You are not logged in.
Hello:
... my PCs with Daedalus ...
... have a file /etc/resolv.conf, no symlink.
I see ...
I take it that this is with connman 1.41-3 installed and modifications to the symlink from you.
Yes?
... file points to my local ethernet router IP as DNS (I am working with fixed IP adresses).
Yes, I think you can set it up with up to two different DNS servers.
Thanks for your input.
Best,
A.
Hello:
All this crap has me so annoyed.
It has kept me looking all over the web for anything related to connman+resolv.conf.
There are a great many posts complaining about this same issue, going back as far as 2017 ...
I did find an interesting tidbit with respect to the origins of connman and networkmanager which I could not have imagined.
ConnMan is a product of Intel, and NetworkManager is a product of RedHat.
--- snip ---
Who would have guessed?
Those two heavyweights behind a simple application ...
[tin foil]
Make me wonder just much interest they would have in other (competing?) and much less complicated not to mention unobtrusive applications such as WiCD failing or simply just being abandoned. Because, besides connman, networkmanager and WiCD ...
What else is there for Linux?
[/tin foil]
I also found the same opinion/idea already seen in a few forums:
Whoever set up ConnMan figures that the sysadmin has no knowledge of use, and so it does everything itself.
See here.
Also a couple of possible solutions:
When connmanctl borks up whatever symlink /etc/resolv.conf points to this week, particularly with ipv6 nonsense, you can easily set it back to the real nameservers:
# select config id of currently connected network current="$( connmanctl services | awk '/^[^ ]/ && $1~/R/{print $3}' )" connmanctl config "${current}" --nameservers 192.168.1.10 192.168.1.11
Or you could just clobber the resolv.conf with a real file and not a symlink to /etc/connman/resolv.conf. Crap like this makes me miss wicd although I'm sure it had its problems too.
See here.
This last one (originally suggested by Ralph.Ronnquist) is what I am interested in doing but I was not so sure as to how to go about it.
Then I found man connman.conf with the explanation to an an entry I had not seen/noticed before and which could be a solution:
FILE FORMAT
The configuration file consists of sections (groups) of key-value pairs.
--- snip ---
Description of sections and available keys follows:[General]
This section is the only mandatory section of the configuration file.
--- snip ---
ResolvConf=string
Path to resolv.conf file. If the file does not exist, but intermediate directories exist, it will be created. If this option is not set, it tries to write into /run/connman /resolv.conf and fallbacks to /etc/resolv.conf if it fails (/run/connman does not exist or is not writeable). If you do not want to update resolv.conf, you can set /dev/null*.
* underlining is mine
See here.
But it happens that the entry for etc/connman/main.conf mentioned above is not present in that file so I think that the best go-to solution would be to clobber the resolv.conf with a real file and not a symlink to /etc/connman/resolv.conf. ie: blunt force, like with a kitchen roach.
Curiously enough, although I have had my box on-line for almost nine hours, so far there has not been a DNS change (!).
So I really do not have a clue as to what is going on with connman or how it decides when to do what it does.
Truth be told, I just want WiCD back. 8^°
Thank your for the instructions, I was thinking of doing more or less the same thing.
What kept me back is the assumption that as it had been set up by connman doing so would break it and things would get worse.
I will wait and see what happens / till I get another DNS change from connman and then proceed.
Thank you very much for your input.
Best,
A.
Hello:
... will report back ...
No dice.
I have found no way to apply chattr to resolv.conf.
Tried with repair mode and booting a Knoppix.
Probably because it is a symlink. (?)
The strange thing is that while my local DNS address is set in both the fibre router settings page and in the connman settings page and it does survive a reboot, connman does not change the resolv.conf file right away eg: at boot time.
... would be to purge resolvconf and reinstate /etc/resolv.conf as a text file.
I find it quite disturbing that the possible solutions to this absurd situation have to be that blunt (so to speak).
We should be able to set the application to do as we want instead of forcefully blocking it from doing something we do not want it to do as is the case with changing the attributes of a file and deleting a working symlink.
Does not speak well of whoever wrote connman and decided that it was a good thing to so.
@ralph.ronnquist
I don't have the resolveconf package installed. so the etc/resolve.conf -> /run/connman/resolv.conf was set up by connman.
