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


Channel log from 16 November 2013   (all times are UTC)

00:18nocturn has left IRC (nocturn!~nocturn@unaffiliated/nocturn, Ping timeout: 260 seconds)
00:25nocturn has joined IRC (nocturn!~nocturn@unaffiliated/nocturn)
00:37simpoir has left IRC (simpoir!~yaaic@sencha.simpoir.com, Ping timeout: 248 seconds)
00:46Phantomas1 has joined IRC (Phantomas1!~Phantomas@ubuntu/member/phantomas)
00:48Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 264 seconds)
00:49Phantomas1 has left IRC (Phantomas1!~Phantomas@ubuntu/member/phantomas, Client Quit)
01:17dsugar100 has left IRC (dsugar100!~dsugar@columbia.tresys.com, Quit: dsugar100)
01:18dsugar100 has joined IRC (dsugar100!~dsugar@columbia.tresys.com)
01:30Enslaver_ has left IRC (Enslaver_!~Enslaver@c-98-196-42-169.hsd1.tx.comcast.net, Read error: Connection reset by peer)
01:32Enslaver has joined IRC (Enslaver!~Enslaver@fedora/Enslaver)
01:37Enslaver has left IRC (Enslaver!~Enslaver@fedora/Enslaver, Read error: Connection reset by peer)
01:40Enslaver has joined IRC (Enslaver!~Enslaver@fedora/Enslaver)
02:29alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
02:47alexqwesa has left IRC (alexqwesa!~alex@109.172.12.47, Ping timeout: 264 seconds)
03:45alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
04:10Enslaver has left IRC (Enslaver!~Enslaver@fedora/Enslaver, Read error: Connection reset by peer)
04:13Enslaver has joined IRC (Enslaver!~Enslaver@fedora/Enslaver)
05:00alexqwesa has joined IRC (alexqwesa!~alex@109.172.12.47)
05:39vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)
05:47Hyperbyte has left IRC (Hyperbyte!jan@middelkoop.cc, Ping timeout: 245 seconds)
05:48Hyperbyte has joined IRC (Hyperbyte!jan@middelkoop.cc)
06:02Hyperbyte has left IRC (Hyperbyte!jan@middelkoop.cc, Ping timeout: 264 seconds)
06:02Hyperbyte has joined IRC (Hyperbyte!jan@middelkoop.cc)
07:08urly has left IRC (urly!4dccfd66@gateway/web/freenode/ip.77.204.253.102, Quit: Page closed)
09:04bennabiy has left IRC (bennabiy!~Thunderbi@24-181-55-79.dhcp.gnvl.sc.charter.com, Ping timeout: 248 seconds)
09:04vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)
09:05bennabiy has joined IRC (bennabiy!~Thunderbi@24-181-55-79.dhcp.gnvl.sc.charter.com)
09:10freedomrun has left IRC (freedomrun!~freedomru@unaffiliated/freedomrun, Read error: Operation timed out)
09:28freedomrun has joined IRC (freedomrun!~freedomru@unaffiliated/freedomrun)
10:37Enslaver has left IRC (Enslaver!~Enslaver@fedora/Enslaver, Read error: Connection reset by peer)
10:40Enslaver has joined IRC (Enslaver!~Enslaver@fedora/Enslaver)
11:01elias_a has left IRC (elias_a!elias@hilla.kapsi.fi, Read error: Connection reset by peer)
11:31klausade has joined IRC (klausade!~klaus@cm-84.215.153.179.getinternet.no)
11:32Fenuks has joined IRC (Fenuks!~Fenuks@212.164.150.126)
12:35freedomrun has left IRC (freedomrun!~freedomru@unaffiliated/freedomrun, Ping timeout: 268 seconds)
12:40siahos_ has joined IRC (siahos_!2eb049bf@gateway/web/freenode/ip.46.176.73.191)
12:40
<siahos_>
#join ts.sch.gr
12:40
join #ts.sch.gr
12:42
help
12:43
join #ts.sch.gr
12:45siahos_ has left IRC (siahos_!2eb049bf@gateway/web/freenode/ip.46.176.73.191, Quit: Page closed)
12:49freedomrun has joined IRC (freedomrun!~freedomru@unaffiliated/freedomrun)
13:10alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
14:15imox has joined IRC (imox!~imox@91-64-148-46-dynip.superkabel.de)
14:46markit has joined IRC (markit!~marco@88-149-177-66.v4.ngi.it)
14:50alexqwesa has left IRC (alexqwesa!~alex@109.172.12.47, Remote host closed the connection)
14:50alexqwesa has joined IRC (alexqwesa!~alex@109.172.12.47)
15:28hs366 has joined IRC (hs366!~hs366@94.254.45.76)
15:41PhoenixSTF has joined IRC (PhoenixSTF!~rudi@78.29.159.239)
16:23vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)
16:42
<alkisg>
vagrantc: should I commit https://bugs.launchpad.net/ltsp/+bug/357268 or do you want me to wait until you release?
