jun

DSM 6.1.x Loader

Recommended Posts

Hi Jun,

 

It seems load the latest drivers 5.3.5.4

 

[sun Nov 20 15:23:03 2016] Intel(R) Gigabit Ethernet Network Driver - version 5.3.5.4
[sun Nov 20 15:23:03 2016] Copyright (c) 2007-2015 Intel Corporation.
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: irq 141 for MSI/MSI-X
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: irq 142 for MSI/MSI-X
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: added PHC on eth1
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: Intel(R) Gigabit Ethernet Network Connection
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: eth1: (PCIe:2.5GT/s:Width x1)
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0 eth1: MAC: 1c:1b:0d:63:b9:38
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: eth1: PBA No: Unknown
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: LRO is disabled
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)

 

 

I unplug the eth0 and plug into eth1 and eth1 cant get IP from DHCP. Later I change back to eth0.

 

[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: irq 141 for MSI/MSI-X
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: irq 142 for MSI/MSI-X
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: added PHC on eth1
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: Intel(R) Gigabit Ethernet Network Connection
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: eth1: (PCIe:2.5GT/s:Width x1)
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0 eth1: MAC: 1c:1b:0d:63:b9:38
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: eth1: PBA No: Unknown
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: LRO is disabled
[sun Nov 20 15:23:03 2016] igb 0000:03:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)
[sun Nov 20 15:27:32 2016] igb 0000:03:00.0 eth1: igb: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX
[sun Nov 20 15:29:04 2016] igb 0000:03:00.0 eth1: igb: eth1 NIC Link is Down

 

I also tried to recompile the module and copy to /lib/modules and use insmod in /etc/rc.local. Still not working.

 

Below is from my 5.2 which is using older version of igb

[sun Nov 20 15:47:58 2016] Intel(R) Gigabit Ethernet Network Driver - version 5.3.3.2
[sun Nov 20 15:47:58 2016] Copyright (c) 2007-2015 Intel Corporation.
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: irq 138 for MSI/MSI-X
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: irq 139 for MSI/MSI-X
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: added PHC on eth1
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: Intel(R) Gigabit Ethernet Network Connection
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: eth1: (PCIe:2.5GT/s:Width x1)
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0 eth1: MAC: 1c:1b:0d:63:b9:38
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: eth1: PBA No: Unknown
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: LRO is disabled
[sun Nov 20 15:47:58 2016] igb 0000:03:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)

 

Maybe not latest drivers always good... let me try to compile again using this version.

 

Should I repack the extra.lzma or insmod is fine?

Share this post


Link to post
Share on other sites

look like insmod only load after. Even I copy igb.ko with lower version to /lib/modules still it load the latest version.

 

I need to use this command to unpack and pack back the igb.ko?

 

unlzma  -c ../extra.lzma | cpio -im

find . | cpio -o -H newc | lzma > ../extra_new.lzma

 

anyone can explain what is the cpio means?

Share this post


Link to post
Share on other sites

Here a few initial experiences with the new DSM loader for 6.0.2 on different bare metal systems... (and one was REALLY strange!)

 

1st system (AMD):

 

ASROCK FM2A87M-ITX+ with 8GB DDR3 and 6x2TB Seagate - boot up without any problems, upgraded and runs fine :cool:

 

2nd system (Intel):

 

MSI ZA77-GD65 with Intel Core i7 2600K, 16GB DDR3 and 6x1TB Seagate (onboard) and 4x1TB WD (LSI HBA) - boot up without any problems, etc. :cool:

 

3rd system (Intel):

 

HP Microserver Gen8, 16GB ECC, Xeon, 4x4TB Seagate - same over here, no problems (I use the SD card) :cool:

 

4th system (Intel) - this cost me 2 days and one sleepless night: :shock::sad::evil:

 

ASUS P9D WS, Xeon, 16GB ECC, 6x4TB Seagate (onboard), Intel X520DA-2 - don't ask what happened but it makes no sense (details below). Note: I have this mainboard/CPU/RAM combo twice (one system on hold for spare parts) and tried it first with a test installation on the spare system from DSM 5.2 with no errors.

