112[02:02:05] <awalis> hello fellows, I am using the 5 first
DVDs of debian 10.8 as my local repo. as 10.9 is out, I'd like
to download the "update DVDs" there is 3 of them in total
on the jigdo page, I'd like to know how many of them I need
(corresponding to the first 5 DVDs I already have)
238[05:32:01] <Wafficus> Hi there, I have a question about the
gqrx port. When I try launching it in terminal I get "Control
port disabled "Pulseaudio error: Connection refused"
239[05:32:16] <Wafficus> I have pulse installed, but am not sure
what to do next to make it actually run
252[05:40:33] <somiaj> hmm, maybe you had something in yoru
config causing problems with pulse, glad you figured it out
253[05:40:35] <Wafficus> ^ got the fix from that thread
254[05:40:49] <Wafficus> yeah that's the weird thing I
think it automatically made some config stuff
255[05:41:00] <Wafficus> I only have ever config'd ALSA on
all my machines
256[05:41:19] <Wafficus> but that's cause I use a Focusrite
Solo interface sometimes for audio recording, and sometimes just
need headphone out, etc, so nothing major
257[05:41:25] <somiaj> Some apps may try to do that, if it keeps
happening you may have to track down which app is saving that file.
417[09:26:24] <dosmanak> Hello. How kernel modules are loaded
after boot nowadays? I don't see anything in /etc/modules nor
/etc/modules-load.d yet I have plenty of modules loaded.
424[09:31:23] <SymbioticFemale> version 2 .onions will be fully
depreciated and non-functional in like 2 months people
425[09:31:41] <dosmanak> somiaj: what about ip_tables e.g.?
426[09:33:10] <somiaj> dosmanak: I'm unsure on the exact
details (and if systemd helps with this at all), but probaly through
dependencies in loading the network hardware
427[09:33:24] <somiaj> also ip_tables is depricated and nftables
should be being used
428[09:34:37] <somiaj> dosmanak: but mostly as things have
become more automatic, I have paid less attention to the details,
and it actually appears that modprobe is doing the work
431[09:35:53] <somiaj> others come from dependencies, and there
might be some core things that are just loaded as part of booting.
Use of things like /etc/modules are for if you happen to have
hardware or module needs that don't get detected appropertly
432[09:35:58] <dosmanak> I am preparing little deb package, that
installs some kernel module as plugin for iptables and I realized I
am not sure how to make it load after reboot.
438[09:38:20] <somiaj> dosmanak: I guess /usr/lib vs /etc
depends on if you want this my_package.conf to be a confflie or not,
putting it in /etc/ will give more freedom for the user to edit it
if they need
439[09:38:26] *** Quits: Filohuhum (~filohuhum@replaced-ip) (Remote host closed the connection)
440[09:39:11] <dosmanak> No need to configure the thing, just
load. I find dh_installmodules not helpful here.
442[09:40:00] <somiaj> On my machine cups does use
/etc/modules-load.d -- sure it maybe just for loading, but putting
in /etc allows a user to comment out/adjust without updates to the
packages changing things
443[09:40:14] <somiaj> I'd personally use /etc, but either
works
444[09:42:10] *** Quits: Trixelit (~trixelit@replaced-ip) (Read error: No route to host)
611[13:35:08] *** Quits: endstille (~endstille@replaced-ip) (Quit: I'll be back.)
612[13:35:39] <ratrace> azy: you can install it from snaps
613[13:35:45] <machine_1490> Hi, I'm on Debian 10 trying to
print via USB cord connection to a Brother 1510 printer. As per
replaced-url
614[13:35:45] <machine_1490> reduction with convert before they
will work). It doesn't seem to be file size dependent (some
large 30MB file prints, while a smaller 1MB won't. I followed
that Debian Wiki instructions and installed the Brother driver
installer. No luck. Printing is still spotty. I tested evince,
okular, firefox... note that this printer works with Windows. How
617[13:37:59] *** Quits: igrtrrt (~igrtrrt@replaced-ip) (Remote host closed the connection)
618[13:38:02] <machine_1490> A possible issue is that I'm
on a 64 bit machine and some of the drivers downloaded via the
installer are 32 bit. But note that the Debian Wiki shows a mix of
64 and 32 bit drivers being used, so shouldn't be an issue,
I'd imagein.
