16.04 problems - install, inactive/suspend, hard disk

[rikona]

Hello Ralf,

Monday, September 12, 2016, 2:03:45 AM, Ralf wrote:

> On Sun, 11 Sep 2016 21:06:31 -0700, rikona wrote:
>>In one of the syslogs there was a line with "soft lockup - CPU#2 stuck
>>for 23s". This line appeared many times in the log. I didn't know what
>>to make of it but it doesn't sound good.

> Somebody claimed that replacing the PSU should have solved this issue,

I saw that, at the end of a long article where many ideas were tried -
they were confused too. :-)

> however, consider to google more and read more about this issue.

I've read many. The problem is when it was solved, it was solved in a
number of different ways - so which one is what is actually going on
in my box. :-)

> https://www.google.de/?gws_rd=ssl#q=soft+lockup+-+CPU%232+stuck+for

->> https://ubuntuforums.org/showthread.php?t=2205211 

>>I ran memtest86 directly from a bootable memtest86 CD, and let it run
>>the default test. It ran for two days and gave no errors, but it did
>>give a "note" stating that "RAM may be vulnerable to high-frequency
>>row hammer bit flips". Could this be a problem? Enough reason to
>>return the new memory?

> I don't know, you need to google yourself.

I did but so far but no conclusive opinion as to being a serious
enough risk to return the memory.

> "Starting from MemTest86 v6.2, the user may see a warning indicating
> that the RAM may be vulnerable to high frequency row hammer bit flips.
> This warning appears when errors are detected during the first pass
> (maximum hammer rate) but no errors are detected during the second pass
> (lower hammer rate). See MemTest86 Test Algorithms for a description of
> the two passes that are performed during the Hammer Test (Test 13).
> When performing the second pass, address pairs are hammered only at the
> rate deemed as the maximum allowable by memory vendors (200K accesses
> per 64ms). Once this rate is exceeded, the integrity of memory contents
> may no longer be guaranteed. If errors are detected in both passes,
> errors are reported as normal." -
> http://www.memtest86.com/troubleshooting.htm

> Perhaps you could limit speed by the BIOS settings, but I don't know if
> this is related to the issue. Google is your friend.

>>Noting earlier that CPU number two was stuck for 23 seconds, I tried
>>to run memtest86 using CPU number two. The result was an instant full
>>lockup crash. Rebooting and trying with the remaining four cores
>>produced the same result. This would seem to indicate that there is
>>some kind of CPU problem, but I'm not absolutely sure that memtest86
>>will do a core selection with a late model AMD64 six core processor.
>>If anyone can confirm that it does, I would appreciate knowing that.

The reason I asked was that I ran cpuburn in addition to memtest86 -
memtest86 crashed with anything other than CPU#1. cpuburn supposedly
runs ALL the cores, and that ran fine for quite a long time. So it is
still not clear if there really is a CPU problem...

> A while back I thought my mobo was borked, but it wasn't. I backuped
> the BIOS settings to an USB stick, turned off the PC, removed the
> onboard button cell battery, moved the jumpers to clear the CMOS RAM as
> described by the mobo's manual, moved the jumpers back, installed a new
> battery, restored the old BIOS settings from an USB stick and
> everything worked again. In the past I several times replaced batteries
> without clearing the CMOS RAM and got rid of serious issues.

I'll try that, although it is a brand new MB.

>>Is it possible, during an install, to put /home on a separate 3GB
>>disk?

> Yes, there is, but this is off-topic now. The installer provides an
> option to manually assign a directory, e.g. /home, to a partition.

I assumed there was. The problem was that with the 3Tb disk the box
crashed - *every time* - part way through the "other" option, so I was
not able to use that to format the 3Tb disk during the install. I
formatted the 3Tb disk separately, then installed it in the box.

When that did not work well I then did a complete reinstall with a
1.5Tb disk, and THAT disk did make it all the way through the "other"
option with no problem, on the first try, and was able to get /home on
the 1.5Tb disk. The other thing that was different is that the 1.5Tb
disk did not crash overnight, which was a regular occurrence with the
3Tb disk.

At this point, I thought that the original but new 3Tb disk might have
a problem, and removed it. I replaced it with another brand new 3Tb
disk, and again tried to do a complete reinstall. Unlike the 1.5Tb
disk, which went through the "other" option with no problem, this 2nd
brand-new 3Tb disk also had a lot of problems going through the
"other" option. The first four times I tried to go through that
option, the box would crash before I made it through to formatting the
disk. However, on the ***fifth*** try I did make it through the rest
of the format. It crashed again at what seemed like the end of the
install. A reboot brought it up again, but with a thoroughly messed up
display - the screen was much too large to display on the monitor and
was unusable. Another reboot did bring up the screen properly.

I'm beginning to think it may be UEFI related - whenever UEFI is
needed, I seem to have a problem. A UEFI boot with the Ubuntu CD does
not work well, but a non-UEFI Ubuntu CD boot is much better. My backup
plan is another reinstall with a 2Tb disk - the max for non-UEFI as I
understand it.

I'm getting soooo irritated with this hardware I'm about to bring it
all back and start over. :-( Never ever had this much of a problem in
building a box...

And again, thanks VERY much for the help!

-- 

 rikona