IRC chat logs for #ltsp on irc.freenode.net (webchat)


Channel log from 30 July 2019   (all times are UTC)

00:57adrianor1 is now known as adrianorg
01:07
<adrianorg>
alkisg: I seem to understand that I will need to create a raw disk image and install an operating system on it using some virtualization software (such as virtualbox), symlink that disk image to $BASE_DIR/ltsp/img_name.img, then run ltsp image img_name, which will export an squashfs image to $BASE_DIR/images/img_name.img. Is that correct?
01:46vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
02:18GodFather has left IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com, Ping timeout: 258 seconds)
02:18GodFather_ has left IRC (GodFather_!~rcc@d53-64-7-141.nap.wideopenwest.com, Ping timeout: 245 seconds)
04:44woernie has joined IRC (woernie!~werner@p57A0EC92.dip0.t-ipconnect.de)
04:51adrianor1 has joined IRC (adrianor1!~adrianorg@177.132.219.253)
04:55adrianorg has left IRC (adrianorg!~adrianorg@186.213.156.2, Ping timeout: 268 seconds)
05:00woernie has left IRC (woernie!~werner@p57A0EC92.dip0.t-ipconnect.de, Remote host closed the connection)
05:13
<alkisg>
adrianor1: exactly, except the symlink goes in $BASE_DIR/img_name.img (without /ltsp/). Thanks a lot!
05:36ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)
05:51statler has joined IRC (statler!~Georg@p54897D17.dip0.t-ipconnect.de)
05:59adrianorg has joined IRC (adrianorg!~adrianorg@179.179.75.220)
06:02adrianor1 has left IRC (adrianor1!~adrianorg@177.132.219.253, Ping timeout: 248 seconds)
06:30woernie has joined IRC (woernie!~werner@p50867386.dip0.t-ipconnect.de)
08:04book` has left IRC (book`!~book`@static.38.6.217.95.clients.your-server.de, Quit: Leaving)
08:08book` has joined IRC (book`!~book`@static.38.6.217.95.clients.your-server.de)
08:45woernie has left IRC (woernie!~werner@p50867386.dip0.t-ipconnect.de, Remote host closed the connection)
08:50statler has left IRC (statler!~Georg@p54897D17.dip0.t-ipconnect.de, Remote host closed the connection)
09:02adrianor1 has joined IRC (adrianor1!~adrianorg@177.156.228.158)
09:05adrianorg has left IRC (adrianorg!~adrianorg@179.179.75.220, Ping timeout: 244 seconds)
09:41statler has joined IRC (statler!~Georg@gwrz.lohn24.de)
11:25adrianor1 is now known as adrianorg
11:36GodFather_ has joined IRC (GodFather_!~rcc@d53-64-7-141.nap.wideopenwest.com)
11:36GodFather has joined IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com)
11:39
<adrianorg>
alkisg: I thank you! :-) I went to read the man page again to check the "without /ltsp/ part" and now $BASE_DIR is expanded in the document (e.g. "/srv/ltsp defaults to /srv/ltsp"). The git history says it's a "Temporary commit", so maybe I should wait for the next commit.
11:56section1 has joined IRC (section1!~section1@178.33.109.106)
12:03Faith has joined IRC (Faith!~Paty_@2001:12d0:2080::231:49)
12:03Faith has joined IRC (Faith!~Paty_@unaffiliated/faith)
13:28professor90 has joined IRC (professor90!bb4c0216@187.76.2.22)
13:28
<professor90>
hi
13:28
any can help me with epoptes?
13:35professor90 has left IRC (professor90!bb4c0216@187.76.2.22, Remote host closed the connection)
13:53vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
14:01
<alkisg>
adrianorg: I thought it would be more readable if I replaced $BASE_DIR with /srv/ltsp in the documentation; but my search/replace made that line that you just mentioned not make sense anymore :)
14:01
Thanks, I just replaced it with: The /srv/ltsp path can be configured using `ltsp --base-dir=`:
14:19
vagrantc: do you happen to know foka@debian, the go-md2man maintainer? We would really appreciate a new go-md2man release, both for ltsp5 and ltsp19+, as it's now using blackfriday v2 and it has a lot of fixes, including the OPTIONS...