631[14:03:10] *** Quits: Nokaji (~Nokaji@replaced-ip) (Quit: "... when the freedom they wished for most was
freedom from responsibility then Athens ceased to be free and was
never free again.” ~ Edward Gibbon (1737-1794) - Decline and
Fall of the Roman Empire, 1909)
648[14:16:52] <cws> terabit: commercial vpns are snake oil
garbage, you'd do well to save your money.
649[14:16:57] <terabit> ratrace: bro, I have absolutely no
reason to justify that to you. By your hostmask, you're not
affiliated with the project,my question was for those who can speak
in an official capacity.
653[14:17:33] <ratrace> terabit: first I'm not a
"bro". Second, no, you don't. I was just curious
about your contradicting yourself. Anyhoo... moving on.
654[14:17:51] <terabit> if it was corporate network I
understand, but for a community driven project this is really bad
655[14:17:55] <cws> terabit: It is, isn't it? Even better
that my opinion is supported by a mountain-range of evidence
approximately equivalent to the Himalayas, and you... are wasting
your money on snake-oil garbage. :)
656[14:18:28] <machine_1490> Sorry all, didn't want to
cause tension with same question in ##Linux and #debian. And thanks
to all.
657[14:18:32] <ratrace> also, maybe _that_ particular VPN
you're using is especially abusive. Some allow free,
unregistered "web proxy" access, easy to abuse.
658[14:19:37] <ratrace> (and third, irc masks aren't
indication of anything. I was FreeBSD committer and maintainer with
a mask that wasn't "freebsd")
659[14:19:38] <terabit> ratrace: it's an informal term
similar to "my friend" , I am aware we're not
related. I wasn't contradicting myself. I simply wanted to
bring to the attention of debian developers and admins. if the
project does not wish to allow people who use Tor, a VPN or some
other "origin hiding" (as you said) IPs to use Debian, I
can respect that decision
661[14:20:02] <terabit> cws: again, that's cute, i Have no
interest in having that debate with you, not the time or place.
662[14:20:18] <ratrace> maybe it's not about all VPNs but
that particular network
663[14:20:28] <terabit> machine_1490: you didn't, I could
have ignored it and moved on, I didn't because it's a
pet-peeve
664[14:20:44] <cws> Which indicates a particular talent in
picking the worst apple from a barrel full of rotten ones.
665[14:21:43] <ratrace> terabit: my comment about
"bro" wasn't about us being related. Please be
inclusive of all genders. I'm not a man, thus, I'm not not
a "bro".
666[14:21:55] <terabit> ratrace: which makes sense, but in
general it's bad practice to ban IPs unless you have no other
option. This would prevent me from using debian and thousands of
other users will be excluded to. my point was not to rant but to let
the devs know this block has serious impact
667[14:23:08] <ratrace> terabit: then maybe try the OFTC chans,
those are more "official".
668[14:23:31] <terabit> ratrace: "bro" just like
"guy" is not a gender specific term in this context unless
you misunderstood as my meaning to call you a relative. It is not
formal english so I can understand why you would misunderstand. I
don't frequent this channel but if I do I'll avoid that
term
669[14:23:41] <terabit> ratrace: thanks, I will
670[14:24:01] <Unit193> terabit: Huh, that may be a specifically
abused VPN. I just tried with tor and at least the exit I'm
using viewed the page fine.
671[14:24:36] <Unit193> Perhaps still odd to block read-only
unless it's specifically agressive.
672[14:25:28] <Ede|Popede> on what level does that block happen?
network, httpd, scripting, ...?
673[14:26:01] <terabit> Unit193: It's proton, they're
main stream. maybe their abuse script didn't like one specific
attack (they just switch scanners so it won't help if it is to
avoid vuln scans). IP blocks don't help with DDOS and if it is
exploit attempts, just fix it because they won't go away
because you blocked an IP, they just move to a new IP usually.
anyway, I made my point
674[14:26:30] <terabit> Unit193: I don't mind solving
captchas or using it behind Cloudflare,etc...
675[14:26:53] <terabit> Ede|Popede: looks like httpd level (403)
676[14:27:16] <ratrace> depends on type of abuse. on our company
servers, we employ WAFs that detect SQL injections and then bans the
ip for 24 hours. We log those and if a network is particularly
abusive in that regard, the whole ASN gets permanently firewalled
677[14:27:38] <ratrace> I have hundreds of ASNs in permaban
ipsets :)
678[14:27:43] <Unit193> Excessive auth attempts tend to trigger
blocks too.