~$ ls -l /etc/resolv.conf
lrwxrwxrwx 1 root root 24 Sep 18 06:48 /etc/resolv.conf -> /run/connman/resolv.conf
~$
So there's no resolveconf to purge.
I'd appreciate instructions on how to proceed with your solution.
---
Edit:
I found this this tidbit from 2015 (!) when this mess was already a thing on the web.
... a patch for connman that prevents it from changing resolv.conf, but you'll have to recompile it from source and use it on your own risk.
... another option - to protect your /etc/resolv.conf file using the chattr command:
chattr +i /etc/resolv.conf
A patch?
Recompile and use at your own risk?
Talk about blunt.
Unless I am far off, it would seem that the option to change /etc/resolv.conf attributes at some time actually worked.
ie: the file was not a symlink, just a text file and chattr +i could actually be used.
But then someone thought it would be a good idea to nip that one in the bud.
Will wonders never cease?
---
Thanks in advance.
Best,
A.
Hello:
And we'll see how that goes.
Badly or not-as-expected.
My conky setup queries my DNS and I can see it on-screen.
One moment it was as-expected and the next time I looked it had changed.
As posted previously, I has left it at this ...
~$cat /etc/resolv.conf
# edited by user 17092024
nameserver 192.168.1.11
~$
... and now it is this:
~$ cat /etc/resolv.conf
domain Home
search Home
nameserver [my fibre provider DNS]
nameserver [my fibre provider DNS]
~$
Maybe I did not do the chattr -f +i /etc/resolv.conf routine correctly?
Cannot say as I am unable to check the file's attrbutes:
~$ lsattr /etc/resolv.conf
lsattr: Operation not supported While reading flags on /etc/resolv.conf
~$
Hmm ...
I finally found what the -f modifier does.
It happens that it does not mean force, it just suppresses error messages.
I will edit the file again, boot into recovery/repair mode and try to set it from there.
What a load of crock ... 8^/
I will report back once I am done.
Best,
A.
Hello:
... /etc/resolv.conf is a link ...
Yes.
mc sees it as @resolv.conf.
... suggests you also have the package resolvconf ...
Checked but no, it is not in my system:
~$ apt list | grep installed | grep resolvconf
--- snip ---
~$
~$ apt list | grep resolvconf
--- snip ---
resolvconf-admin/stable 0.3-1 amd64 # not installed
resolvconf-admin/stable 0.3-1 i386 # not installed
resolvconf/stable,stable 1.91+nmu1 all # not installed
~$
Would that be a bit less confusion?
... happens regardless of which networking software you use; whether it's plain ifupdown, wicd, network-manager or connman.. or whatever.
I see.
The thing is that (for the time being) everything works as per my needs with the chattr fix for /etc/resolv.conf.
The mystery regarding the how/why of the presence of NetworkManager files in my system remains.
ie: I seriously doubt I had anything to do with that, this system has (at least) been upgraded from ascii or maybe even Jesse.
But I recall always having used WiCD.
Maybe the default in Jesse or ascii was NetworkManager and was not properly/totally purged? Would not be surprised.
But it is too long ago for me to remember the details.
Q1: what to do with all these?
/etc/NetworkManager
/etc/NetworkManager/NetworkManager.conf
/etc/NetworkManager/dispatcher.d
/etc/NetworkManager/dispatcher.d/ntpsec-ntpdate
/usr/lib/NetworkManager
/usr/lib/NetworkManager/conf.d
/usr/lib/NetworkManager/conf.d/no-mac-addr-chan
ge.conf
... purge resolvconf and reinstate /etc/resolv.conf as a text file.
Makes sense ...
But /etc/resolve.conf links to /run/connman/resolv.conf which is a file produced by connman, not NetworkManager or resolveconf which is not installed in my system:
~$ readlink -f /etc/resolv.conf
/run/connman/resolv.conf
~$
Q2: would that not break something?
The more I go down this hole, the more (and more) I agree with that chap from stackexchange. 8^°
I have now edited /etc/resolv.conf.
From this:
~$ cat /etc/resolv.conf
# Generated by Connection Manager
search Home
nameserver 192.168.1.11
~$
To this:
~$cat /etc/resolv.conf
# edited by user 17092024
nameserver 192.168.1.11
~$
And we'll see how that goes.