14:20
<vagrantc>
alkisg: not off the top of my head
14:20
<alkisg>
ok
14:20
<vagrantc>
alkisg: a wishlist bug to update it isn't uncommon
14:21
<alkisg>
Ah, cool, will do
14:21
<vagrantc>
especially with an offer to help, but be careful or you end up the maintainer :)
14:21
<alkisg>
Nah, I'm at -50% free time now :D
14:21
<vagrantc>
heh
14:22
<alkisg>
I'd love to try uploading ltsp19 myself though, with sponsoring or whatever it takes
14:22
<vagrantc>
but just describing the new features you want/need in the newer versions briefly and nicely sometimes works :)
14:22
<alkisg>
Later though, finishing the code comes first; packaging a bit later
14:22
Oh just the new go-md2man release, nothing other than that
14:23
<vagrantc>
sure, happy to help with that when you think it's ready
14:23
<alkisg>
Cool. Will use the PPA for a brief testing period in around 2 weeks, then another week to get feedback, then I'll ping you for whenever you have time
14:24
I've set up an ltsp-community repository in github, to replace the mailing list and the wiki.ltsp.org for ltsp19+
14:25
While ltsp.github.io will contain "upstream only" documentation, including the man pages
14:42shored has left IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi, Read error: Connection reset by peer)
14:42shored has joined IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi)
14:51
<alkisg>
vagrantc: btw, the existing ronn gives muuuuch better output than even the newer go-md2man; why are its build dependencies so significant? To make local pbuilder builds faster?
14:52
<vagrantc>
that was one reason
14:52
<alkisg>
I was trying to convert from the go-md2man output to html, to simulate e.g. debian manpages like https://manpages.debian.org/unstable/ltsp-server/ltsp-update-image.8.en.html, but it's kinda awful
14:52
<vagrantc>
it also increases the liklihood that it gets blocked from transitioning the larger the dependency chain is
14:52
<alkisg>
I.e. the go-md2man produced man pages are of low quality, and aren't converted to html well
14:52
<vagrantc>
i think at the time ronn was even maybe not going to be in the release
14:53
<alkisg>
There's also pandoc, but I think that's even bigger
14:53
<vagrantc>
i was happy enough with the results, but you keep bringing it up so i guess you're not :P
14:53
<alkisg>
I guess I can run ronn locally and we'll have them nice in our ltsp.github.io web site, but they won't display nice in all other sites with go-md2man...
14:54
vagrantc: have a look: https://manpages.debian.org/unstable/ltsp-server/lts.conf.5.en.html
14:54
Although you didn't use options or anything, the spacing is still too big
14:54
Go to this line: KEEP_SYSTEM_SERVICES="acpid avahi-daemon bluetooth cups"
14:55
8 empty lines above, 3 below...
14:56
<vagrantc>
looks a bit odd, sure
14:56
<alkisg>
If I were a user trying to read the lts.conf man page from firefox, I'd be disappointed
14:56
<vagrantc>
my old eyes appreciate a bit of whitespace, though
14:56
<alkisg>
It's a lot more whitespace than `man lts.conf` gives
14:56
It's just a lot of bugs involved there
14:56* vagrantc shrugs
14:56* alkisg reminds vagrantc that his eyes are older :P
14:57
<vagrantc>
just must be wearing mine out faster
14:57
<alkisg>
Hehe
14:59
On the other hand, the older lts.conf, is a lot more readable online: manpages.ubuntu.com/lts.conf
15:00
Ah it's on debian.org too: https://manpages.debian.org/unstable/ltsp-docs/lts.conf.5.en.html
15:01* vagrantc really didn't want to continue to maintain all that xml/docbook stuff
15:01
<alkisg>
Oh sure, that part's fine
15:01
Let me see if ronn does the trick
15:02
Otherwise we may ...sed the go-md2man output or something
15:03
<mwalters>
alkisg: if you want to start putting issues up in the ltsp/ltsp.github.io repo, it'd help people contribute
15:04
(i.e., "todo" issues)
15:04
<alkisg>
mwalters: issues? Like, bug reports or missing features?