679[14:29:14] <cws> ratrace: wow, that's nasty :P
680[14:29:15] <Ede|Popede> oh, not even an edit link w/o being
logged in.
682[14:30:23] <machine_1490> Update re printer: I just
hadn't selected the proper driver in the printer app after
using the Brother installer.
683[14:30:33] <terabit> ratrace: I've seen companies that
even block entire countries,not surprising tbh. I've also seen
companies where IP blocks are a huge deal because of the type of
business they have. If it avoids actual operational impact and it is
the best solution it makes sense. but if someone banned my VPN IP
and it was a site I was doing business with I'd just move on.
720[15:39:06] <TheBigK02> echoSMILE: as long as u dont have any
apt dependency conflicts... i would assume everything works in
parallel ... but no guarantees on that from me :)
721[15:40:27] <TheBigK02> echoSMILE: or did u compile those
frameworks manually from source or something like that?
724[15:41:32] <echoSMILE> TheBigK02: nop from source. I am
trying to compile a xfce plugin from source, but he is asking for
gtk+3
725[15:42:20] <TheBigK02> echoSMILE: i dont understand... u did
also compile xfce from source?
726[15:42:41] <TheBigK02> echoSMILE: as long as u use the debian
native libraries for the dependencies, u should be fine
727[15:43:20] <echoSMILE> TheBigK02: nop. only one plugin
728[15:43:57] <TheBigK02> echoSMILE: as long as u can install
those dependency libraries without any conflicts in apt, u should be
on the working side
729[15:44:03] *** evilbug_ is now known as evilbug
730[15:44:21] <echoSMILE> let me try.
731[15:44:23] <echoSMILE> tks
732[15:45:00] <TheBigK02> but thats no garantee since ur working
outside of the debian package system... always having a working
backup is a good start :)
733[15:45:16] <TheBigK02> god i cant type today... sorry for the
mistakes ;)
765[16:27:47] <maxf> Hey, when I plug an android device, I can
browse its file in a gui file manager (I'm using gnome/debian
sid). But I can't figure out what's its actual path is (so
I can do stuff from CLI). My file manager tell me the device is at
`mtp://redacted_reference_here/`. How can I find where the device
really is ?
766[16:29:20] <crimsonhawk009> maxf: did you poke around in
/run/user/ ?
767[16:29:32] <maxf> crimsonhawk009, I did nothing there :/
768[16:30:01] <crimsonhawk009> maxf: take a peak in there once
you have it connected
780[16:40:48] <maxf> oxek, I don't want to have a shell on
the device, but to be able to do stuff with the files mounted on my
local computer (like move them somewhere on my computer, from the
terminal).
781[16:41:04] <oxek> maxf: do you have gvfs-backends package
installed?
782[16:41:45] <oxek> and gvfs-fuse
783[16:42:22] <maxf> gvfs-backends yes, gvfs-fuse no
809[17:01:43] <oxek> tbf I'm not a fan of debian having
separate onion addresses for each service
810[17:02:25] <oxek> doesn't really make much sense, since
one could simply be a proxy for all the services, and not even use
the full 3 hops for onion services
811[17:04:01] <SymbioticFemale> i don't think onions
support subdomains
812[17:04:15] <oxek> SymbioticFemale: they do
813[17:04:36] <SymbioticFemale> reducing the number of hops on
the other hand is at the discretion of whomever controls the torrc
814[17:04:44] <SymbioticFemale> oxek: that's cool.
i've never seen someone use one.
815[17:04:58] <maxf> oxek, cool, now the device is available
under /run/user/1000/gvfs/ !
819[17:05:48] <oxek> it's always either the lack of
gvfs-fuse or gvfs-backends, since they are not installed by default
on many systems
820[17:05:57] <topoi> Heyho, I try to stream from a webcam to a
server (can be local, so the stream gets tunneled through VPN). I
already tried motion, but the framerate is quite unsatifying
although I enjoy the simplicity of streaming without any fuzz (maybe
due to creating JPEG's?). Regarding VLC I cannot find a decent
working example that fits the need. Has anybody some hints?
853[17:45:24] <karlpinc> I missed how to mount mtp (adb)
connected devices. If there's a quick summary I'd like to
hear it.
