You are not logged in.
Hello:
I do not use IPv6 and long ago (ascii) disabled it via the usual kernel command line stanza.
ie: ipv6.disable=1
I case that were not enough, I also blacklisted the module:
$ cat /etc/modprobe.d/blacklist-ipv6.conf
# Blacklist IPv6 module.
blacklist ipv6
$
My dmesg printout at boot time tells me about it ...
$ sudo dmesg | grep -i ipv6
--- snip ---
[ 2.834452] IPv6: Loaded, but administratively disabled, reboot required to enable
[ 2.834736] mip6: Mobile IPv6
---snip ---
$
... but it clearly says IPv6: Loaded, which is not what blacklisting the module should achieve.
And it seems that, as a result (?), mip6 is also loaded.
That said, neither module show up with lsmod.
What's going on here?
Am I blacklisting the wrong module?
Best,
A.
Hello:
You were on the wrong archive page.
I beg to differ: I was on the right page.
The one belonging to lists.cups.org, now inaccessible.
Try this one ...
Great find ... 8^D
But that is a Wayback Machine snapshot, unsearchable (?).
Thanks for your input.
Best,
A.
Hello:
Thanks ...
You're welcome.
... didn't search much cause ...
That's the first thing to do before posting.
You don't know what you don't know till you look for it. 8^°
Funny how HP support ...
Actually it is tragic.
For years now, Bill Hewlett and David Packard (founders of the original HP) have been turning in their graves three times a minute.
To think what HP was and see the piece of crap that it has become ...
Uncanny.
Best,
A.
Hello:
... when I shut down my HP ZBook 15 G2 ...
Check here: https://discussion.fedoraproject.org/t/ … own/124911
TL;DR
Disable the “Wake on LAN” option in the BIOS.
If that does not work, check this link where you will find many hits with posts by people having the same issue.
Sometimes, Google can be a friend. 8^°
Best,
A.
Hello:
Ever since I installed CUPS in my Devuan systems, my go-to source of information (besides Dev1) was the https://www.cups.org/ web site.
A couple of days go I had an issue I wanted to search for and went to the archives page, only to be greeted with this:
---
Forbidden
You don't have permission to access this resource.
---
I thought that to be very strange as this was not a web page (404 Not found) problem.
I recall having accesed the archives page last May without seeing this.
But I was still receiving posts by list members and was able to post to the list. ie: posts did not bounce.
But unable to check the archives to see what was going on.
So I sent an email to cups-owner asking about this and got this rather cryptic reply:
More recent cups related stuff can be found at openprinting.org.
When apple bought cups, those lists went to their servers.
Mike quietly left apple years ago, and it appears that apple has removed the lists.
They own "cups". So Mike had to make a new name.
'Mike' refers to Michael R. Sweet, original developer of CUPS and Gimp-Print /Gutenprint who left Apple late December 2019.
Much to my chagrin, a second email asking for additional information was not replied to.
I then posted to the OpenPrinting GitHUb page asking about this and right away received a reply from M.R. Sweet himself:
Unfortunately, lists.cups.org is an Apple-managed site and we have no control over its contents or configuration...
The printing-users and printing-architecture lists on kernel.org are the current place for discussing printing-related issues.
Right ...
Whatever has transpired here is not good, not at all.
Unless there is an accesible mirror hidden somewhere, the lists.cups.org archives are, to all intent and purpose, gone.
Sequestered by Apple Inc. and with it, many years of useful CUPS related information.
Heads up:
These are the links to the current place for discussing CUPS printing-related issues.
printing-users Linux printing list for end users to discuss printing issues / feature requests
printing-architecture Printing architecture under Linux
The layout is definitely strange ie: not the usual mailing list layout.
But it is there.
[rant]
Over the years, I ended up getting used to Apple and the shenanigans I read about in the press.
Eventually nothing surprised me.
But this? ... 8^ | <---- Oracle did worse yet with the Sun Microsystems lists and file downloads.
[/rant]
That's all.
Best,
A.
Hello:
thanks ...
You're welcome.
... seems connected with this "systemd" thing,
Yes, your post is.
What I meant to say was that it is not connected to the OP by @Fjalar.
That said, let's see if we can make some sense out of what seems to be going on.
... this package has been function since "etch".
etch is a Debian release from April 2007 and Devuan did not exist at that time.
Devuan Jesse was released May 2017, roughly ten years later.
$ uname -a tells us your system runs Devuan Daedalus.
apt list | grep -i installed | grep -i open-iscsi tells us the open-iscsi package installed is open-iscsi_2.1.8-1_amd64.
