Last week I had issue where SmartOS wouldn't boot suddenly(off USB), it just hanged there after "mDNSPlatformRawTime went backwards by 3684530 ticks..." warning. I think this one used to appear at times, but I didn't pay attention, however this is the point where it would get stuck booting. Then, since my Zpool didn't contain that much important data, I just unplugged this disk and used another identical one to create fresh pool, as I thought maybe the disk is the issue (was on cheaper end). Also I updated USB stick with latest SmartOS image and then it went past previous hanging point, but resulted in kernel panic, disabling a module. FMD-8000-2K
Long story short, because this machine is dual-boot, in Windows 7 I noticed that out of 6 sticks of RAM(total 12 GB), it sees only 2GB - one stick. Recently I had some spare time, and cleaned up dust, plus there was some hair where RAM connects, and there we go, 12GB back, so I definitely think this was the issue, and might've been the problematic module.
But when I get back to smartos and even on new zpool/disk, when I try to install latest bootstrap, it says that / filesystem is full.
I have not yet mastered Illumos/SmartOS, even though its my plan to delve as far as possible, so I'm not aware how metadata on specific Zpool works, or what part is changed on USB boot stick itself, after kernel panic..
Now, is it possible that on new Zpool I still have to locate disabled module and enable it?
UPDATE:
I managed to access my original pool after RAM was cleaned up and when I plugged the first disk back in. I made use of second disk as mirror now. But this is what still worries me - why is ramdisk:a such low capacity and full:
# df -h
Filesystem Size Used Available Capacity Mounted on
/devices/ramdisk:a 289M 289M 66K 100% /
/devices 0 0 0 0% /devices
/dev 0 0 0 0% /dev
ctfs 0 0 0 0% /system/contract
proc 0 0 0 0% /proc
mnttab 0 0 0 0% /etc/mnttab
swap 4.76G 980K 4.76G 1% /etc/svc/volatile
objfs 0 0 0 0% /system/object
bootfs 0 0 0 0% /system/boot
sharefs 0 0 0 0% /etc/dfs/sharetab
/devices/pseudo/lofi@1:disk
432M 360M 72.4M 84% /usr
/usr/lib/libc/libc_hwcap1.so.1
432M 360M 72.4M 84% /lib/libc.so.1
When I tried to get newer bootstrap using curl, I get / filesystem is full again, after certain download %.
So the main remaining issue now is / being full, but I'm not sure how it is mounted, or mechanisms behind it, or what I need to look into. Note: also I'm not sure what was the status of / before failing RAM, could be it was also low capacity and highly loaded, but upon first configuration 5 months back, I managed to pull and install bootstrap-2019Q2-x86_64.tar.gz without any problems.
Thanks for any help you can provide.