15:04
Ah, not yet
15:04
I need to go into beta first
15:04
I would appreciate someone willing to test though, in a couple of weeks
15:04
<mwalters>
absolutely, I'll get a test box together
15:05
are you targeting 19.04 for ubuntu?
15:05
<alkisg>
E.g. I might be thinking "since XRANDR_COMMAND_* doesn't make sense with wayland, I'll just provide a method to create a xorg.conf", but users might convince me otherwise
15:06
No no, I think I'll leave ltsp5 in 20.04, while in parallel, the ppa and greek schools will have ltsp19 for 20.04, so ltsp users can choose
15:06
And bullseye can ship with ltsp19
15:06
<mwalters>
gotcha
15:07
<alkisg>
(and 20.10 with ltsp19+ too)
15:07
<mwalters>
They replaced the carpet in our day school. Now the director can't print... :D
15:07
I guess I need to go look at that
15:07
<alkisg>
Haha, I guess they needed to unplug the printer :D
15:16
<vagrantc>
carpet based networking is all the rage
15:26vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Ping timeout: 276 seconds)
15:28
<alkisg>
vagrantc, when you come back, compare with ronn output: https://ltsp.github.io/ltsp-image.html
15:28
I don't know what you say, I prefer that a whole lot better :)
16:37vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
17:15
<alkisg>
(06:28:26 PM) alkisg: vagrantc, when you come back, compare with ronn output: https://ltsp.github.io/ltsp-image.html (06:28:41 PM) alkisg: I don't know what you say, I prefer that a whole lot better :)
17:15
I can have a script that syncs them from our wiki, ronn's them, and pushes them to our sources, so that we ship compiled man pages instead of requiring ronn at build time :P
17:18
<vagrantc>
would violate the GPL, may as well just depend on ronn if you're set on it :)
17:20
<alkisg>
OK; but... /me blinks.... violate the gpl how<
17:20
Isn't troff a valid source?
17:21
<vagrantc>
not if you're preferred way of modifying it is not troff
17:21
e.g. would you accept patches to the generated .troff, or would you then require then to edit the .md files?
17:22
<alkisg>
Hrm, that would make all images in debian violate gpl, because each person maintains them as e.g. gimp source and publishes them as png...
17:22
I see the logic, but I'm not sure... will think about it another time :)
17:22
ty
17:23
<vagrantc>
how do you take patches for images?
17:23
it sounds like you prefer ronn at this point, so i'd say go with it
17:23
<alkisg>
vagrantc: I can also do this:
17:23
we use go-md2man anyway, and manpages.debian.org/our manpages/ suck,
17:24
but, I locally run ronn, and ltsp.github.io/manpages are fine
17:24
If you prefer that, I don't mind doing it
17:24
<vagrantc>
i'd just say go with ronn; it's simpler
17:25
<alkisg>
OK
17:25
<vagrantc>
the large dependency chain slows down build times slightly and might occasionally slow transition from unstable to testing in debian, but not so much as to maintain two separate systems
17:26
when possible, better to use the same thing everywhere
17:30statler has left IRC (statler!~Georg@gwrz.lohn24.de, Remote host closed the connection)
17:49spaced0ut has left IRC (spaced0ut!~spaced0ut@unaffiliated/spaced0ut, Quit: Leaving)
18:12shored has left IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi, Read error: Connection reset by peer)
18:12shored has joined IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi)
19:12shored has left IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi, Read error: Connection reset by peer)
19:13shored has joined IRC (shored!~shored@87-92-94-52.bb.dnainternet.fi)
19:18statler has joined IRC (statler!~Georg@p54897D17.dip0.t-ipconnect.de)
19:40woernie has joined IRC (woernie!~werner@p57A0EC92.dip0.t-ipconnect.de)
19:54
<Sleaker>
what's with ltsp clients always complaining about not being able to connect to nbd9?