Yet for some reason, attempting to run the application gets you a classic systemd error in a system that by definition does not use the systemd package for init.
In an attempt to get to the bottom of the problem, I decided to see if I could reproduce the problem:
Installation
# apt install open-iscsi
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
The following additional packages will be installed:
libisns0 libopeniscsiusr
Recommended packages:
finalrd
The following NEW packages will be installed:
libisns0 libopeniscsiusr open-iscsi
0 upgraded, 3 newly installed, 0 to remove and 0 not upgraded.
Need to get 464 kB of archives.
After this operation, 2152 kB of additional disk space will be used.
Do you want to continue? [Y/n]
Get:1 http://deb.devuan.org/merged daedalus/main amd64 libisns0 amd64 0.101-0.2+b1 [92.0 kB]
Get:2 http://deb.devuan.org/merged daedalus/main amd64 libopeniscsiusr amd64 2.1.8-1 [59.7 kB]
Get:3 http://deb.devuan.org/merged daedalus/main amd64 open-iscsi amd64 2.1.8-1 [312 kB]
Fetched 464 kB in 2s (187 kB/s)
Preconfiguring packages ...
Selecting previously unselected package libisns0:amd64.
(Reading database ... 188375 files and directories currently installed.)
Preparing to unpack .../libisns0_0.101-0.2+b1_amd64.deb ...
Unpacking libisns0:amd64 (0.101-0.2+b1) ...
Selecting previously unselected package libopeniscsiusr.
Preparing to unpack .../libopeniscsiusr_2.1.8-1_amd64.deb ...
Unpacking libopeniscsiusr (2.1.8-1) ...
Selecting previously unselected package open-iscsi.
Preparing to unpack .../open-iscsi_2.1.8-1_amd64.deb ...
Unpacking open-iscsi (2.1.8-1) ...
Setting up libopeniscsiusr (2.1.8-1) ...
Setting up libisns0:amd64 (0.101-0.2+b1) ...
Setting up open-iscsi (2.1.8-1) ...
Processing triggers for libc-bin (2.36-9+deb12u10) ...
Processing triggers for man-db (2.11.2-2) ...
Processing triggers for initramfs-tools (0.142+deb12u1) ...
update-initramfs: Generating /boot/initrd.img-6.1.0-33-amd64
live-boot: core filesystems dm-verity devices utils udev blockdev iscsi dns.
#
The package installed properly.
Attempt to run the application using your same parameters:
$ iscsiadm --mode discovery --type sendtargets --portal 192.168.178.100
sh: 1: /bin/systemctl: not found # <---- same error
iscsiadm: can not connect to iSCSI daemon (111)!
iscsiadm: Could not make /etc/iscsi/send_targets: Permission denied # <---- this neeeds root
iscsiadm: Could not add new discovery record.
Same error.
Try as root, again using your same parameters:
# iscsiadm --mode discovery --type sendtargets --portal 192.168.178.100
sh: 1: /bin/systemctl: not found # <---- same error
iscsiadm: can not connect to iSCSI daemon (111)!
iscsiadm: Could not scan /sys/class/iscsi_transport.
sh: 1: /bin/systemctl: not found # <---- same error
iscsiadm: can not connect to iSCSI daemon (111)!
iscsiadm: Cannot perform discovery. Initiatorname required.
iscsiadm: Could not perform SendTargets discovery: could not connect to iscsid
#
Same error.
---
From what I see, it seems that this is a problem with the open-iscsi_2.1.8-1_amd64 package.
It requies a Devuan bug report to the maintainers.
See here on how to do that.
Note: Being a topic not related to the original post, please continue anything related to this issue on another, new thread.
Best,
A.
Hello:
... nothing with the systemd gets installed ...
Indeed.
As it should be.
This provided that the package installed (like the printout you posted shows) actually comes from the Devuan repository.
Like I pointed out, the problem* posted by kapqa (not related to the original post) would seem to indicate that the package installed may not be from the Devuan repository. * printout points to a typical error in Linux systems using systemd.
That said, note that the link posted leads to a Debian open-iscsi wiki page.
@kapqa
You may want to consider opening a separate thread with your problem and start by posting the terminal printout from this:
$ uname -a
$ cat /etc/apt/sources.list
$ apt list | grep -i installed | grep -i open-iscsi
Best,
A.
Hello:
... using "open-iscsi" there is error with devuan
https://wiki.debian.org/SAN/iSCSI/open-iscsisystemctl not found
The printout points to a typical error in Linux systems using systemd.
Devuan is not one of them.
From the web:
If systemctl is not found on your system, it usually means the systemd service manager is not installed or configured correctly.
