Re: Bug: Buster 10.0 Testing 20190617 Image LiveCD Will Not Boot
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello Michael,
There was no key sequence to interrupt the splash screen
The kernel boot delay parameter was tried as suggested via
suggestions of replies I had via the
debian-live@lists.debian.org. This despite my expecting to
have no effect on being able to see the boot messages flashing
past on the splash screen. The kernel boot delay parameter did
delay the kernel boot, but has no effect on the original issue
of splash error messages flashing by very very quickly. This
means I determined the real kernel boot parameter was
boot_delay and not the one suggested via reply to me via
debian-live@lists.debian.org.
The kernel boot delay parameter, the exact messages I
determined via my own technical efforts are in my replies to
debian-live@lists.debian.org. Because the
debian-live@lists.debian.org does not have a related NNTP I
could not reply in manner that would place my replies with the
related message I was replying to. You need to look at my
replies to see what I did, what I tried per suggestions, and
what I found as the live boot error messages. For example my
replies to your messages are not being placed with my prior
replies to debian-live@lists.debian.org.
Be aware there were much more serious bugs, I do mean bugs, I
never mentioned nor posted that I suspect are deeper level of
issues with the live image boot.
I had no problem reproducing the issue as I reported to and
asked for assistance how to obtain the flashing error messages
on the splash screen. The bug occurred 100% of time.
Testing via VM/KVM has limited value for most part. I suggest
you refer to:
<https://lists.debian.org/debian-live/2019/06/msg00040.html>
<https://wiki.debian.org/Teams/DebianCD/ReleaseTesting/Buster_r0>
and my replies to debian-live@lists.debian.org as to why KVM/VM
testing of images is of limited scope/value.
PC based VM unlike IBM VM that has much much less reduction in
scope difference compared to bare metal instance of VM or OS
Guest under VM. I made no mistake about VM under VM with IBM
VM. It is very common for various reasons I will not get into
why instances and nested instances of VM are run under VM with
IBM VM. Likewise I will skip the VM in the hardware microcode of
last number of years that is another level of VM abstraction on
top of any IBM VM OS based IBM VM. This unlike PC based VM
that is basically crude and temperamental compared to IBM VM
even of the 1980s. IBM had VM as Operating System least by
early 1970s as a business OS environment choice for customers.
Most customers using VM also had one or more other IBM OSs
running under VM. I certain has systems experience from
technical, systems programming, and operations experience with
systems running 10, often more, VM guests on an instance of VM
let alone multiple VMs on different physical machines VM
communicating at VM level as easy as buying the same multiple
item in a store at same or different time/day. At least as of
few years ago IBM VM was still able to boot and emulate IBM
CPUs from the 1960s just fine (had different PSWs for starters).
Do not assume the issues and testing that I did that lead to
the Buster LiveCD issue I reported and the more serious Buster
issues I reported were done on a laptop, let alone an AMD
E-350. The fact is the Buster LiveCD boot issues were on
Lenovo hardware, that was not a laptop machine. The more
serious Buster LiveCD boot issues likewise not on laptop. The
Stretch LiveCD and prior to Stretch LiveCD boot images have had
a number of issues, some similar to the serious LiveCD boot bug
I have not reported and others persisted for few Major Debian
releases all while other Dabian based LiveCDs of respective
Debian release not having any LiveCD boot issues. This means
for a number of years on various hardwarwe Debian releases
have had bugs of serious nature with the LiveCDs.
I understand the limited time Debian maintainers, et al have,
but to spend time wasting time and spinning on issues that
others clearly do have when faced with comparisons on exact
same hardware that do not have any issues is actually reducing
by orders of magnitudes these Debian peoples time and
productivity. I dare say to point of little movement for
Debian as distribution and causing many to be turned away from
Debian as place to consider. I can say that from the Buster
point of view I would not use buster as it has way too many
serious basic issues. I have not reported these other
bugs/issues and will not waste my time trying to report.
As of this eMail I will not spend more time dealing with the
issue as I reported and asked for assistance to find the
error messages that flashed on the splash screen.
I wish you the best in the issues you wish to report.