19:54
looks like something in ldm rc.d init script maybe?
19:56
the boot parameter clearly is /dev/nbd0 for the ltsp boot image stuff, but then it seems like ltsp is trying a bunch of file system mounts on nbd1-12?
19:57section1 has left IRC (section1!~section1@178.33.109.106, Remote host closed the connection)
20:06
<vagrantc>
Sleaker: it's a check if the image has been updated since it was booted
20:10woernie has left IRC (woernie!~werner@p57A0EC92.dip0.t-ipconnect.de, Remote host closed the connection)
20:25
<alkisg>
!nbd
20:25
<ltsp>
I do not know about 'nbd', but I do know about these similar topics: 'disable-nbd-proxy', 'nbd-proxy', 'nbd-compression', 'quick-nbd-tests', 'nbd-client', 'nbd-server-bug', 'nbd-trusty-bug', 'nbd-client-breaks-nm', 'nbd-bug', 'nbd-checkupdate'
20:26
<alkisg>
!nbd-checkupdate | echo Sleaker:
20:26
<ltsp>
Sleaker: nbd-checkupdate: To avoid bug https://github.com/NetworkBlockDevice/nbd/issues/59, put this in lts.conf: INIT_COMMAND_RM_NBD_CHECKUPDATE="rm -rf /usr/share/ldm/rc.d/I01-nbd-checkupdate"
20:26* alkisg wonders why we haven't completely removed that check yet :P
20:35
<Sleaker>
oooo thanks
20:44Faith has left IRC (Faith!~Paty_@unaffiliated/faith, Quit: Leaving)
20:45
<Sleaker>
has anyone tried to build buster images?
20:46
<vagrantc>
not since the last upload to buster
20:46
<Sleaker>
I think it's asking for input on the apt-get install command :-/
20:47
I think I forgot to set NONINTERACTIVE
20:47
ahh yah need to set the deb frontend to non-interactive to bypass the samba questions.
20:52
<vagrantc>
oh.
20:52
no.
20:52
there is an AWFUL bug in apt.
20:52
<Sleaker>
what's that?
20:52
<vagrantc>
oh, it's probably not the apt issue...
20:53
what are you using to create the image?
20:53
<Sleaker>
no I'm pretty sure this was me running apt-get -y install $PACKAGE_LIST
20:53
which includes smbclient
20:53
and not having DEBIAN_FRONTEND=noninteractive set
20:53
<vagrantc>
ah, yeah ... if you manually install packages, you'll have to set that
20:53
<Sleaker>
yuo
20:53
yup*
20:53
ltsp-build-client --no-squashfs-image --dist buster --arch amd64 --base $BASE_DIR --purge-chroot | tee $BUILD_LOG
20:53
and then: ltsp-chroot -p -d -a amd64 -b $BASE_DIR /tmp/client-script.sh | tee -a $BUILD_LOG
20:54
to setup the 'extras'
20:54
<vagrantc>
you could also add the packages with one of the --*packages* options
20:54
<Sleaker>
it includes custom packages that are being copied into the chroot
20:55
<vagrantc>
no repository for it to download from? ... hrm.
20:55
<Sleaker>
although those are just being dpkg -i /tmp/*.deb
20:55
locally managed repo
20:55
<vagrantc>
then should point them to the repository
20:55
<Sleaker>
it's easier to build the squash image from mainline
20:56
and then copy the packages locally
20:56
<vagrantc>
and you can also use "apt install ./path/to/package.deb" instead of "dpkg -i /path/to/package.deb" and it will actually attempt to resolve dependencies
20:56
<Sleaker>
at this point in our build the apt repository isn't live with the changes from the current build
20:56
<vagrantc>
and mark it correctly as istalled or not
21:12vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
21:36ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)
21:40GodFather_ has left IRC (GodFather_!~rcc@d53-64-7-141.nap.wideopenwest.com, Ping timeout: 248 seconds)
21:40GodFather has left IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com, Ping timeout: 244 seconds)