854[17:47:27] <greycat> the few times I've seen it come up,
I believe the answer has ended up being that you *can't* mount
them as dumb disks, because the android OS gets in the way
856[17:47:54] <greycat> so you have to use special mtp-ish
programs with special GUIs and shit
857[17:48:27] <karlpinc> greycat: Up above (before I got
disconnected) there was talk of a fuse something or another. So I
thought maybe there was hope.
861[17:49:56] <karlpinc> I bricked a 2009-ish macbook pro. (And
I don't see a model number on the case or I'd google.)
I'm wondering if Debian can be installed on it. (Being a mac
there's no "normal bios", etc. I know there's
keys that can be pressed during the boot sequence but don't
know more.)
862[17:50:30] <sney> debian wiki has some macbook pro pages, if
anything an older model is more likely to have published workarounds
864[17:50:45] <karlpinc> sney: Thanks. I'll take a look.
865[17:50:50] <sney> though you might end up with fun situations
like "haha, no wifi for you"
866[17:51:44] <karlpinc> sney: Yah. That would be less exciting,
and put it in the same catagory as my Asus Aspire 1 that has no
wifi.
867[17:51:45] <crimsonhawk009> Yeah on a macbook air (2011) I
got everything installed just fine just no wifi
868[17:52:29] <sney> apple likes to use broadcom chips for wifi,
and that junk is annoying to get working on regular pc hardware,
never mind something that was purpose-built to run OSX only
870[17:52:54] <sney> aiui it is sometimes possible depending on
how much time you want to spend fighting with the various b43
options.
871[17:52:55] <karlpinc> I think I could get the asus wifi to
work if I tried. At first, there was no FOSS driver for it and I had
no wifi. Then, messed around with the proprietry driver and got
something working. But I think the FOSS driver has evolved and would
work.
872[17:53:30] <sney> yeah, that's usually the progression.
873[17:53:31] <karlpinc> The propetary driver stopped being
supported and no longer works.
874[17:54:10] *** Quits: Ericounet (~Eric@replaced-ip) (Quit: Je m'en vais ...)
881[17:58:42] <karlpinc> Does not look like the macbook is going
to work, considering it won't even boot OSX. Oh well.
882[17:59:52] <sney> apple is able to repair those things
somehow, I bet there's a formerly-secret firmware recovery mode
that's documented on a forum somewhere
887[18:02:59] <sney> command-option-o-f was the firmware key
combo for the powerpc macs, I wonder if they kept using it. but if
it's not that it'll be something.
965[19:25:14] <PackageCompare> Hi All. I have installed openssl
from source on Debian 10, and now my apt install is broken because
of broken dependencies that rely on openssl. What are my options to
resolve this?
967[19:27:04] <oxek> PackageCompare: give an example, or
outputs, of what you mean by broken
968[19:27:44] <PackageCompare> So I originally had openssl
installed but it wasnt the version that I needed, so I uninstalled
it and installed from source
969[19:27:46] <PackageCompare> One moment
970[19:28:13] <PackageCompare> ca-certificates : Depends:
openssl (>= 1.1.1) but it is not going to be installed
971[19:28:15] <imMute> PackageCompare: you either need to
package your version so it satisfies the dependency, or create a
fake package with nothing in it (that also satisfies the
dependency).
977[19:29:33] <imMute> apt and dpkg don't know or care
about stuff installed by hand, so such things cannot satisfy
dependencies.
978[19:30:25] <sney> !equivs
979[19:30:25] <dpkg> equivs is a package that enables you to
create dummy packages that tell <apt> you really have
installed (through some other means) the package. apt install
equivs, and read /usr/share/doc/equivs/*, see also <usrlocal>.
A better plan is often to adapt the Debian packages to your needs,
ask me about <package recompile> <uupdate> <ssb>.
981[19:30:37] *** Quits: maggotbrain (~maggotbra@replaced-ip) (Remote host closed the connection)
982[19:30:51] <greycat> imMute: that's nonsense.
Dependencies are managed by dpkg based on what is actually
installed, and it doesn't KNOW or CARE that you installed
something by hand, or via apt, or via some other tool.
998[19:33:27] <GNU\colossus> security issues that .1k fixes for
upstream releases are fixed in Debian stable's 1.1.1d version
as well
999[19:34:12] <PackageCompare> I have an application using
msodbcsql17 and it looks like the version of openssl is causing
conflict when trying to connect to our sql server 12.