 

After a successful test on the spare system I made the upgrade on the live system - worked without errors. Then I had the great idea to install a LSI 9200-8e (FW P19) to connect to a HP P2000 G3 SAS (12x2TB in a RAID10 vdsik). So I shut down DSM, stripped the power supply, plugged in the LSI and fired it up again. The HBA acknowledged the P2000 and DSM began to boot. But nothing happened. No ping, no putty, nothing. So I shut it down (thought of a driver problem with the HBA), removed the LSI and startet again. Same sh*t... Next thing I tried was the reinstall mode within the boot menu. Syno assistant found it immediately through DHCP and I reinstalled DSM 6. During the reboot I manually selected the 1st entry within the boot menu and let it go. Same result. No ping, no DHCP lease, etc. I changed the USB stick (with modifications in grub.cfg) - nothing. Switched USB 2.0 ports - nothing. Tried DSM 5.2 which only offered a recovery (useless). Booted Ubuntu 16 (live version) and gained access with mdadm - everything there, no disk errors, etc.

 

Note: I began at 9 in the morning - at this time it was 4 p.m.

 

The situation became more and more odd. Nothing worked with the DSM 6 loader (no reinstall or force install mode, I even didn't get access). I set the BIOS to defaults, tried the spare system, absolutely nothing. Then I switched to another Intel system (Z77 chipset) where I installed the 6 S-ATA HDDs. Changed the MAC (for onboard LAN) within the grub.cfg, started the system and bingo - welcome back. I reinstalled DSM 6 again and the system ran fine with no data loss. DSM ran a parity check for 3-4 hours. And now the funny thing: I switched it off, plugged the HDDs back to the P9D WS system, changed the MAC again (same USB stick), powered it on and hello! WTF???????

Share this post


Link to post
Share on other sites
Prompt me please, I can normally upgrade to Update3, if now I use a Hybrid UEFI/BIOS baremetal image setsunakawa (on 10 page), but with replaced the image/DS3615xs/ramdisk.lzma in the boot partition from the image in your, Jun, first post (on page 1).

 

If your bootloader works OK, it's not needed to touch on it. Just update to DSM 6.0.2 8451 Update3 only.

Hi.

 

Thanks, I updated to DSM 6.0.2 8451 Update3 properly. Can I upgrade to DSM 6.0.2 8451 Update4? In a my server with Asus p8z68 installed and running website. Nothing will break?

 

Thankyou.

Share this post


Link to post
Share on other sites
Prompt me please, I can normally upgrade to Update3, if now I use a Hybrid UEFI/BIOS baremetal image setsunakawa (on 10 page), but with replaced the image/DS3615xs/ramdisk.lzma in the boot partition from the image in your, Jun, first post (on page 1).

 

If your bootloader works OK, it's not needed to touch on it. Just update to DSM 6.0.2 8451 Update3 only.

Hi.

 

Thanks, I updated to DSM 6.0.2 8451 Update3 properly. Can I upgrade to DSM 6.0.2 8451 Update4? In a my server with Asus p8z68 installed and running website. Nothing will break?

 

Thankyou.

 

Some other people have without problems. The choice is yours.

Share this post


Link to post
Share on other sites

 

If your bootloader works OK, it's not needed to touch on it. Just update to DSM 6.0.2 8451 Update3 only.

Hi.

 

Thanks, I updated to DSM 6.0.2 8451 Update3 properly. Can I upgrade to DSM 6.0.2 8451 Update4? In a my server with Asus p8z68 installed and running website. Nothing will break?

 

Thankyou.

 

Some other people have without problems. The choice is yours.

 

Yes. I'd like nothing to break, because jun wrote:

 

The new kernel will not in effect, unless manually copy it to the right place.

Since Intel machines are in fact using unmodified synology kernel, I am considering restructure the boot partition's to closely match synology's, so that new kernel will take effect automatically, but this has the potential risk of a small upgrade may break a working system.

Share this post


Link to post
Share on other sites

Human being is not perfect and therefore things they've made also not perfect either. You think you have no risk if you buy Windows or Mac?

If you don't trust it, do use it. Why bother?

Share this post


Link to post
Share on other sites
I need to use this command to unpack and pack back the igb.ko?

 

unlzma  -c ../extra.lzma | cpio -im
find . | cpio -o -H newc | lzma > ../extra_new.lzma

 

The first line decompress lzma archive and cpio command extract files to current directory.

 

The second line pack all files and directories in current directory to cpio format and then apply lzma compression on it

 

anyone can explain what is the cpio means?

 

:arrow:https://en.wikipedia.org/wiki/Cpio

Share this post


Link to post
Share on other sites
I need to clear one last detail before updating, anyone can confirm if creating new disk group is working ?