Thank you very much for your input.
Best,
A.
Hello:
Rather embarrassed, I am at a loss to explain or understand what is going on, so I'd appreciate some input on this.
As per this thread regarding whatever goes on these days with /etc/resolv.conf and how to solve the DNS issues involved, I understood that my box running a default Devuan Daedalus/XFCE installation ran with network-manager installed.
I had no doubts with respect to that being so because /etc/resolv.conf clearly read:
~$ cat /etc/resolv.conf
# Generated by NetworkManager <- # this
# search home
--- snip ---
~$
Not only that, the file was being constantly rewritten with the # Generated by NetworkManager banner at the top.
The file's timestamps was sufficient proof, so what to think?
There are also NetworkManager files all over the system:
/$ locate NetworkManager
/etc/NetworkManager
/etc/NetworkManager/NetworkManager.conf
/etc/NetworkManager/dispatcher.d
/etc/NetworkManager/dispatcher.d/ntpsec-ntpdate
--- snip ---
/usr/lib/NetworkManager
/usr/lib/NetworkManager/conf.d
/usr/lib/NetworkManager/conf.d/no-mac-addr-change.conf
/$
Not so fast ...
This morning, while doing further research on the matter I found a thread that led me to have a look at /etc/init.d, coming across this surprise:
~$ ls /etc/init.d | grep network-manager
~$
Ooops ...
A further manual inspection showed me this:
~$ ls -1 /etc/init.d -v
README
--- snip ---
connman~
connman
--- snip ---
networking
nfs-common
--- snip ---
~$
ie: there is a connman entry but there is no network-manager entry
So I asked apt:
~$ apt list | grep installed | grep network-manager
--- snip ---
~$
~$ apt list | grep installed | grep connman-
--- snip ---
connman-gtk/stable,now 1.1.1+git20180626.b72c6ab-3 amd64 [installed]
connman-ui/stable,now 0~20150623-1+b2 amd64 [installed]
~$
What is going on?
Could it be possible that connman does not over write the whole /etc/resolv.conf file and just edits the DNS addresses?
And where did the # Generated by NetworkManager banner come from?
I do not/can not recall using anything but WiCD from Jesse or ascii onwards.
This is a Beowulf -> Chimaera -> Daedalus upgrade which had quite a few issues, all finally solved.
Beowulf ran with the last available WiCD, no idea what Chimaera upgraded to as the desktop and most XFCE settings were mangled up during the upgrade making it unusable so I decided to continue on to Daedalus.
Only then was I able to finish settings things up properly.
For the moment the fix posted by fsmithred seems to have worked. (via chattr)
ie: survived this morning's boot but I'll have to wait and see what happens if/when my fibre provider decides to do one of their remote router resets.
In any case, from what I have read on-line, both connman and network-manager have a go at /etc/resolv.conf, so it will be back to WiCD when it finally gets to the Debian/Devuan repositories.
Any comments welcomed.
Thanks in advance,
A.
Hello:
... something wrong. Well, there was: the provider's equipment.
I don't think so.
For reasons I can only adscribe to syncronicity, serendipity, karma or just bad luck my box started behaving like yours.
This in spite of having set my fibre router to use my specific DNS.
ie: PiHole/Unbound in a VM running Devuan Chimaera.
Before that, it happened once every two weeks or so, when the fibre provider did a remote reset of the router, the result being the DNS1 and DNS2 settings being restored to their default IP addresses.
But, if my provider's router is set to use my own DNS (192.168.1.11) and NetworkManager's settings are set to use that very same IP adress why in holy fuck's name is this POS NetworkManager setting /etc/resolv.conf to an IP address reflecting one of my fibre provider's DNS servers?
I found the answer here.
It seems that every clueless moron who writes some half-arsed network management tool or script thinks that it is acceptable to blow away a hand-crafted /etc/resolv.conf and replace it with some garbage that only works in the imaginary fantasy-land of the author's imagination, not in the real world. If you don't want your /etc/resolv.conf being mangled by programs like systemd-resolved or network manager, then you need to either a) configure them to stop doing that, or b) stop using them. In your case, something (probably systemd-resolved) has replaced your /etc/resolv.conf with a symlink. –
Unfortunately, I have not been able to make option a) work and b) needs WiCD to get made into the Debian/Devuan repositories.