1000[19:34:20] <PackageCompare> sorry sql server 11
1002[19:35:16] <GNU\colossus> can you show/demo what specific
problem you are running into? (using a pastebin)
1003[19:36:17] <greycat> To me this is sounding like "I have
my Debian libs and my Debian packages, but I also have my weird
production stuff, and my locally compiled libssl to support that,
and I keep them strictly separated".
1004[19:36:48] <greycat> I would not try to mix the two.
1005[19:37:04] <PackageCompare> So I found the solution here:
replaced-url
1006[19:38:01] *** debhelper sets mode: +l 1065
1007[19:38:11] <PackageCompare> The exact error is:
SQLDriverConnect: 10054 [Microsoft][ODBC Driver 17 for SQL
Server]TCP Provider: Error code 0x2746
1009[19:39:05] <GNU\colossus> the SO problems reads to me like
the server would like to handshake TLS 1.0, and the client with its
default config won't let that happen, due to Debian's
relatively strict default policy
1010[19:39:23] <GNU\colossus> that CERTAINLY isn't a reason
to install openssl from source into the system, fubar'ing many
other things for good
1011[19:39:29] <PackageCompare> But updating to the latest
version, theres no issues?
1012[19:39:43] <GNU\colossus> that version probably brings a less
trict policy/config with it
1054[20:01:26] <GNU\colossus> the retry connecting to the server
with your existing method
1055[20:02:43] <GNU\colossus> (the *real* fix should be to ask
the SQL Server admin to install a properly SHA-2 signed X.509/TLS
certificate for the service)
1056[20:04:15] <PackageCompare> I think that requires an update.
I dont manage the windows servers
1077[20:14:04] <avernos> How can i create a AP/hotspot and be
connected via wifi to internet? i have set the conf files, but wpa
loads the conf for the same interface as the AP
1078[20:14:46] <GNU\colossus> PackageCompare, looks good to me.
1079[20:14:50] <avernos> main problem is to separate the conf
files of one wifi to another :S
1080[20:16:07] <PackageCompare> In regards to sql server, I dont
see any certs installed
1156[21:12:17] <dpkg> Stretch is the codename for the current
<oldstable> release, Debian 9, released 2017-06-17.
"Stretch" is the rubber octopus in Toy Story 3, see
replaced-url
1157[21:12:21] <johnfg> jhutchins: Yes. I'm asking there
too. But I wonder if it's a debian problem with their service
file, perhaps.
1159[21:12:54] <greycat> if you're seriously considering
trying to use bleeding-edge development crap on a 4+ year old Debian
version, you may have your work cut out for you
1160[21:13:24] <imMute> johnfg: I didn't read back, but you
need to "systemctl enable" a service if you want it
started at boot (and the .service needs a proper [Install] section)
1182[21:22:29] <johnfg> I just did 'sudo apt-get -t
buster-backports install syncthing' and it returned: syncthing
is already the newest version (1.0.0~ds1-1+b11).
1189[21:26:12] <oxek> a possiblity would be to contact the
maintainer of syncthing in debian, and ask them to update syncthing
and backport it, but it's too late now because bullseye is in
freeze
1190[21:26:40] <oxek> hence bullseye will be released with an old
version of syncthing (1.12.1) and you'll need to wait for
bullseye-backports
1202[21:29:21] <oxek> imMute: -iname before -type, so that it
doesn't stat every file
1203[21:29:27] <maroloccio> oxek: what if I did not want to walk
the whole filesystem and leverage some pre-existing index? updatedb
from locate would give me that. does tracker give me that or what
else?
1249[21:42:42] <jmcnaught> johnfg: have you looked at
"journalctl -u syncthing@johnfg.service" ?
1250[21:44:40] <maroloccio> I just discovered that this is
freaking instantaneous: tracker search --music
1251[21:44:44] <alexrelis[m]> Is it possible to clone a drive
with `dd` if the source drive is larger than the destination drive,
but the partition of the source drive is shrunk to fit the size of
the destination?
1266[21:51:39] <tanja84dk> Just curious because I havent faced
this before what could the reason be that a server is holding 53
back in a dist upgrade and what advise would be best to forward?
1267[21:51:57] <sney> !idfma
1268[21:51:57] <dpkg> Insufficient Data For Meaningful Answer
1269[21:52:17] <maroloccio> tanja84dk: tell us more please
1275[21:53:10] <johnfg> jmcnaught: That was for you.
