
pci2="0000:00:11.0"

 

 

 

I can even use two NVMe disks for read-write cache 

 

[Kaneske]

Hmm testing on Promox VM:

 

TCRP built, then bringfriend destroys me grub, so that it only will boot in rescue…

 

Any clues @pocopico?

[pocopico]

1 minute ago, Kaneske said:

Hmm testing on Promox VM:

 

TCRP built, then bringfriend destroys me grub, so that it only will boot in rescue…

 

Any clues @pocopico?

 

Thats new ... Do you maybe have two loader disks attached on this system ? While on TCRP you can recover the GRUB contents if the partitions are OK.

 

to boot from rescue now type the following ...

set root=(hd0,msdos3)
linux /vmlinuz64 loglevel=3 cde waitusb=5 vga=791
initrd /corepure64.gz
boot

 

[Peter Suh]

12 hours ago, cferra said:

Yeah it works on Baremetal. I’ve got that going now. 

 

As a result of my testing, there were two cases where NMVe cache worked in TCRP friends even on baremetal.

 

The NVMe cache of the onboard internal M.2 slot did not work,

 

In the case of using the M.2 PCIE extension adapter, it worked for me as well.

 

May I know which of the two is the case?