Other available options seem to be worse.
The author also (quite eloquently) said this:
... said it before but apt-get purge is effective but unsatisfyingly inadequate, there should be a --kill-it-with-fire or --banish-to-hell option for miserable system-breaking junk like that. –
I wholeheartedly agree with his opinion, NetworkManager fits the description perfectly.
In short:
I tried all the solutions I found posted on line and (up to now) the only one that worked (jury still out) is the one posted by fsmithred here.
But there's a twist.
As /etc/resolv.conf is a symlink, the right syntax to use is this:
~$ sudo chattr -f +i /etc/resolv.conf
Otherwise you get an error:
~$ sudo chattr +i /etc/resolv.conf
--- snip ---
chattr: Operation not supported while reading flags on /etc/resolv.conf
~$
So now we know ... 8^°
Best,
A.
Hello:
Nothing stopping another developer (to be?) to stand up ...
I guess not.
But the fact is that between the last release by a member of the WiCD project team and Andreas' starting to work on it, over 12 years passed.
We can shorten that to 7 if we count Pieter Leclerc's contribution (not listed as a team member) by releasing 1.7.3 (12/2014) and squashing ~30 bugs. That was his only (albeit very valuable) contribution.
But that is it, no one else till Andreas stepped up.
There is even a request for membership dating from 2021 still pending approval.
What I fail to understand is the lack of communication from the owner/members of the WiCD project team.
eg:
We do not have any time or willingness to further develop/maintain WiCD and as a result the project will be permanently archived as of mm/yyyy.
If interested in continuing development/maintenance of the WiCD project, please contact us so we can transfer the licence and control of the code/repository.
... advantage of getting that code away from Microsoft's (github) control.
Indeed.
Would be a very good thing the whole thing moved away from github.
Best,
A.
Hello:
Andreas Messer is no stranger to Devuan.
Or to me ... 8^)
We exchaged quite a few emails with respect to WiCD and testing back in 2022.
Best,
A.
Hello:
... that Takahiro Yoshizawa will donate his work to the official branch ...
... that Andreas will accept the code for mainline.
I don't have much of a clue as to all that would entail.
eg: with respect to the person listed as the owner of the project or the people listed as active members of the 'Wicd-devel' team.
Andreas Messer joined in 09/2021 and started to work on the WiCD project which, to all intents and purposes, had been dead-in-the-water since the bugfixing release (1.7.3 - 12/2014) authored by Pieter Leclerc who, for whatever reason, is not listed as a member of the team.
Things probably got complicated for Andreas and in spite of all his efforts nothing much happened till now but in all fairness, he seems to be the only active member of the WiCD project since 04/2012 with the release of 1.7.2 by David Paleino.
ie: over 12 years ...
With the publishing of the python2 -> python3 work done by Takahiro Yoshizawa, WiCD is now back in the spotlight.
Kudos to him for that.
Hopefully, the WiCD project will soon be put under new management (or forked?) and, in time, recover its place in the Debian/Devuan repositories.
And in all my Devuan installations, where it is sorely missed.
Best,
A.
Hello:
... happy like a child ...
That > | makes > | two > | of > | us!
And probably many more.
I know four members from Dev1.
... how to thank hanaguro ...
I'm not sure but it seems (?) hanaguro may be Takahiro Yoshizawa's handle.
(see authors list to the right of the page)
He is the author (launchpad) of every commit from 2024-06-05 onwards.
He has been busy,
Now I have to find out how to thank hanaguro ...
I had an UbuntuOne account from long ago and logged in but found no way to contact the chap.
Thanks for the heads-up.
Best,
A.
Hello:
... from "professors" who haven't tried the distro
Surprised?
"Genuine question, what's the issue with systemd in terms of reliability for a day to day desktop focused distro?"
Indeed ...
As genuine as a wooden nutmeg.
Best,
A.
Hello:
NetworkManager resets the resolv.conf file ...
Quite so.
... the ISP's problem.
No.
It is the default behaviour for NetworkManager. 8^°
--- snip ---
dns
Set the DNS (resolv.conf) processing mode.default: The default if the key is not specified. NetworkManager will update
resolv.conf to reflect the nameservers provided by currently active connections.dnsmasq: NetworkManager will run dnsmasq as a local caching nameserver, using a "split
DNS" configuration if you are connected to a VPN, and then update resolv.conf to point
to the local nameserver.none: NetworkManager will not modify resolv.conf.