1276[21:53:41] <tanja84dk> its a old debian 8 server that
I'm in the middle of trying to upgrade to debian 9 and then
afterwards to 10 ( to not jusp to much as once )
1277[21:54:07] <sney> tanja84dk: apt is a powerful tool that
manages a lot of overlapping package dependencies, so there are a
lot of different reasons that could happen, depending on the
packages, environment, 3rd party repos, and many other factors. it
is impossible to say for sure without knowing details.
1278[21:54:15] <oxek> !bat
1279[21:54:15] <dpkg> [Basic Apt* Troubleshooting]. To diagnose
your problem, we need ALL OF THE FOLLOWING information: 1. complete
output of your apt-get/apt/aptitude run (including the command used)
2. output from "apt-cache policy pkg1 pkg2..." for ALL
packages mentioned ANYWHERE in the problem, and 3. "apt-cache
policy". Use
replaced-url
1280[21:54:37] <oxek> also sources.list, and so on.
1281[21:54:43] <maroloccio> I want to schedule a clean-up action
on a server from which I deleted user MP3 files. There is no
`at'. What should I use instead? I could double fork and
reparent a shell command group starting with sleep but is there
something else at my disposal on Deb10? It's one off, so I
don't think systemd timer units should be the go-to..
1282[21:54:45] <oxek> maybe you used `apt-mark hold` too
1290[21:56:24] <maroloccio> oxek: I think the diff there is that
at would survive a reboot if it happened in-between and sleep n;
cmd1 would not, right?
1291[21:56:39] <oxek> it wouldn't survive reboot, true
1292[21:56:52] <tanja84dk> This is the output from apt upgrade
replaced-url
1293[21:56:52] <oxek> but neither might accurate timekeeping
1294[21:57:09] <imMute> maroloccio: systemd-run with
--on-calendar
1295[21:57:28] <maroloccio> imMute: aha! let me look into this..
(thanks!)
1317[22:04:35] <tanja84dk> and no I have not tried apt
full-upgrade because I were following the guide didnt mention that,
so i have only apt dist-upgraded and normal upgraded after I changed
to stretch repo. And thanks alot going to try
1318[22:04:49] <sney> dist-upgrade and full-upgrade are the same
thing
1319[22:05:13] <oxek> `apt-get dist-upgrade` is same as `apt
full-upgrade` (and `apt dist-upgrade` works too)
1320[22:05:44] <oxek> johnfg: you can try
1321[22:05:59] <oxek> might at least show the maintainer that
someone cares about the package that they maintain
1322[22:06:00] <tanja84dk> I did the apt-get dist-upgrade and
that was where it marked them as held back
1335[22:11:29] <greycat> I keep the google-chrome repo in place
when upgrading, and it doesn't cause problems, but that's
a special case. Other third-party repos tend to be more intrusive.
1339[22:11:30] <sney> yeah, a repo that you know for sure only
has 1 or 2 packages is rarely an issue. but it's still ideal to
disable anything not-debian during the upgrade, and then re-enable
it if needed once done.
1343[22:11:47] <johnfg> oxek: With that page showing for action
needed, the latest version, do you think that's sufficient?
1344[22:12:05] <tanja84dk> :/ hmm great that really destroyed
alot :( it messed up the keyring and posible more. Guess there only
is one option now delete the server
1350[22:15:12] <alexrelis[m]> I am trying to take a Debian
install from one laptop and move it to another laptop by taking out
the NVME drive from the old laptop and putting it in the new laptop.
I also happened to have /boot mounted on a USB flash drive. When I
did this drive swap, the new system won't boot Grub from USB.
The new system doesn't even detect anything. Both systems have
UEFI. What could be the problem?
1351[22:15:15] <oxek> johnfg: I can't speak for the
maintainer.