16:43
I think we should commit it so that it gets wider testing...
16:50
<vagrantc>
alkisg: groupadd is debian-specific ...
16:51
or rather, it's API is debian-specific
16:51
<alkisg>
vagrantc: I thought addgroup was debian specific...
16:51
<vagrantc>
groupadd/addgroup often exist on both debian and redhat based systems, but their behavior and commandline options are often flip-flopped
16:52
they're not interchangeable between distros
16:53
this seems like a case for vendor_functions or something, though.
16:53
<alkisg>
vagrantc: we could easily do it with some standard library, I'm guessing those are the same between distros?
16:53
<vagrantc>
i.e. move the current *slow* code into generic functions, and replace them with vendor-specific functions
16:53
<alkisg>
I don't think there's need for the current code at all
16:54
<vagrantc>
alkisg: does your new code not use /var/cache/ltsp-localapps at all?
16:54
<alkisg>
The problem was in groups with spaces, but with id -G that was solved
16:54
Since it's always ran, I thought we shouldn't use that localapps directory
16:54
The local user will always be created now, even if localapps are off
16:55
<vagrantc>
ok.
16:55
<alkisg>
And it's just for 2 files now instead of 5, so I don't think they're worth their own dir...
16:55
<vagrantc>
alkisg: does it wipe out the user somehow?
16:56
<alkisg>
It's the previous code, on logout, on X crash, and on the next login, the previous passwd and group are restored
16:56
<vagrantc>
we do it in both places, and you remooved the "extra" one?
16:57
<alkisg>
I didn't remove anything , it's still in 3 places
16:58
<vagrantc>
alkisg: i see it removed from localapps-cleanup
16:58
doh.
16:58
and implemented always, rather than aas a conditional
16:59
<alkisg>
I just moved it outside the if, because we agreed we want to always create the local user, right?
16:59
<vagrantc>
alkisg: i see some changes in the LOCAL_APPS_EXTRAMOUNTS code ...
16:59
alkisg: yeah, that's fine, i just missed it
17:00
alkisg: switched from "${LOCAL_APPS_EXTRAMOUNTS}" to "$LOCAL_APPS_EXTRAMOUNTS" and similar changes with other variables
17:00
is that needed?
17:00
<alkisg>
Yes, some devs are mixing the ${} semantics with the "$" semantics,
17:00
<vagrantc>
same with CUPS_SERVER
17:00
<alkisg>
e.g. I think that they think that this works even with spaces:
17:00
sshfs -o ${follow_home_symlinks}allow_other,nonempty,ControlPath=${LDM_SOCKET} ${LDM_SERVER}:${LDM_HOME} ${LDM_HOME}
17:00
${LDM_HOME} doesn't work if it contains spaces
17:01
"$LDM_HOME" and "${LDM_HOME}" do
17:01
So I think that by replacing ${} with "$", other devs will follow up and the code will work with spaces and other extra characters too
17:01
Of course we could use "${}"...
17:02
Anyways, whatever, I just want the code to work PLUS some consistency
17:02
So, I want quoting, and I don't care about ${} or not as long as the code is consistent
17:03
<vagrantc>
hmmm...
17:04
alkisg: ok, so it's just some coding consistancy changes mixed in with the new feature?
