installing directly to usb?

[ipanini]

Hi,

I've been out of the picture for a while.

Can the iso image be installed to usb directly? Maybe using UNetbootin?

I'm not sure since we're talking solaris here.

I would prefer this as it's quite a hassle to go back to working CD / DVD drives on my equipment.

Erwin

[ipanini]

Follow up:

- Perhaps outdated solution(s) are found here.
- Tried to create a usb with UNetbootin but it fails to boot (not 100% decisive as burning the iso to CD-R also failed 2 times before succeeding the 3rd time)

[Andre Lue]

It's possible to install from windows, but a .usb image is needed. I plan to release .usb images going forward, it is simply a conversion of the iso image.

More of an fyi, if you already have a working install, you don't need to create a cd/dvd.

-a

--
You received this message because you are subscribed to the Google Groups "EON ZFS Storage" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eonstorage+...@googlegroups.com.
To post to this group, send email to eonst...@googlegroups.com.
Visit this group at https://groups.google.com/group/eonstorage.
For more options, visit https://groups.google.com/d/optout.

[ipanini]

Thanks for replying;

No bootable system available yet.

In fact - much worse - I'm trying to pick up again on this situation.

At that time you wrote:

"Pre-req: all disks for current pool and new pool must be connected properly on the system"

At the moment I also need to check the correct connecting order to the controller.

Is there any reference that I can check to now which disk was in which order in the pool?

Will zfs figure this out by itself?

Or is all data lost if one can not figure out the correct disk-to-pool order?

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

See the attached screenshot how far the system boots...

Any clues?

Thx,

Erwin

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

Did the system boot ok from cd/dvd?

--

[ipanini]

No, it hangs at the screenshot.

I've tried with both 1.0b-151 and the new release.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

Which version/release is the screenshot booting?

Is it the latest 64bit eon-cifs-151a9-071215.iso?

Does the system have a removeable pci-express card/device that coincides with the hang point?

windows installer I mentioned earlier, requires a .usb vs .iso image: http://devzone.sites.pid0.org/OpenSolaris/opensolaris-liveusb-creator

I will post a .usb version of the image after I complete the new binary kit for the community release.

--

[ipanini]

Yes it is the latest 64bit eon-cifs-151a9-071215.iso

To be honest I don't see a difference in the hanging of either the latest and 1.0b-151.

Do you want me to reboot and post new pictures?

Does the system have a removeable pci-express card/device that coincides with the hang point?

->yes, that is the supermicro sat2-mv8

Not in a hurry with the usb images - thanks anyway

Erwin 

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

These are the attached drives.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

can you send output of  'grep marvell /etc/driver_aliases' for the version that boots and also the version that doesn't boot (if you can remove the card and test boot)?

--

[ipanini]

Extremely odd:

I had 2 pci cards in the machine, 1 intel pro/1000 nic interface as the mobo only has 10 mb on board if I remember correctly.

The other is the supermicro sata card.

I've pulled both, reenabled the onboard nic in the bios, and the system still does not boot... 

Last 2 lines of verbose boot:

PCI Express-device: pci1849,55f@2,1, ehci0

ehci0 is pci@0,0/pci1849,55f@2,1

I'll reboot and check if I find strange things in the bios

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

Hmm, what to do?

I tried a couple of cd / dvd's I had lying around:

- tried eon = hangs

- open solaris 0906_x86 = hangs <= I'm not sure if any bootable iso will boot with an AMD cpu...

- nexenta sunos 5.11 134e = hangs at loading kernel

- ubuntu 10.04 boots no problem

- freenas live 0.7rc1.4735 boots no problem

Came across a review of the mobo stating no compatibility problems with this mobo under Linux or Solaris.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

So the version that was booting before, no longer boots?

--

[ipanini]

Since the moment I started trying to revive this system it has not booted into eon or solaris.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

I'm bit stumped ... without the cards, does it hang/halt at the same point/msg?

The boot args are "-v -m verbose" or ?

--

[ipanini]

Yes, it hangs each and every time at the same point/msg.

I'm booting with -v verbose, have not tried -m / what does -m do? can it help?

I seem to remember something like part of the memory is used for I don't know what and that osol picks this up as ehci0 device..

Does this ring a bell to you?

One thing I know is it has been running..

I have an old usb but it is damaged, has been hit and there seems to be a missing connection between the usb connector and the print.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

Solved the main problem: 

It seems to have been my usb DVD drive that throws the ehci0 msg.

Scouring the web hinted often at usb..

Hence I connected a sata BD drive I have, and disconnected the USB DVD drive.

We're booting now.

I'll reassemble the system and see how it goes..

Thanks for sticking with me!

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

Ok - booting works now.

But  - tried an elderly eon usb - and got a similar usb msg halt. Now i haven't really waited for 10 - 15 minutes or so, because I remember at times it took really that long before boot went through..

- re-inserted supermicro sata controller

- boot from latest 64bit eon-cifs-151a9-071215.iso, boot succeeds

 

See screenshot with message about pool.

What I'd like to now:

- find some usb stick the system accepts to boot with (for new install of eon)

- add 4 or 8 TB drive to onboard sata

- proceed with the replication of the old pool to the new big drive

How do you suggest to proceed?  