1352[22:15:21] <tanja84dk> bu thanks alot for the help
1353[22:15:27] <tanja84dk> *but
1354[22:15:33] <oxek> sometimes packages only get upgraded when
maintainers are prompted by users
1356[22:16:26] <oxek> tanja84dk: that doesn't look like a
big problem
1357[22:16:56] <oxek> the security repo no longer updates the
signature for stretch
1358[22:17:45] <tanja84dk> oxek, its just when it has messed that
up and the system held so many packages back then it could be all
the packages there are having issues and flawn in their config
1359[22:17:58] *** Quits: supercilious (~thomas@replaced-ip) (Quit: ahhhh am ded)
1360[22:18:14] <oxek> not really
1361[22:18:26] <tanja84dk> that was just a copy from one og them
every repo reports that, so its the keyring it has fucked up
1362[22:18:47] <oxek> the proper order of things is: 1) Read
release notes 2) Change sources.list 3) `apt update` 4) `apt
upgrade` 5) `apt dist-upgrade`
1363[22:19:28] <oxek> tanja84dk: also, try to use the proper
archive repo, not the one from linode
1364[22:19:30] <johnfg> oxek: Email sent. Thanks for the help
today!
1365[22:19:41] <tanja84dk> But I am downloading the latest backup
from the off site place right now from the server to roll back on it
even thats eol
1369[22:20:44] <tanja84dk> well their repo has nevewr failed and
they set it up automaticly its just the same as if I chose thwe one
in sweeden from the installer
1370[22:20:58] <oxek> deb.debian.org never failed me
1371[22:21:42] <maroloccio> Maybe they save bandwidth with their
own repo? Outgoing connections
1373[22:22:37] <oxek> tanja84dk: anyway, after reboot, you should
be good to go to upgrade to buster
1374[22:22:47] <tanja84dk> they do plus local connection for the
costumers so better speeds and not having to pay for the bandwich or
counting to our limit
1383[22:31:23] <oxek> and upgrades are trivially easy
1384[22:31:40] <tanja84dk> well the upgrades does also break
everything so one bad thing or the other
1385[22:32:00] <oxek> tanja84dk: but it doesn't break it -
the "error" you received can be safely ignored
1386[22:32:08] <oxek> and you can continue to debian 10 from
there
1387[22:32:10] <tanja84dk> and all that reconfigurating and fix
things from upgrades would take months
1388[22:32:46] <oxek> well, in months there will be debian 11, so
you'll have even more things to work on, and even fewer people
that remember upgrades from 8->9
1389[22:33:38] <tanja84dk> it did break things checked in htop
and there were several things there didnt start after the dist
upgrade
1391[22:33:57] <sney> some vps images are known to not upgrade
cleanly, so in many cases it's wise to spin up a new image and
migrate your sites and services, rather than trying to upgrade in
place.
1393[22:34:08] <tanja84dk> actually no because next will be a
rolling realease linux on all servers
1394[22:34:15] <sney> this also mitigates any issues found by
upgrading mysql to mariadb, or php5 to php7, etc
1395[22:34:29] <sney> rolling release on a server? haha well if
you like excitement
1396[22:35:18] <tanja84dk> well at least there arent those big
issues. Like its the third time I have had issues upgrading a debian
server
1397[22:35:54] <tanja84dk> back when lenny welt eol it also
failed from debian 5 to 6 where it took the email server down
1398[22:36:04] <sney> you will just have to learn from experience
why this is a bad idea, I suppose.
1399[22:36:13] <tanja84dk> dovecot couldnt handle that upgrade
1400[22:37:29] <jmcnaught> Moving from Debian stable to a rolling
release would be like going from dealing with upgrade issues every
couple of years, to dealing with them all the time.
1410[22:41:38] <greycat> Debian is designed to do in-place
upgrades. Most Debian users never reinstall. Ever.
1411[22:42:02] <oxek> tanja84dk: you really don't
1412[22:42:12] <karlpinc> As in, doing things like installing 3rd
party packages and other software not in the debian repos, and
installing such things the wrong way.
1413[22:42:13] <tanja84dk> not even ubuntu server have had so
many issues upgrading, but unfortunally the overhead on ubuntu is
the reason I wont use them in the cloud
1415[22:42:37] <somiaj> tanja84dk: you are an outlier, most
(provided they follow the release notes) don't have to much
trouble with upgrades, and find them smooth.
1420[22:44:32] <wintersky> no one consciously wants arch
1421[22:44:47] <tanja84dk> karlpinc, nope never arch on a server
that is suicide way to bleading egde. My plan is testing with
opensuse in the summer holidays and if it goes well over the summer
rolling it out in the fall
1429[22:46:18] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
1430[22:46:20] <ens> hello. i am on an ancient system that i am
trying to get some package manager working. It's debian sarge.
