Re: sysupgrade failure logs

2021-02-14 Thread V S




On 2/14/21 7:02 PM, Judah Kocher wrote:

Hello folks,

I am having an issue with sysupgrade and I have had trouble finding 
the source of the problem so I hope someone here might be able and 
willing to point me in the right direction.


I have 6 small systems running OpenBSD -current and I have a basic 
script which upgrades to the latest snapshot weekly. The systems are 
all relatively similar. Three are the exact same piece of hardware, 
two are slightly different, and one is a VM configured to match the 
first three as closely as possible with virtual hardware.


The script checks the current kernel version, (e.g. "GENERIC.MP#302") 
logs it, runs sysupgrade, and after the reboot it checks the kernel 
version again. If it is different it logs it as a "success" and if it 
is still the same it logs it as a failure.


All 6 systems were configured using the same autoinstall configuration 
and the upgrade script is identical on each unit. However, two of the 
three identical units always fail. When I remote into either system 
and manually run the upgrade script it also fails. I was able to get 
onsite with one of them where I connected a monitor and keyboard and 
manually ran the script to observe the results but oddly enough it 
succeeded so I learned nothing actionable. However it continues to 
fail the weekly upgrade. I have confirmed that the script permissions 
are identical on the working and nonworking units.


The 4 units that successfully upgrade leave a mail message with a log 
of the upgrade process. However I have been unable to find any record 
or log on the systems that are failing to help me figure out why this 
isn't working. The only difference I can identify between the systems 
is that "auto_upgrade.conf" and "bsd.upgrade" are both present in "/" 
on the two systems that fail, but are properly removed on the 4 that 
succeed.


I would appreciate any suggestions of what else I can try or check to 
figure out what is causing this issue.


Thanks

Judah



Care to show a script ? otherwise it looks like rather lengthy 
mathematical problem with quite some variables.




Re: not deleting sysupgrade, was: sysupgrade failure logs

2021-02-15 Thread V S




On 2/15/21 6:23 PM, Luke A. Call wrote:

On 2021-02-15 09:33:03+, Ottavio Caruso  
wrote:

On 14/02/2021 23:44, Theo de Raadt wrote:

When we get reports like this where people "touch the insides",   both
Florian and I regret that sysupgrade ever arrived in the system.
We want to delete sysupgrade.

If this is not just a provocative statement, +1 from me.
I've never liked unattended, automatic, Debian-style system upgrades. A lot
of things can go wrong.

I think I stay in the box, and definitely appreciate sysupgrade (etc).  It
has made my openbsd use more secure and easier (given that I am not near
your level of expertise here), so, thanks for it being there.

Luke Call
http://lukecall.net



Noobs like me would not be able to run snaps without sysupgrade..
Think of such noobs. But even I understand that running snaps is not
for unattended systems, but only ones I interact with mostly. I might
just suggest that OP rather make primitive scripts for unattended boxes
syspatch
pkg_add -u

Keep it Simple Stupid,
sysupgrade always works for me.



Re: sysupgrade failure logs

2021-02-15 Thread V S




On 2/15/21 7:21 PM, Judah Kocher wrote:

Hello Theo,

I never for a moment intended to convey that anyone "owed" me support 
of any kind for my outside-the-box use of this tool. While I don't 
understand your vitriolic response to someone else's application of 
your software for their own personal use in a way you do not condone, 
you are certainly entitled to be as outraged as you please. I remain 
grateful for the work you and others put into the OpenBSD operating 
system. It has been made clear on multiple occasions that use of 
sysupgrade with anything other than default responses is heretical and 
cancel-culture worthy but I don't mind breaking things while 
experimenting and do not blame anyone else when this happens, nor do I 
particularly care if anyone else is bothered by it as long as no 
actual harm is being done.


If anyone cares to read my original query from an intellectually 
honest perspective I think they would be hard pressed to respond as 
you have. I never claimed my "sysupgrade use was completely normal" 
nor did I blame the sysupgrade tool for the issue I am attempting to 
diagnose. I did not mention my usage of it because logically it does 
not seem to be relevant and I was concerned it would become an excuse 
for people to fly off the handle. I only had and still only have one 
question.


Does sysupgrade leave any kind of logging behind which could help me 
to pinpoint why it is failing on one system while working on another 
apparently identical system?