Also can you expand volume ?

 

If anyone tested i would like to know.

 

So far I have migrated 4 of this from v5 to v6 you need to change de vid/pid and then edit synoinfo.conf to enable SHR, so far I have been able to expand with no problems, to edit synoinfo.conf you must ssh and then follow this to enable SHR with sudo:

 

edit /etc.defaults/synoinfo.conf

comment out/delete

CODE: SELECT ALL

supportraidgroup="yes"

 

then add

CODE: SELECT ALL

support_syno_hybrid_raid="yes"

 

Thanks for those infos, i didn't even know SHR wasn't supported natively. Is it over with DSM 6.0 or just a "bug" in this loader ?

Share this post


Link to post
Share on other sites

So far I have migrated 4 of this from v5 to v6 you need to change de vid/pid and then edit synoinfo.conf to enable SHR, so far I have been able to expand with no problems, to edit synoinfo.conf you must ssh and then follow this to enable SHR with sudo:

 

edit /etc.defaults/synoinfo.conf

comment out/delete

CODE: SELECT ALL

supportraidgroup="yes"

 

then add

CODE: SELECT ALL

support_syno_hybrid_raid="yes"

 

Thanks for those infos, i didn't even know SHR wasn't supported natively. Is it over with DSM 6.0 or just a "bug" in this loader ?

 

It's because Synology removed it's support for SHR for this specific model (we're emulating). Previous releases would include this out of the box but this is a more barebone release. So you'll have to do it manually.

Share this post


Link to post
Share on other sites

But if i migrate my current SHR raid to this dsm version, when i edit the synoconf it's after installation so would it wipe my data or will it boot nonetheless but just won't show it as SHR in the storage manager ?

And so i install dsm 6.0, i edit synoconf, reboot and voila all is good ?

Share this post


Link to post
Share on other sites

Are there any plans to fix the error with Intel X520-DA2 cards and similar models? Or are any workarounds available?

Share this post


Link to post
Share on other sites
But if i migrate my current SHR raid to this dsm version, when i edit the synoconf it's after installation so would it wipe my data or will it boot nonetheless but just won't show it as SHR in the storage manager ?

And so i install dsm 6.0, i edit synoconf, reboot and voila all is good ?

As you'll be migrating the SHR volume will show up & work ok, you just won't be able to create new SHR volumes unless you make the change in synoconf

Share this post


Link to post
Share on other sites

MOBO:GA-H170N-WIFI

 

I manually load different version of module also doesn't work. tried 5.3.3.2, 5.3.4.4, 5.3.5.4

rmmod /lib/module/igb.ko

insmod /lib/module/igb.ko

 

modified the sequence in /etc/rc.modules still the e1000e will load before igb.

 

in ramdisk, which file can control the sequence of loading the modules?

can I do the repack also with ramdisk? maybe I just remove the e1000e.ko and igb.ko inside then repack it again.

Share this post


Link to post
Share on other sites

Hello all

 

i tried Juns V1.01 loader but ist not working for me i cant finde the NAS on my router.

I have checked the module Folder and saw that forcedeth.ko is not included so thats the reason why my n570 Card is not working. can someone add it or tell me how i can do this by myself

 

thanks in advanced

Share this post


Link to post
Share on other sites

Hi all, can someone please helpme on how to make it work under vmware esxi? i've already tried to deploy using the v1.01 ovf and vmdk! but it fails with an "error uploading synoboot.vmdk to server IO error" did read the past posts, but can't find a "for dummys how to on this". sorry if missid some post... :smile:

Share this post


Link to post
Share on other sites

So I tried reversing the order ramdisk.lzma gets loaded in order to get i210 nic to work but every time i tried my 9211-8i will not get recognized

 

Is there any other way for the i210 to work?

Share this post


Link to post
Share on other sites
Has anybody successfully built a Nas based on new DSM 6.0.2 loader on VMWare Workstation?

Do I need an iso or img file to boot it up?

Thanks a lot.

Yes I did it. Check fiew post, i describes what I did.

 

If you want your vm to autoboot, create a batch file with "vmware - x yourvmfile".

I store 1 VMK / disk. So if one disk fails, i add a new one, and recreate the missing vmk

 

 

Envoyé de mon SM-T700 en utilisant Tapatalk

Share this post


Link to post
Share on other sites

@jun Since the upgrade in DSM to update 4 will not actually update kernel, then that means some critical security fixes mentioned in update 3/4 will not be applied even after the update, correct?