17:04
alkisg: er, bugfix :)
17:04
<alkisg>
Not just consistancy, they're there for cases where $LDM_HOME has spaces too
17:04
Or other special characters...
17:05
I do that regularly when I change some script...
17:05
Maybe it'd be better to do it for all scripts at once, at a separate commit, but it's a bit big task...
17:06markit has left IRC (markit!~marco@88-149-177-66.v4.ngi.it, )
17:07
<vagrantc>
alkisg: so it's two bugfixes plus coding consistancy changes
17:07
alkisg: it's just a big enough diff as is, wanted to make sure i wasn't missing something.
17:07
<alkisg>
Yup, you didn't miss anything
17:07
The main goal was to "always create the local user"
17:08
I didn't like the code, it was slow and there was the bug report etc so I thought I'd rewrite some bits
17:08
By removing /var/cache/ltsp-localapps, I had to change many files just for that line,
17:09
and, there were a few cases of quoting + removing ${}
17:09
But since groupadd is debian-specific, I have to revisit it now... :)
17:10
<vagrantc>
alkisg: at least, i think it is. i remember having troubles using the various *group* and *user* commands across distros
17:11
<alkisg>
vagrantc: I want to use them to notify accountsservice etc about the user change, but we'll have to `sed` afterwards anyway to support spaces...
17:11
...although /etc/passwd and group weren't meant to have spaces in the usernames there, the tools seem to work fine with them
17:11
<vagrantc>
 groupadd --force --gid "$gid" --non-unique "$groupadd_name"
17:11
that could cause two groups to have the same gid ?
17:11
<alkisg>
Yes, and I think that's what we want
17:12
<vagrantc>
can cause very unpredictible behavior if those groups ever clash
17:13
alkisg: does that also renumber an existing group, or add another group with the same name and different gid ?
17:14
<alkisg>
Neither of those two, I think
17:15
All those cases need documentation, on what we choose to do and why
17:16
I think that we don't want to renumber existing groups, nor add another group with same name / different gid, and that we do want the same gid in the first ^ case...
17:16
...but some use cases should be mentioned to justify those...
17:17* alkisg boots a fedora.iso...
17:17* vagrantc waits patiently
17:18
<vagrantc>
fuse is one group that's frequently different GIDs on the thin client and the server, for example.
17:18
at least, that's one where i've actuaally run into problems
17:19
<alkisg>
We want to reuse the local gid there, to have access to the devices, right?
17:19
vagrantc: I see the same manpage for groupadd as for debian
17:19
<vagrantc>
right.
17:19
alkisg: really?
17:19
maybe things have improved.
17:20
<alkisg>
OK, let's test on that fedora VM
17:21
First case, group name that already exists?
17:21
<vagrantc>
it only matters if the gid is different, in that case
17:23
<alkisg>
It does nothing and returns success no matter if the gid is the same or not
17:23
That's what we want, right? To keep the local gid...
17:24
<vagrantc>
seems like strange behavior... but ok, yeah, that's what we want.
17:24
<alkisg>
Second test, different name, existing gid
17:24
It added it and returned success
17:25* alkisg tries to think of a use case for that...
17:26
<vagrantc>
so two names with conflicting gids?
17:26
<alkisg>
Yup...
17:26
<vagrantc>
or, two names with same gid
17:27
<alkisg>
Suppose I'm on a group g1:1234 locally, and connect to the server where I'm on a group g2:1234
17:27
<vagrantc>
alkisg: basically, if there are files with a group in the user's homedir or in extramounts, it may show up with the wrong group name
17:28
<alkisg>
I think it depends on the file system and the mapping it uses
17:28
<vagrantc>
and then you get into interesting games with "rsync -a" or "tar p" and such
17:28
alkisg: i have some confidence it will be unpredictible :)
17:29
alkisg: i think the current code creates a new GID for GID conflicts?
17:29
which can lead to cascading gid conflicts...
17:29
<alkisg>
That's easy to do, we can just ommit the --non-unique
17:29
We just need to find out what we prefer
17:29
Is NFS always doing name mapping?
17:30
Or some times it's using UIDs/GIDs?