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

Sounds more like a usb issue. Did you try booting from all usb ports(front, back panel, etc)?

For now you can proceed with the iso boot, the only real loss is because you cannot preserve the hostid, you will have to force import the pool on reboots. Everything else will run fine.

To answer your KVM question, its exactly KVM used in Linux. So most howto will work. I have a basic starter conf, launcher on sourceforge you can check out. Will require i3 or higher though, sorry no AMD.
https://sourceforge.net/projects/eonzfsstorage/files/KVM%20labs/

--

[ipanini]

I've booted with a usb stick plugged in - so far so good.

Ran all the checks as stated in your getting started wiki. Looks ok.

- Is there some cmd to tell the nic speed? I'm on the built in one for the mo, but I guess it's a slow one.

- From what point on can I use ssh?

I'll try to do a setup to the usb stick now.

Will try to leave the existing pool untouched for now.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[Andre Lue]

Try dladm show-phys

--

[ipanini]

Strange enough I keep having problems with usb.

Installed to previously used stick. Took very long to format..

Next boot hangs again at ehci0

Tried various ports.

All the same.

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

1. After several usb sticks throwing the same problem each time, I have added a 64GB ssd as boot medium instead.

I've proceeded with the install to the internal ssd and this is booting allright.

Are there any caveats as to executing config cmd's because this is not a usb stick anymore?

I'm following the wiki's burn - boot - store

Executing updimg.sh throws an error: /usr/.backup no such file or dir

2. network speed:

dladm show-phys does not show anything

but ipadm show-if gives result, only I can't see the link speed...

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

Hmm, there's a mistake in my previous post - (Is it possible to edit a post? don't seem to be able)

- I forgot to take out the CD, so the system was booting from the CD and not the sad

- it seems I cannot select the sata ssd as a boot option in bios... (seems strange...)

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

Recap:

Where are we?

- forget usb on this system. I even tried on old pci card with 5 usb ports. It worked but I got msg from solaris spread over several lines. In short I understood it was deprecated, usb type 1 etc. Tried to boot from it. Sticks were seen allright, but system hanged anyway. Tried disabling onboard usb to see if the PCI usb card would take over. Didn't work. So I quit with usb on this system.

- added a 64GB ssd, at first could not select it in bios to boot from. Finally found a way, had to cycle the 7th disk to the 1st one of that controller (in bios). Then I could select it as bootable option

- found that onboard nic did not work (saw no IP, and could not ping) disabled and re-added the Intel Pro.

- did the slinky r and c thing

- ran setup twice (after adding the Intel Pro)

- did updimg.sh

System boots ok

Mentions that pool can be imported with -f

I replaced my BD drive with a 6TB WD drive. Hopefully we can recover some or part of the data from pool0

Msg: The pool cannot be imported due to damaged devices or data

config:

pool0       unveil      missing device

   c0d0     online

   c2t0d0  online

   c2t4d0  online

   c2t2d0  online

   c2t3d0  online

As far as I can see some of the connected drives don't come online.

I have

c2t7do that does not show up. I believe (but need to check) it to be a WD green WD20EZRXTB. Maybe I have another one of these disks around and can try to swap the controller electronics.

As for now I prefer to await your advice.

It would be helpful to be able to login using ssh. Easier to handle from the desk and easier to grab msg's and logs.

Thanks for sticking with me! 

 

 

 

 

 

On Tuesday, 5 April 2016 14:22:37 UTC+2, ipanini wrote:

[ipanini]

a couple of screenshots

[ipanini]

Back in business!

After lots & lots more checking and re-checking I managed to get *ALL* disks online.

This leaves me with 2 possible pools to import with -f option. (imported with -f -r at the moment)

Last action to complete is to replicate the JBOD of 6 HD's to 1 HD.

Would you mind to help me out with this step by step if possible?

As far as I can see:

- need to make snapshot - how to do this?

- then send receive - I can follow the syncing-data FAQ

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

eon:2:~>df -h

Filesystem             Size   Used  Available Capacity  Mounted on

/devices/ramdisk:a     366M   185M       145M    57%    /

/devices                 0K     0K         0K     0%    /devices

/dev                     0K     0K         0K     0%    /dev

ctfs                     0K     0K         0K     0%    /system/contract

proc                     0K     0K         0K     0%    /proc

mnttab                   0K     0K         0K     0%    /etc/mnttab

swap                   1.3G   328K       1.3G     1%    /etc/svc/volatile

objfs                    0K     0K         0K     0%    /system/object

sharefs                  0K     0K         0K     0%    /etc/dfs/sharetab

/usr/lib/libc/libc_hwcap2.so.1

                       366M   185M       145M    57%    /lib/libc.so.1

fd                       0K     0K         0K     0%    /dev/fd

/dev/dsk/c3t0d0s0       59G   174M        58G     1%    /mnt/eon0

swap                   1.3G     0K       1.3G     0%    /tmp

swap                   1.3G    56K       1.3G     1%    /var/run

pool0                  2.6T    20G       351G     6%    /pool0

pool0/data             2.6T   394G       351G    53%    /pool0/data

pool0/public           2.6T   1.4T       351G    80%    /pool0/public

pool00                 2.7T   144K       2.7T     1%    /pool00