--- snip ---
https://people.freedesktop.org/~lkundra … .conf.html
See these two posts for a solution:
https://dev1galaxy.org/viewtopic.php?pid=47932#p47932
and
https://stackoverflow.com/questions/517 … esolv-conf
Best,
A.
Hello:
~$ cat /etc/resolv.conf
# Generated by NetworkManager
search home
nameserver 192.168.1.1
nameserver 2a02:1210:3236:f500:a2b5:49ff:feb7:9ea0
~$
Right ...
Let's try something.
As sudo, please edit resolv.conf so that it ends up like this:
~$ cat /etc/resolv.conf
# Generated by NetworkManager
# search home
#
nameserver 8.8.8.8
nameserver 8.8.4.4
#
# nameserver 192.168.1.1
# nameserver 2a02:1210:3236:f500:a2b5:49ff:feb7:9ea0
~$
Save the file, reboot and see if your problem is still there.
If things work, now you know that you have some DNS configuration problem.
TL;DR
192.168.1.1 is (most probably) a/your router's IP address.
2a02:1210:3236:f500:a2b5:49ff:feb7:9ea0 is a valid IPv6 address but I cannot say what it belongs to, maybe the router also.
You can check it here: http://sqa.fyicenter.com/1000334_IPv6_A … idator.htm
By editing resolv.conf you will have set Google's DNS servers as nameservers for your box and things should work as expected now.
Bear in mind that Google's DNS servers are not the only ones you can use, you can also use your service provider's DNS servers or freely available public DNS servers.
Let us know how you fared with this.
Best,
A.
Hello:
traceroute never responds ...
Hmm ...
Could you please post the output of:
$ cat /etc/resolv.conf
Thanks.
Best,
A.
Hello:
... new to this forum.
Welcome to Dev1.
... install Pinta but it is currently only in the archived repositories ...
Packages in the Devuan repositories come directly from the Debian repositories 'as they are' unless they need to be sanitized/fixed because of their needing systemd. If for whatever reason that cannot be done, the package is then banned from Devuan repositories.
See: https://www.devuan.org/os/packages
Devuan package repositories are exclusive. Other repositories, including Debian, Ubuntu, Mint etc, should not be used directly.
If Pinta was once available for Devuan, it must have been (at some point) available in the Debian repositories.
ie: if it is not in the Debian repositories, it will not be in the Devuan repositories either.
The Pinta package seems to have been removed from Debian unstable last year.
See:
https://tracker.debian.org/news/1448883 … -unstable/
Probably because of a bug, cannot really say.
See:
https://bugs.debian.org/cgi-bin/bugrepo … ug=1041606
Package: ftp.debian.org
Severity: normal
User: ftp.debian.org@packages.debian.org
Usertags: remove
Control: affects -1 + src:pintaPlease remove pinta. It is RC-buggy and was not contained in several
releases. It depends on gtk2. There are several new upstream releases
which nobody cared to import to fix the RC issue.
But it is not in their repositories any longer.
See:
https://tracker.debian.org/pkg/pinta
Maybe contacting the package maintainer will get you more information.
See: https://www.pinta-project.com/contact
Best,
A.
Hello:
deb.devuan.org remains blocking ...
Still good here:
~$ date
Thu Sep 12 07:07:39 -03 2024
~$
~$ ping deb.devuan.org
PING deb.rr.devuan.org (95.216.15.86) 56(84) bytes of data.
64 bytes from megumin.packet-gain.de (95.216.15.86): icmp_seq=1 ttl=47 time=257 ms
64 bytes from megumin.packet-gain.de (95.216.15.86): icmp_seq=2 ttl=47 time=257 ms
--- snip ---
--- deb.rr.devuan.org ping statistics ---
12 packets transmitted, 12 received, 0% packet loss, time 12187ms
rtt min/avg/max/mdev = 256.589/256.769/256.948/0.120 ms
~$
As you can see, a ping to deb.devuan.org will act on deb.rr.devuan.org which would indicate that it is (?) working correctly.