This can happen if the system is not fully updated or if you've removed systemd accidentally.
To fix this, you'll need to install or reinstall systemd.
systemd has been intentionally removed from Devuan, no failed update or accident to speak of there. 8^°
The installed open-iscsi application is looking for systemctl (part of the systemd package) but cannot find it, hence the error printout.
At the risk of stating the obvious, this should not happen* if the installed open-iscsi package is the one present in the Devuan repositories.
* unless there has been some error in the processing / sanitising for use in Devuan for which a bug report should be filed with the Devuan maintainers.
HTH.
Best,
A.
Hello:
... spelling mistake is one thing ...
Crikey !...
Seems espresso had not kicked in yet.
But not a proper excuse though.
The thing is that I'm just lowly hack who most always finds the answers to difficult stuff under the same two names.
ie: fsmithred / ralph.ronnquist
Yes, that's the best (true one) I can come up with. 8^D
Best,
A.
Hello:
... disable this kernel module either modularly ...
Does not seem possible.
I have not found a way to disable any of those modules.
ie: ima, evm, selinux, etc.
Whatever methods I found searching on-line did not work.
The main thing to disable would be LSM which seems to orchestrate all of them, including this latest Microsoft contribution to the Linux kernel.
But I have not been able to find a working method.
... or when building the kernel ...
Right ... 8^°
... distros may integrate ...
Debian obviously does, no options to disable or heads-up given.
No surprise there ...
As a result, Devuan is stuck with all this crap.
Best,
A.
Hello:
3 years later ...
... anything with "viber" in it. # <- Actually it was ralph.ronnquist
There is no viber package in the Debian repositories.
As a result, there is no viber package in the Devuan repositories either.
The reason being (most probably) that viber is "Proprietary cross-platform IM and VoIP software".
With respect to it working on Arch, bear in mind that the package in queston is part of their user repository ie: not an official AUR package.
DISCLAIMER: AUR packages are user produced content. Any use of the provided files is at your own risk.
Seems that none of that has changed in the last three years.
Best,
A.
Hello:
That is one of the basic characteristics these security features have.
The main one one being that they are both installed and enabled by default / without your consent or knowledge.
Some fresh news with respect to LSM (Linux Security Modules).
Yes, those modules.
The ones you cannot disable or have any control over.
Microsoft's newest open-source contribution to the Linux kernel being proposed is ... Hornet,
a Linux security module (LSM) for providing signature verification of eBPF programs.
About eBPF
TL;DR
What is eBPF used for?
eBPF lets you gather detailed information about low-level networking, security, and other system-level activities within the kernel.
Better yet, it works without requiring direct modifications to kernel code.
Absolutely wonderful !!!
Just what we were needing.
Linux kernel with Microsoft security module integration doing all of that.
Check the Phoronix note here.
Best,
A.
Hello:
... same errors at boot time.
These are not errors.
What you are seeing is information related to EVM being enabled, not errors.
And it seems that there is no easy or documented way to avoid / disable security which (for a desktop system) is probably not needed.
As always, YMMV.
That said, the usual/basic way to check for errors is to look at dmesg in a terminal:
1. in a line by line fashion:
sudo dmesg | more
2. using grep:
sudo dmesg | grep -i "error\|warning\|fail\|segfault\|fatal\|not"
3. sifting by type of message
sudo dmesg --level=alert,crit,err,warn
4. with a real time rolling printout:
sudo dmesg -wH
As this is Linux*, there are probably more ways to get that done as there are other logfies in human readable format which you can look at to get a more detailed idea as to what is going on.
* 8^D !
Best,
A.
Hello:
... sysvinit follows a design principle which states that designs and/or systems should be as simple as possible.
Maybe I'm just as dumb as a (rusty) doornail, but after all these years (~12) with Linux I still fail to grasp the need for anything but one properly working init package for my systems* which have always worked wthout a hitch.
* desktop, netbook, RPi
Of course, as many others have, I came across more than one issue.
But none in any way related to the default init software, which was always* sysvinit.
* save experiments with Debian/Debian based dists which I abandoned as systemd was adopted.
... *nix method of do one thing and do it well, a concept seemingly lost to many of today's programmers.
+1
Best,
A.
Hello:
... sounds like a pain to get rid of and/or disable ...
Indeed ...
That is one of the basic characteristics these security features have.
The main one one being that they are both installed and enabled by default / without your consent or knowledge.
Something that should be getting everyone thinking about it and the reasons for it being so.
We are slowly but steadily arriving at the point where booting a computer will require signatures, code or features over which you will have no control or access to.
Unless certain requirements are met, of course.
Best,
A.
Hello:
... if you haven't disabled apparmor ...