17:30
<vagrantc>
not sure
17:30
<alkisg>
(and also ltspfs...)
17:35Enslaver has left IRC (Enslaver!~Enslaver@fedora/Enslaver, Read error: Connection reset by peer)
17:35
<bennabiy>
alkisg vagrantc: Isn't the reason that spaces work fine in /etc/passwd is because the IFS is mapped to : instead of a space or tab
17:35
so names can have spaces
17:36
<alkisg>
bennabiy: the shadow tools don't allow adding users with spaces in their names
17:36
<vagrantc>
"work fine" is an interesting way of putting it
17:36
so many tools fall over when that happens
17:36
<alkisg>
That's why we have to sed afterwards, to force a minimal compatibility with ldap/ad
17:36
<bennabiy>
I am sorry, I read wrong, I was not talking about username. I was talking about gecos
17:36
<alkisg>
(names)
17:36
Sure gecos can have spaces...
17:37
<bennabiy>
I guess it helps to read the whole thing before commenting
17:37
sorry about that
17:38Enslaver has joined IRC (Enslaver!~Enslaver@fedora/Enslaver)
17:38
<alkisg>
bennabiy: no worries :)
17:40
vagrantc: I want to download some .iso's to be able to cross-distro test such things, which distros would you consider OK for something to go in the "main" ltsp source and not in some distro-specific function?
17:41
debian, ubuntu, fedora, suse, gentoo, are those enough?
17:42
<vagrantc>
alkisg: you're a good soul. sure, those seem fine.
17:42* alkisg hopes there's a gentoo iso and that he doesn't need to build everything from source... :D
17:44
<vagrantc>
heh.
17:51khildin has joined IRC (khildin!~khildin@ip-213-49-83-192.dsl.scarlet.be)
17:55Fenuks has left IRC (Fenuks!~Fenuks@212.164.150.126, Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
18:22xet7 has left IRC (xet7!~xet7@a88-112-147-81.elisa-laajakaista.fi, Read error: No route to host)
19:38imox has left IRC (imox!~imox@91-64-148-46-dynip.superkabel.de, Quit: imox)
19:52
<alkisg>
vagrantc: gentoo has the same /usr/sbin/groupadd program and manpage too
19:53
opensuse will have to wait until tomorrow, those distros only seem to have 4+ Gb live dvds... :)
19:54
and `xprop -spy` seems generally available too
19:57alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
20:01khildin has left IRC (khildin!~khildin@ip-213-49-83-192.dsl.scarlet.be, Quit: I'm gone, bye bye)
20:18PhoenixSTF has left IRC (PhoenixSTF!~rudi@78.29.159.239, Ping timeout: 245 seconds)
20:27xet7 has joined IRC (xet7!~xet7@a88-112-147-81.elisa-laajakaista.fi)
20:51awilliams has left IRC (awilliams!~awilliams@unaffiliated/mistik1, Ping timeout: 264 seconds)
21:26Enslaver has left IRC (Enslaver!~Enslaver@fedora/Enslaver, Ping timeout: 272 seconds)
21:34gbaman has joined IRC (gbaman!~gbaman@host81-130-48-226.in-addr.btopenworld.com)
22:10Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
22:18awilliams has joined IRC (awilliams!~awilliams@unaffiliated/mistik1)
22:40webcomp has joined IRC (webcomp!18017019@gateway/web/freenode/ip.24.1.112.25)
22:45
<webcomp>
Need help on locking down clients - running on Ubuntu 12.04. Just running a WEB kiosk with Chrome in --kiosk mode. Want to disable everything else I can i.e., Launcher, function keys, filesystem, etc. - need assistance please!
23:17hs366 has left IRC (hs366!~hs366@94.254.45.76, Read error: Connection reset by peer)
23:22gbaman has left IRC (gbaman!~gbaman@host81-130-48-226.in-addr.btopenworld.com, Remote host closed the connection)
23:32gbaman has joined IRC (gbaman!~gbaman@host81-130-48-226.in-addr.btopenworld.com)
23:55xet7 has left IRC (xet7!~xet7@a88-112-147-81.elisa-laajakaista.fi, Quit: Lähdössä)