Can I just extract the zImage from the update 4 pat file, then replace the loader's zImage in synoboot1 partition, if it boots correctly, then I actually updated the kernel too? And the same logic applies to ramdisk, so I have to extract the rd file from pat, and repack it with community compiled kernel modules with driver supports?

Share this post


Link to post
Share on other sites
Refer to my previous question, I was able to install DSM 6 on my SSD drive. But the OVF file had a 50MB virtual Disk default setting (assume thats where the DSM install to) which also recognized by DSM itself as a HDD in Disk Manager (See attached photos). I am wondering if I have done anything step wrong to cause this issue? Don't seem to see other people experience the same. Any thing I can do to disable it from Disk Manager? Thanks.

 

YTLvc4K.jpg

 

amZNtul.jpg

 

I refer to my previous question posted here, I tried to diagnosis the cause of additional unwanted HDD in my DSM and realise the DSM is actually not booting from the USB drive I prepared (plugged inside MicroServer Gen8 internal USB port). Instead it boot from the 50MB vmdk file created (and saved on my SSD) when I deploy the OVF file. That also explains why the serial number and MAC number I edited in the grub and burnt to USB drive didn't change and show up when I set up the DSM when connected to find.synology.com. The plugged USB drive simply did nothing.

 

Problem is, I tried to remove the 50MB vmdk from virtual machine settings, added USB controller and USB xHCI controller to virtual machine settings. With the actual USB drive plugged in, I also added such USB drive to the virtual machine settings. However when I power the virtual machine on, I could not select it from the BIOS boot order, nor would it auto boot up.

 

So, sincerely, can someone running xpenology 6.0 under esxi on a HP Microserver Gen8 could share you thoughts and setup to help me out here??

 

Thanks so much

Share this post


Link to post
Share on other sites
@jun Since the upgrade in DSM to update 4 will not actually update kernel, then that means some critical security fixes mentioned in update 3/4 will not be applied even after the update, correct?

Can I just extract the zImage from the update 4 pat file, then replace the loader's zImage in synoboot1 partition, if it boots correctly, then I actually updated the kernel too? And the same logic applies to ramdisk, so I have to extract the rd file from pat, and repack it with community compiled kernel modules with driver supports?

 

I was asking myself the same thing and I have done some testing on my gen8 with jun's 1.0.1

I replaced only zImage and gen8 boot succesfully

Before

admin@gen8:/$ uname -a
Linux gen8 3.10.77 #8451 SMP Wed Aug 17 05:12:41 CST 2016 x86_64 GNU/Linux synology_bromolow_3615xs

After

admin@gen8:~$ uname -a
Linux gen8 3.10.77 #8451 SMP Wed Nov 9 13:45:32 CST 2016 x86_64 GNU/Linux synology_bromolow_3615xs

So I think the kernel has been updated... but no version change!? What does it mean?

After I tried to replace also ramdisk.lzma... the gen8 boot, stay up only few seconds (Synology Assistant show it) and then it shutdown...

 

[edit]

After rebooting the gen8 with correct ramdisk.lzma from jun's I can see this error on system events >> Abnormality detected on gen8. All volumes have been unmounted.

 

ciao

ema

Share this post


Link to post
Share on other sites
I Update whit Jun 1.0.1 and its working fine everything

 

Just a quick question

 

DSM 6.0.2-8451 Update 3 is available, someone know if its safe to update?

Updated successfully 2 VMs on eSXI and Proxmox :wink:

 

Care to share how you were able to make it work on Proxmox? Thanks.

Share this post


Link to post
Share on other sites
I need to use this command to unpack and pack back the igb.ko?

 

unlzma  -c ../extra.lzma | cpio -im
find . | cpio -o -H newc | lzma > ../extra_new.lzma

 

The first line decompress lzma archive and cpio command extract files to current directory.

 

The second line pack all files and directories in current directory to cpio format and then apply lzma compression on it

 

anyone can explain what is the cpio means?

 

:arrow:https://en.wikipedia.org/wiki/Cpio

 

Hi Arcao,

when I extract the ramdisk.lzma, I got the data is corrupt. But it not with extra.lzma

 

# unlzma -c ../ramdisk.lzma | cpio -im

unlzma: ../ramdisk.lzma: Compressed data is corrupt

45997 blocks

Share this post


Link to post
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.