Does a traceroute from this end of the world render any useful data?
~$ traceroute deb.devuan.org
traceroute to deb.devuan.org (103.146.168.12), 30 hops max, 60 byte packets
1 192.168.1.1 (192.168.1.1) 1.755 ms 1.706 ms 1.672 ms
2 XXX.XX.XXX.X (XXX.XX.XXX.X) 3.094 ms 3.149 ms 3.125 ms
3 10.192.4.36 (10.192.4.36) 3.152 ms 3.430 ms 3.947 ms
4 * * *
5 176.52.255.29 (176.52.255.29) 132.642 ms 10.192.17.1 (10.192.17.1) 4.898 ms 176.52.255.29 (176.52.255.29) 132.540 ms
6 213.140.39.116 (213.140.39.116) 4.794 ms * 1.374 ms
7 * 176.52.255.29 (176.52.255.29) 121.977 ms 127.727 ms
8 * ix-be-26.ecore1.mln-miami.as6453.net (66.110.72.30) 120.381 ms *
9 if-ae-45-2.tcore1.mln-miami.as6453.net (63.243.152.34) 148.806 ms 148.678 ms *
10 if-bundle-37-2.qcore1.mln-miami.as6453.net (66.110.9.112) 149.087 ms * *
11 if-ae-45-2.tcore1.mln-miami.as6453.net (63.243.152.34) 148.287 ms 147.865 ms 147.742 ms
12 * if-bundle-2-2.qcore2.aeq-ashburn.as6453.net (216.6.87.9) 148.239 ms *
13 if-ae-12-2.tcore4.njy-newark.as6453.net (66.198.155.33) 148.661 ms * 149.464 ms
14 if-ae-1-3.tcore3.njy-newark.as6453.net (216.6.57.5) 147.936 ms * *
15 if-ae-12-2.tcore4.njy-newark.as6453.net (66.198.155.33) 148.054 ms 209.58.124.6 (209.58.124.6) 343.277 ms if-ae-12-2.tcore4.njy-newark.as6453.net (66.198.155.33) 149.337 ms
16 * * *
17 121.240.236.6.static-Delhi.vsnl.net.in (121.240.236.6) 354.318 ms 354.241 ms 209.58.124.6 (209.58.124.6) 343.408 ms
18 103.196.223.166 (103.196.223.166) 355.967 ms 356.725 ms 354.336 ms
19 121.240.236.6.static-Delhi.vsnl.net.in (121.240.236.6) 355.794 ms 103.146.168.12.ipacct.in (103.146.168.12) 363.173 ms 361.877 ms
~$
Best,
A.
Hello:
... something wrong?
No, at least not where I am:
~$ date
Wed Sep 11 12:37:42 -03 2024
~$
~$ sudo apt update
---- snip ---
Get:1 http://deb.devuan.org/merged daedalus InRelease [43.0 kB]
Get:2 http://deb.devuan.org/merged daedalus-updates InRelease [33.4 kB]
Get:3 http://deb.devuan.org/merged daedalus-security InRelease [33.2 kB]
Get:4 http://deb.devuan.org/merged daedalus/main i386 Packages [8931 kB]
Get:5 http://deb.devuan.org/merged daedalus/main amd64 Packages [9043 kB]
Get:6 http://deb.devuan.org/merged daedalus-updates/main i386 Packages [2488 B]
Get:7 http://deb.devuan.org/merged daedalus-updates/main amd64 Packages [2492 B]
Fetched 18.1 MB in 7s (2663 kB/s)
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
1 package can be upgraded. Run 'apt list --upgradable' to see it.
~$
Also, have a look here.
May have to do with connman, dnsproxy and how they interact.
Best,
A.
Hello:
... FUD whether you accept it or not.
In my opinion, the concept of FUD is to a great extent and to say the least, subjective.
ie: without a proper evaluation of intent, purpose and context labelling something as FUD can be quite difficult if not risky.
To wit:
Alter Kim's post at the [devuan-dev] list was thoughtfully replied to by Mark Hindley (arguably Devuan's most prominent member) with a follow up by member tempforever with the addition of more information.
In both instances without any mention of FUD spreading and such. ie: intent, purpose and context were evidently considered.