I disabled apparmor from the first time I saw it has been installed and enabled without my intervention.
So no, I do not have the problem you have.
My way of dealing with it is adding security=none apparmor=0 nmi_watchdog=0 to my kernel command line.
That said, I am not too sure the stanza is quite as effective as I believe it is because early on, my dmesg printout also reveals this:
--- snip ---
[ 3.066032] evm: Initialising EVM extended attributes:
[ 3.066218] evm: security.selinux # <-
[ 3.066338] evm: security.SMACK64 (disabled)
[ 3.066493] evm: security.SMACK64EXEC (disabled)
[ 3.066660] evm: security.SMACK64TRANSMUTE (disabled)
[ 3.066842] evm: security.SMACK64MMAP (disabled)
[ 3.067009] evm: security.apparmor # <-
[ 3.067132] evm: security.ima # <-
[ 3.067239] evm: security.capability # <-
[ 3.067369] evm: HMAC attrs: 0x1
--- snip ---
As you can see, evm* does not indicate apparmor (and other security features) as being disabled.
ie: adding security=none to the kernel command line should have disabled all that.
* https://linux-ima.sourceforge.net/linux … l-20110907
Edit:
At some point, someone posted a request to disable EVM and IMA.
While reading and testing LSM code, I found IMA/EVM consume per inode
storage even when they are not in use. Add options to diable them in
kernel command line. The logic and syntax is mostly borrowed from an
old serious [1].
I have tried (adding lsm= to the kernel command line but it does not work, at least in the latest Daedalus 6.1.0-33-amd64:
$ grep -o "lsm=.*" /proc/cmdline
lsm=
$
$ ls /sys/kernel/security/
evm ima integrity lockdown lsm # <- this should read "integrity lsm"
$
Apparently it requires a patch. (?)
Best,
A.
Hello:
Filed using ...
Hmm ...
Good luck with that. 8^°
Last time I followed an issue here at Dev1, it required a Debian bug report, which I encouraged the author of the OP to file.
The reply was this:
Since sysvinit is not enabled by default in Debian, I do not consider this
bug as release-critical. Downgrading the bug severity to "normal"*.
* "won't fix" "get lost".
You may (or may not) want to consider reading the whole Dev1 thread and the Debian bug report to get an idea of the severity of the (properly investigated) problem the OP reported to see where any non-systemd bug report stands.
Best,
A.
Hello:
Thank you ...
You're welcome.
... seems to be in Caja.
... it's nasty ...
... 100's of mb's ...
... critical error ...
... unattended to ...
... permanently screw up a solid state drive.
Best,
A.
Hello:
... have the .xsession-errors down to almost zero ...
As long as they generate no issues ... *
... i'd like to figure out these things and address them ...
I thought likewise at one time but all the answers I found were in line with that I say above.
Like you, I don't like those errors either, but with most it's like chasing a rainbow.
So I ended up resorting to having a size limit on the /home/.xsession-errors file.
Check out this post.
* gist of most if not all the replies I got when asking about the entries in the huge /home/.xsession-errors file.
Best,
A.
Hello:
... a huge amount of .xsession-errors ...
Yes, /home/.xsession.errors can get quite fat with just harmless informative printout.
See here: https://dev1galaxy.org/viewtopic.php?id=2412
and
here: http://www.daniloaz.com/en/how-to-preve
Best,
A.
Hello:
... postinst script of xz is buggy ...
I see.
... doesn't recognize that the "alternatives" link for lzma already points correcly.
Right.
Makes sense. 8^)
Maybe the script ...
... when the new "alternatives" link get installed ...
... ends up with the same value that it already had.
Thank you for taking the time to figure that out.
I was worried that an unknown something had mucked around the system.
ie: the (manually or by a script) part of the postinst printout.
Not having been me, the only alternative offered was a script.
Not good.
My only intervention (as far as my memory serves me) in update-alternatives was long ago to set the default editor.
For obvious reasons, not vim.
Or emacs. 8^° !
Remember that the postinst script doesn't "see" those pathnames the way you see them in the error message.
Remember?
I did not have a clue, I'll have to learn it first.
Thanks for the heads up.
... sure there are other possible stories that could explain ...
And I'm quite sure that you have worked out the most probable | feasible | likely one.
So I will mark this one as [solved].
As always, thank you very much for your input.
Best,
A.
Hello:
... upgraded the xz-utils package ...
Yes, of course.
My post was about the warnings and whatever reason could have caused them.
Do have a read at the OP.
Best,
A.
Hello:
... because you updated the lzma package(s) for two different arches?
No idea.
That package is probably in my system since ascii or Jesse and I am now running on Daedalus, to which I arrived through succesive dist-upgrades.
Now ...
Looking at yesterday's timeshift snapshot of /etc/alternatives/lzma I see this:
/etc/alternatives/lzma: symbolic link to ../../usr/bin/xz
In the current version, I see the same thing:
/etc/alternatives/lzma: symbolic link to ../../usr/bin/xz
/usr/bin/xz exist in both cases, same size but different timestamp: 02/12/2023 and 03/04/2025 respectively.
So I don't get the part where apt reports ... /lzma has been changed ... and ... /xz is unknown.
Edit:
I checked to see how update-alternatives read:
$ update-alternatives --get-selections | grep lzma
lzma auto ../../usr/bin/xz
$
So it seems it is on auto and not manual, maybe apt found it had been changed?
Edit_2:
I recalled I had once seen /var/log/alternatives.log.
The printout is rather unreadable but I have tried to split it so as to see what it is all about:
$ cat /var/log/alternatives.log
update-alternatives 2025-04-06 09:30:57: run with
--install /usr/bin/lzma lzma /usr/bin/xz 20
--slave /usr/share/man/man1/lzma.1.gz lzma.1.gz /usr/share/man/man1/xz.1.gz
--slave /usr/bin/unlzma unlzma /usr/bin/unxz
--slave /usr/share/man/man1/unlzma.1.gz unlzma.1.gz /usr/share/man/man1/unxz.1.gz
--slave /usr/bin/lzcat lzcat /usr/bin/xzcat
--slave /usr/share/man/man1/lzcat.1.gz lzcat.1.gz /usr/share/man/man1/xzcat.1.gz
--slave /usr/bin/lzmore lzmore /usr/bin/xzmore
--slave /usr/share/man/man1/lzmore.1.gz lzmore.1.gz /usr/share/man/man1/xzmore.1.gz
--slave /usr/bin/lzless lzless /usr/bin/xzless
--slave /usr/share/man/man1/lzless.1.gz lzless.1.gz /usr/share/man/man1/xzless.1.gz
--slave /usr/bin/lzdiff lzdiff /usr/bin/xzdiff
--slave /usr/share/man/man1/lzdiff.1.gz lzdiff.1.gz /usr/share/man/man1/xzdiff.1.gz
--slave /usr/bin/lzcmp lzcmp /usr/bin/xzcmp
--slave /usr/share/man/man1/lzcmp.1.gz lzcmp.1.gz /usr/share/man/man1/xzcmp.1.gz
--slave /usr/bin/lzgrep lzgrep /usr/bin/xzgrep
--slave /usr/share/man/man1/lzgrep.1.gz lzgrep.1.gz /usr/share/man/man1/xzgrep.1.gz
--slave /usr/bin/lzegrep lzegrep /usr/bin/xzegrep
--slave /usr/share/man/man1/lzegrep.1.gz lzegrep.1.gz /usr/share/man/man1/xzegrep.1.gz
--slave /usr/bin/lzfgrep lzfgrep /usr/bin/xzfgrep
--slave /usr/share/man/man1/lzfgrep.1.gz lzfgrep.1.gz /usr/share/man/man1/xzfgrep.1.gz
update-alternatives 2025-04-06 09:30:57: status of link group /usr/bin/lzma set to manual
update-alternatives 2025-04-06 09:30:57: auto-repair link group lzma
update-alternatives 2025-04-06 09:30:57: status of link group /usr/bin/lzma set to auto
$
Thanks for your input.
Best,
A.
Hello:
Thanks for the fast reply.
Much appreciated.
The link has been renewed ...
I was referring to what came afterwards:
--- snip ---
update-alternatives: warning: /etc/alternatives/lzma has been changed (manually or by a script); switching to manual updates only
update-alternatives: warning: forcing reinstallation of alternative ../../usr/bin/xz because link group lzma is broken
update-alternatives: warning: current alternative ../../usr/bin/xz is unknown, switching to /usr/bin/xz for link group lzma
--- snip ---
About the lzma package:
$ apt list | grep -i lzma
lzma-alone/stable 9.22-2.2 amd64
lzma-alone/stable 9.22-2.2 i386
lzma-dev/stable,stable 9.22-2.2 all
lzma/stable 9.22-2.2 amd64
lzma/stable 9.22-2.2 i386
~$
$ aptitude why lzma
i font-manager Suggests file-roller
p file-roller Suggests lzma
$
Seems apt found something amiss in /etc/alternatives/lzma.
ie: something ... has been changed ...
But what | why | who changed it?
That said, do I even need it?
Thanks for your input.
Best,
A.