If the answer is no, that's easy enough to say. If the answer is yes, 
that's also easy enough if anyone is willing to share where those logs 
would be found. If the answer is, "Maybe, but no one owes you that 
information" that is also perfectly true while kind of pointless to 
even bother saying, although a world where people only offer help to 
others when there is a financial obligation would be a dismal place 
indeed.


I did not and do not expect anyone else to solve my problem for me. If 
you have reason to believe that my "mis-"usage of sysupgrade has 
anything at all to do with this issue, I'd be curious to know how you 
would explain it working on 4 out of 6 systems. Since it seems 
unlikely that the exact same tool would work two different ways on two 
identical systems then logically I would assume that some subtle 
difference exists between them and was hopeful that any records of the 
sysupgrade process would help me identify that difference. I have been 
using this script on these and other less similar systems ever since 
the sysupgrade tool was released with no issues, and therefore I think 
it's reasonable to to conclude that using it this way, while not 
officially sanctioned, has nothing to do with what's going on in this 
particular case.


Thank you again for your work on OpenBSD, including sysupgrade.

To everyone else on the mailing list, I do not apologize for asking a 
question but I do apologize for the drama it provoked.


Judah


On 2/14/21 6:44 PM, Theo de Raadt wrote:

You are outside the box, by changing tons of stuff.

People who operate inside the box won't be able to help you.

And it is even less likely when you are dishonest in the original email.
You claimed your sysupgrade use was completely normal, but it isn't.

It is far from normal.

When we get reports like this where people "touch the insides",    both
Florian and I regret that sysupgrade ever arrived in the system.

We want to delete sysupgrade.  Or, every month or so change the 
internals

so that it will delete some people's machines.

Does sysupgrade recommend what you do?  No.  But you do it.  Do you 
understand

the concept of "you own all the pieces"?




The drama was there, but not of Theo, but mine and someone else who
kinda derailed this splitting
this topic in separate threads "deleting sysupgrade" and "not deleting
sysupgrade", however,
It might be better to use stable for unattended systems, while just
having 'syspatch && pkg_add -u'
script in cron for systems that are not intended to be so much
interactively used.

If sysupgrade(8) does not mention any logging done by it it likely
doesnt, and as 'whereis sysupgrade'
shows it is at /usr/sbin/sysupgrade, and its a shell script which can be
viewed by an unprivilleged user
using something like less. And it is not that long to investigate or
improve on your own.

As for your use of it, you might be better with your own custom way of
upgrading, that you can
handle and log on your own.

Just my 2 cents.

Keep it Simple Stupid.

P.S. Sorry, but I've never used mailing lists before (since few days ago)
and I just sent this reply to bugs instead of here.



D-Link DGE-530T H/W Ver.:D2

2021-02-26 Thread V S
I've bought few of these nics some months ago. I suspect OpenBSD re 
driver needs some adjustments and I could send one of them to an 
interested developer.



P.S. I'm getting watchdog timeouts in some hard to reproduce fashion. 
And I'm way below the
capability to fix it myself and it seems that only these devices are 
currently getting those timeouts
on a router. I'll try booting freebsd on the machine trying 'reproduce' 
(more like waiting till it strikes),
just to see. But I dont want to switch distro for a router if thats 
really just driver needs adjustment.




dino broken on snapshot from about 6hours back

2021-07-11 Thread V S

$ dino
Could not connect: No such file or directory

(dino:7399): GLib-CRITICAL **: 12:24:48.665: g_bytes_get_data: assertion 
'bytes != NULL' failed


(dino:7399): GLib-CRITICAL **: 12:24:48.665: g_bytes_ref: assertion 
'bytes != NULL' failed


(dino:7399): GLib-CRITICAL **: 12:24:48.665: g_bytes_get_data: assertion 
'bytes != NULL' failed

**
GLib:ERROR:../glib-2.68.3/glib/gvariant-serialiser.c:1360:g_variant_serialised_n_children: 
assertion failed: (g_variant_serialised_check (serialised))
Bail out! 
GLib:ERROR:../glib-2.68.3/glib/gvariant-serialiser.c:1360:g_variant_serialised_n_children: 
assertion failed: (g_variant_serialised_check (serialised))

Abort trap (core dumped)