John L. Males
Toronto, Ontario
11 July 2019 18:24
================================================================
2019-07-11 21:27:22.236583311+0000-UTC Time:
1562880442.239757579
11 Jul 21:27:22 ntpdate[9817]: ntpdate 4.2.8p10@1.3728-o Sun
Feb 25 21:22:56 UTC 2018 (1)
11 Jul 21:27:36 ntpdate[9827]: step time server 132.246.11.227
offset -0.005288 sec
Linux 4.9.0-9-amd64 #1 SMP Debian 4.9.168-1 (2019-04-12)
Modified Debian GNU/Linux 9.9 (stretch)
(Alternative to Debian determined, work in progress)
cat /proc/cpuinfo (Selected):
model name : AMD E-350 Processor
vmstat -s:
16026156 K total memory
7462292 K used memory
13107036 K active memory
1465060 K inactive memory
1121912 K free memory
59936 K buffer memory
7382016 K swap cache
0 K total swap
0 K used swap
0 K free swap
189957226 non-nice user cpu ticks
77103656 nice user cpu ticks
232489899 system cpu ticks
582013003 idle cpu ticks
2681070 IO-wait cpu ticks
0 IRQ cpu ticks
5009761 softirq cpu ticks
0 stolen cpu ticks
306880449 pages paged in
55217707 pages paged out
0 pages swapped in
0 pages swapped out
749330804 interrupts
2602289249 CPU context switches
1557283473 boot time
547412536 forks
/proc/vmstat (Selected):
pgpgin 306880449
pgpgout 55217707
pswpin 0
pswpout 0
pgfree 67739341875
pgfault 116367091499
pgmajfault 944674
/proc/meminfo (Selected):
Mlocked: 160 kB
VmallocTotal: 34359738367 kB
VmallocChunk: 0 kB
HugePages_Total: 0
vmstat --partition /dev/sda8 (Swap):
partition was not found
sar -b:
Linux 4.9.0-9-amd64 (debian) 07/11/2019
_x86_64_ (2 CPU)
12:00:01 AM pgpgin/s pgpgout/s fault/s majflt/s pgfree/s
pgscank/s pgscand/s pgsteal/s %vmeff
09:05:01 PM 2465.54 0.00 25578.34 15.52 17601.36
1370.44 0.00 1146.60 83.67 09:15:03 PM
5629.87 2.21 15255.68 28.42 12426.68 2968.83
0.62 2614.10 88.03 09:25:01 PM 559.82 3.79
34624.58 3.46 22633.22 376.50 12.01 172.93
44.51 Average: 70.92 0.18 22531.85 0.39
13006.09 39.84 0.10 33.09 82.85
ps -A:
%CPU START TIME C CLS COMMAND TIME NI PID
POL PRI SZ RSS VSZ SIZE MAJFL MINFL SCH STAT
TIME WCHAN
0.0 May 8 52:00 0 TS kswapd0 00:52:00 0 35
TS 19 0 0 0 0 0 0 0 S
00:52:00 -
Message replied to:
Date: Wed, 10 Jul 2019 11:05:26 +0200
From: Michael Kesper <mkesper@schokokeks.org>
To: jlmales@gmail.com
Cc: debian-boot@lists.debian.org, debian-live@lists.debian.org
Subject: Re: Bug: Buster 10.0 Testing 20190617 Image LiveCD
Will Not Boot
>
> Hi John,
>
> > There were no replies to my post to
> > <mailto:debian-boot@lists.debian.org> of
> > <https://lists.debian.org/debian-boot/2019/06/msg00147.html>.
> >
> > I then posted to <mailto:debian-live@lists.debian.org>
> > starting with
> > <https://lists.debian.org/debian-live/2019/06/msg00016.html>
> > where there were various replies, suggestions and my
> > responses/actions to same.
> >
> > The net my attempts to report a bug by validating my
> > concerns of bugs in the Buster Live CD image were such I
> > gave up. If I could not be just a bit successful in
> > confirming to the Debian Community of bugs, then experience
> > has already demonstrated opening a bug report would be far
> > more time consuming and way too much spinning of others and
> > more.
> >
> > I had no guidance how to capture the error messages
> > flashing on the splash screen.
>
> Can you interrupt them by hitting the "Break" key on your
> keyboard? (Might require some key combination on laptops)
> Otherwise, there was a "bootdelay" parameter mentioned.
> As can be seen on
> https://www.kernel.org/doc/html/v4.14/admin-guide/kernel-parameters.html
> the parameter really is "boot_delay" with an underscore.
>
> > I managed using my own technical
> > skills, knowledge and alot of time to obtain the complete
> > and exact error messages that were flashing on the boot
> > splash screen. Despite having the exact error messages and
> > what conditions the error messages displayed in I still
> > received basically an opinion of not a Buster issue.
>
> Sorry, I did not find them in that thread?
>
> > I gave up as too much wild goose chasing advise or m being
> > told not a bug at all. In my opinion the facts were not
> > being not accepted of what was clear to me a bug (I
> > actually discovered a few bugs and only reported the easy
> > bug in my technical opinion) and not much help in the ask I
> > posted for how to obtain the error messages flashing on the
> > splash screen.
>
> In this case you might need the same (or at least similar:
> AMD CPU/GPU, legacy BIOS boot) hardware to reproduce this
> issue so it's not that easy.
>
> [...]
>
> > A familiar Open Source project response is they do not have
> > the time to work on bugs/issues, works for them so cannot
> > exist, nobody else has reported the issue/bug so cannot
> > exist. The latter two are actually far from truth of
> > determining if a bug exists. I worked many years in roles
> > where the issue reported was a bug and often serious, but
> > only one person/client often was only experiencing the
> > issue. I have worked on bugs that only occur once every
> > few months. I have found the cause of bugs in less than an
> > hour the software engineering teams have not been able to
> > identify in two years. This is not me bragging or
> > similar. The point here is it takes a special set of soft
> > skills, technical skills and problem solving skills that
> > seems to be less and less understood and strived for with
> > software/firmware.
>
> I can relate to this.
>
> I don't have any solution though, sadly.
> As Debian is completely made up of volunteers, everyone does
> what they can but time is limited for everyone.
>
> [...]
>
> > As to the trying the released Buster Live, I happened to do
> > so for secondary reasons this past weekend. Serious issues
> > in my opinion that may or may not be Live image specific.
> > I do not know as I was not going to investigate.
>
> Unfortunately I cannot reproduce your specific bug.
> When testing the live image with kvm though, I noticed many
> squashfs errors, though, when shutting down.
> Will try to report that when I find the time.
>
> Best wishes
> Michael
>
-----BEGIN PGP SIGNATURE-----
iF0EARECAB0WIQQxRId2q5JPHFiozTr5X9dS0HpoEAUCXSe3IAAKCRD5X9dS0Hpo
EGrHAKDArbSqHp2vhEYdn+nrzlVf72FfZACfbqXDLFI5Jdh/bXd4LmFMNWBe3LU=
=PgtP
-----END PGP SIGNATURE-----
Reply to: