Please help: system now hangs at boot "trying to mount root from..."

[ptmixer]

My new FreeNAS 8 system was installed in a VMWare ESXi 5.0.0 virtual machine using the iso of FreeNAS 8.0.2-RELEASE. I created a 2GB boot virtual drive, and three large data virtual drives. Worked fine. Used it and rebooted it several times. Today I take it down to check something, and it will not boot up anymore. It goes to the FreeNAS text-art splash screen, then I have the usual boot options. It starts booting, then gets to this line:

Code:

Trying to mount root from ufs:/dev/ufs/FreeNASs1a

It just hangs there. In verbose mode it goes on to say:

Code:

ct_to_ts([2011-11-11 12:47:55]) = 132101.5675.00000000
start_init: trying /sbin/init
t_delta 16.021cdaeb542aabe0 too long

and the t_delta message slowly repeats with a slightly different number. [edit - subsequent googling suggests the t_delta message is unrelated to the boot problem]

Suggestions on how to fix would be appreciated. I have attempted to also boot again from the iso to try to repair it, but I get the same result -- I can't guarantee that I'm booting off the CD but I have VMWare and the virtual machine's bios set correctly to do so I think.

Thank you.

 

[Durkatlon]

Sounds like perhaps a bum USB stick. Do you have another you can try, preferably a 4 gig one?

 

[ptmixer]

Thanks, Durkatlon, but I'm not using a USB stick. It is all (virtual) hard disks. If something did mysteriously go wrong with the boot drive, how could I repair it? Foolishly, I did not back up my configuration so if I have to re-install will be a bit of a pain.

 

[Durkatlon]

Sorry, my bad. I missed the bit about ESX :D. Yeah, I was saying USB stick because I've had a stick go bad and the behavior is similar. With ESX I have no clue... Probably best to reinstall and take a configuration backup this time!

 

[ptmixer]

I got it to boot from the installer ISO again (had gotten corrupted), tried to do an in-place "upgrade" (reinstall) on my failing boot device. That went through, but when it tried to actually boot again it complained of problems and rebooted itself.

I created a new boot drive, an 8GB thin disk this time, installed anew, created my groups and users, imported my ZFS volume (auto import worked perfectly, thank you!) worked on some new settings and am back where I was I think. I even saved my config!

There still is a complaint I see whizzing by during boot that da1 has a geometry mis-match or something like that. Hmm. Maybe I'll try a reboot now and make sure it's all happy.

 

[Durkatlon]

There still is a complaint I see whizzing by during boot that da1 has a geometry mis-match or something like that.

That is actually normal. The expected geometry of the OS drive is something that is baked into the FreeNAS image. So if you use a different size stick or - as in your case - a virtual HDD instead of a stick, this message will pop up during boot.

 

[ptmixer]

It rebooted fine. Here's the actual line about the geometry which I had on the broken and still on the fresh one too:

Code:

Nov 11 18:05:42 freenas kernel: GEOM: da0s1: geometry does not match label (16h,63s != 255h,63s).

A little higher up, the description of da0:

Code:

Nov 11 18:05:42 freenas kernel: da0 at mpt0 bus 0 scbus1 target 0 lun 0
Nov 11 18:05:42 freenas kernel: da0: <VMware Virtual disk 1.0> Fixed Direct Access SCSI-2 device
Nov 11 18:05:42 freenas kernel: da0: 320.000MB/s transfers (160.000MHz, offset 127, 16bit)
Nov 11 18:05:42 freenas kernel: da0: Command Queueing enabled
Nov 11 18:05:42 freenas kernel: da0: 8192MB (16777216 512 byte sectors: 255H 63S/T 1044C)

Ticking time bomb, or just weirdness that had nothing to do with my boot failure?

[thanks, Durkatlon, I posted this before seeing your reply]

 

[ptmixer]

Upon reflection, I think it was an actual disk error that caused the issue.