When i try to apt-get update it gives 404s trying to find the
package list. Currently the source is: "replaced-url
1431[22:46:34] <greycat> !sarge sources.list
1432[22:46:34] <dpkg> methinks sarge sources.list is "deb
replaced-url
1433[22:46:49] <wintersky> karlpinc: have you ever heard of
debootstrap?
1436[22:47:03] <ens> thanks greycat i'll give that a shot.
1437[22:47:04] <wintersky> it allows you bootstrapping a very
minimal debian system
1438[22:47:14] <wintersky> i guess this is what you are looking
for
1439[22:47:31] <greycat> after a few years, all the releases get
moved to archive.debian.org
1440[22:48:36] <karlpinc> ens: Don't try a major release
upgrade without upgrading one at a time to successive releases. (And
at this point it may make sense to start with a clean install
anyway.)
1451[23:03:09] <ens> yeah, i just want to grab a compiler and
maybe a few other basic utils like a text editor. upgrading the
release on such an ancient box sounds like a daunting challenge
indeed.
1452[23:03:45] <wintersky> you might rather want doing a clean
install
1453[23:04:18] <wintersky> will save you time and effort
1454[23:04:20] <ens> unfortunately that isn't an option as
i'm not the one responsible for provisioning this system.
1456[23:04:42] <wintersky> ok then do release-after-release
upgrades
1457[23:04:59] <wintersky> you'll dump a day on it
1458[23:07:11] <ens> i'm considering it, since it's
only got about 35k/s internet link. so maybe doing release upgrades
first would save me the pain of having to also upgrade all the pkgs
i install beforehand.
1465[23:10:07] <ens> to my knowledge it doesn't have
anything like that. it's just a really cut-down debian system
that serves primarily as a developer tool i am given to access files
on a dev server locally and do some basic admin tasks.
1468[23:10:48] <tanja84dk> oxek, btw gave it a chance rebooting
because waiting on the backup to get ready dowwnloadfing ( misspoke
prior ). and I did see now even also interfaces are f**ked up some
how like after the reboot some there should be there arent and some
weird are there instead
1550[23:23:30] <sney> if gnome 4 is coming, it could possibly be
included in bookworm, ~3 years from now, depending on how ready it
is for production at that point.
1551[23:23:49] <horribleprogram> why is literally all the
documentation of X11 400 years old
1576[23:34:22] <somiaj> Thete: the longest freeze was about a bit
over a year, since then the release team has added some additional
policy to help deal with long standing rc-bugs and the last few
feezes have been about 6 months
1600[23:48:48] <somiaj> kinda, sometimes the fix to an rc-bug is
remove the package, some rc-bugs are 'ignored' if they are
deemed blocking the release, and there is no easy fix or there are
ways to midagate them
1601[23:48:57] * phogg is not a source of official information, treat
his comments as rumor
1602[23:49:17] *** Quits: b1anc (~user@replaced-ip) (Remote host closed the connection)
1603[23:49:19] <somiaj> in general, yes the rc-bug count reahces
zero at the release, but after the release many new ones are found
(some that only affect the new testing)
1604[23:49:27] <phogg> or sometimes they're reclassified so
as to pretend they're not a problem, if they are indeed not
really that critical after all
1605[23:49:44] <somiaj> more they are deemed not critical enough
to block the release
1606[23:50:02] <phogg> indeed, technically rc critical but not
"hold the whole release" critical
1607[23:50:19] *** Quits: submariner (~v32itas@replaced-ip) (Remote host closed the connection)
1615[23:52:11] <somiaj> so though in theory rc-bugs should be
zero at the time of the release, as the number of packages
increases, things get more complicated and interconnected, they are
sometimes dealt with in other ways, mostly to keep the freeze from
not taking over a year.
1616[23:52:27] *** Quits: |shark| (~shark_@replaced-ip) (Remote host closed the connection)
1617[23:52:28] *** Quits: Nitori (~kappa@replaced-ip) (Quit: "I hate eggs with no yolk. My friend Herensuge was
beaten to death by one.")
1624[23:52:58] *** Quits: noettore_afk (~noettore@replaced-ip) (Quit: No Ping reply in 180 seconds.)
1625[23:53:06] *** Quits: horribleprogram (~horriblep@replaced-ip) (Quit: Where I came from the Great Wild 'n shit, where you
can get shot if you crack smiles and shit...)