In a rather surprising follow up, my post here at Dev1 in which I cited the OPs post was met with a rather different demeanor, even after my posting a reply with an explanation of sorts.
@ralph.ronnquist
While I have the utmost respect for your knowledge and contrbution to the Dev1 project, I cannot but strongly disagree with your characterisation of my post as FUD.
So I'll leave this at that and (as far as I am concerned) agree to disagree, so to speak.
Best,
A.
Hello:
A "bug report" is not necessarily a "bug".
Indeed ...
I was citing a post at [devuan-dev] and thought it was something to be taken into account.
See: https://lists.dyne.org/lurker/message/2 … a8.en.html
But also this:
Like the subject reads, this is just a heads-up on my behalf.
I know zilch about all this ie: is it really a concern?
--- snip ---
Opinions/suggestions on how to proceed from those who understand this better are welcome.
I think my post is a very (very) long way from even the possibility of being characterised as the spreading of FUD.
Or anything of the sort.
Same for the OP at [devuan-dev] who clearly acted in good faith and did his research
I did not see his post characterised as FUD by anyone there.
Quite the contrary.
As for me, after over seven years and 1.527 posts at Dev1 ...
FUD?
Do lighten up. 8^P !!!
Best,
A.
Hello:
thanks ...
... we'll keep watching ...
You're welcome.
Concurrently with the bug report to Devuan, this was posted to the [devuan-dev] list.
So I expect that comments/clarifications will get posted there first.
I wonder ...
Does this only affect Devuan? Debian is not affected?
Best,
A.
Hello:
Just received this.
My box runs on Devuan Daedalus, upgraded yesterday to 6.1.106-3:
~$ uname -a
Linux devuan 6.1.0-25-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.106-3 (2024-08-26) x86_64 GNU/Linux
~$
I ran the test and it seems my system suffers from this bug*:
*?
~$ ssh -G 2>&1 | grep -e illegal -e unknown > /dev/null && echo "System clean" || echo "System infected"
System infected
~$ uname -a
Like the subject reads, this is just a heads-up on my behalf.
I know zilch about all this ie: is it really a concern?
So I'll have to start reading up on it now, but not after I take my daily ration of espresso. 8^°
Opinions/suggestions on how to proceed from those who understand this better are welcome.
In any case, my workstation has no ssh access (port 22 closed), only the headless VM running PiHole+Unbound.
Thanks in advance,
A.
Hello:
Nothing looks amiss ...
... Daedalus is running fine.
Same here.
Maybe the reason is that all these updated packages only work properly on/with 6.1.0-25.
ie: not the previous version 6.1.0-23
Thanks for your input.
Best,
A.
Hello:
An odd question, if I may ...
My box runs on Devuan Daedalus:
~$ uname -a
Linux devuan 6.1.0-23-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.99-1 (2024-07-15) x86_64 GNU/Linux
~$
I keep it up to date by checking with apt at least once a week and my upgrades are usually for few packages.
This morning my sudo apt update revealed an unusual (?) number of packages to update:
~$ sudo apt update
--- snip ---
Fetched 67.1 MB in 21s (3147 kB/s)
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
67 packages can be upgraded. Run 'apt list --upgradable' to see them. # 67 packages
~$
I don't recall ever having seen this save when updating a system that had gone more than a few months without an update.
According to /var/log/apt/history.log my last update was on 20243108 when only four packages were updated.
Start-Date: 2024-08-31 07:16:44
Commandline: apt upgrade
Requested-By: user (1000)
Upgrade: libjavascriptcoregtk-4.1-0:amd64 (2.44.2-1~deb12u1, 2.44.3-1~deb12u1), libjavascriptcoregtk-4.0-18:amd64 (2.44.2-1~deb12u1, 2.44.3-1~deb12u1), libwebkit2gtk-4.1-0:amd64 (2.44.2-1~deb12u1, 2.44.3-1~deb12u1), libwebkit2gtk-4.0-37:amd64 (2.44.2-1~deb12u1, 2.44.3-1~deb12u1)
End-Date: 2024-08-31 07:16:49
Is anything amiss, did something happen to trigger such a large package update?
Or is it because of linux-headers-6.1.0-25-amd64 | linux-headers-6.1.0-25-common | linux-image-6.1.0-25-amd64?
Just asking.
Thanks in advance.
Best,
A.