|01:07||AlexPortable has left IRC (AlexPortable!uid7568@gateway/web/irccloud.com/x-iwmnhpyaxngxrklb, Quit: Connection closed for inactivity)|
|01:26||gp has joined IRC (gp!~Nickname1@adsl-068-016-236-123.sip.asm.bellsouth.net)|
|01:29||Nickname123 has left IRC (Nickname123!~Nickname1@adsl-068-016-236-123.sip.asm.bellsouth.net, Ping timeout: 244 seconds)|
|04:55||vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)|
|05:06||work_alkisg is now known as alkisg|
|06:09||mikkel has joined IRC (firstname.lastname@example.org)|
I'm still having trouble with the hosts overrides function of the dns forwarder (or it could be a DHCP issue). Most clients correctly resolve the host override, but some do not. when I use dig, they are using an external DNS server. the config is identical with other machines, no custom DNS or anything like that on client machinaecs
any ideas where I could start looking to resolve this?
Under the DHCP service settings, I have the pfsense VIP (LAN) IP address, and nothing else, so DHCP clients should in theory be querying pfsense
gdi2k: what is a "hosts override"? To put a name in /etc/hosts so that it overrides your dns server?
oh so sorry, I'm in the completely wrong place
alkisg, I'm supposed to be on the pfsense chat! :/
|06:29||* alkisg assumed that much :) :)|
|07:12||ricotz has joined IRC (ricotz!~rico@ubuntu/member/ricotz)|
|07:27||telex has left IRC (email@example.com, Remote host closed the connection)|
|07:28||telex has joined IRC (firstname.lastname@example.org)|
|07:50||alkisg is now known as work_alkisg|
|08:05||markit has joined IRC (email@example.com)|
|08:27||work_alkisg is now known as alkisg|
|08:39||vmlintu has joined IRC (firstname.lastname@example.org)|
|08:46||vmlintu has left IRC (email@example.com, Ping timeout: 255 seconds)|
|09:01||vmlintu has joined IRC (firstname.lastname@example.org)|
|09:29||vervelak has joined IRC (email@example.com)|
|11:14||Faith has joined IRC (Faith!~paty@unaffiliated/faith)|
|11:15||alkisg is now known as work_alkisg|
|11:30||danau11 has joined IRC (firstname.lastname@example.org)|
|11:34||danau11 has left IRC (email@example.com)|
|12:04||markit has left IRC (firstname.lastname@example.org, Quit: Konversation terminated!)|
|13:32||championofcyrodi has left IRC (email@example.com, Remote host closed the connection)|
|13:39||championofcyrodi has joined IRC (firstname.lastname@example.org)|
|13:48||mikkel has left IRC (email@example.com, Quit: Leaving)|
|14:02||adrianorg has left IRC (firstname.lastname@example.org, Ping timeout: 272 seconds)|
|14:03||adrianorg has joined IRC (email@example.com)|
|14:25||AlexPortable has joined IRC (AlexPortable!uid7568@gateway/web/irccloud.com/x-xedqfcplufcnwtea)|
|14:26||vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)|
|14:44||championofcyrodi has left IRC (firstname.lastname@example.org)|
|15:06||telex has left IRC (email@example.com, Remote host closed the connection)|
|15:08||telex has joined IRC (firstname.lastname@example.org)|
|16:11||championofcyrodi has joined IRC (email@example.com)|
|16:22||epoptes_user5 has joined IRC (epoptes_user5!18abfbd6@gateway/web/freenode/ip.188.8.131.52)|
Im new with epoptes
How does an execute command works?
|16:25||epoptes_user5 has left IRC (epoptes_user5!18abfbd6@gateway/web/freenode/ip.184.108.40.206, Client Quit)|
Do all epoptes users suffer from ADHD or something?
|16:32||* vagrantc suspects it has to do with not understanding how irc works|
|16:53||gvy has joined IRC (gvy!~mike@altlinux/developer/mike)|
|17:21||work_alkisg is now known as alkisg|
Yeah I should probably change the epoptes IRC link to an applet applet with instructions on top to "wait for 1h for answers" :)
we need a netflix-like support platform : support on demands ;)
vagrantc: how about doing an ltsp sprint in debcamp? will anyone else from the ltsp camp be there?
alkisg: i'd love to; not sure if anyone else is planning on being there though
What are the main things on our TODO list for that sprint? There's not enough time to ditch LDM then, is there?
(so that I do some preparation if I can...)
Ah, how about merging the Ubuntu packaging at last? stgraber, will August be a good time for that?
And possibly adding ltsp-pnp to the installer
alkisg: was hoping to get the ubuntu sync sooner
ltsp-pnp support would be good too
|17:37||* vagrantc needs to figure out what to do regarding nfs+overlay incompatibilities ASAP|
I'd love to, but stgraber seems very unresponsive these days... and soon I'll be to busy to help, at least until summer vacations
vagrantc: is that bug reported somewhere?
|17:38||vmlintu has left IRC (firstname.lastname@example.org, Ping timeout: 258 seconds)|
the obvious easy thing is to switch to NBD
alkisg: i should report it, yeah.
basically, there are two bugs ... the NFS+overlay incompatibility
and the overlay support in ltsp
The overlay support is just a couple of lines away, isn't it?
Btw, I don't see why nfs wouldn't be supported as lowerdir... ok upperdirs are more restrictive...
it's already in bzr, but haven't uploaded to Debian yet
alkisg: if you want to get it fixed upstream, that'd be my preferred option! :)
vagrantc: I think I lost you somewhere, do you mean the ltsp initramfs fix?
Is that in debian/ and not on upstream ltsp?
|17:44||* alkisg was reading the kernel lists for the nfs/overlayfs issue...|
alkisg: i mean in kernel
Well, if the kernel doesn't support overlayfs with lowerdir=nfs, I don't think we can workaround it from the ltsp code... :)
I think we can either wait for them to solve it, or switch to nbd, aoe...
Switching to nbd by default might not be too bad, it would make the tutorials easier too, as they would be the same for debian + ubuntu
And with fat clients being used more and more, it does have a good speed benefit there..
It does sound possible for them to fix it...
my favorite approach was NFS+squashfs IMG+overlay
well, my favorite is actually to keep NFS+overlay ...
but to get the NFS+squashfs.img+overlay to work, i need to get some fixes into initramfs-tools ... fixes i mostly want anyways
just switch to CIFS+overlay :)
|17:54||* ogra_ hides in a hole|
been meaning to submit an updated patch for https://bugs.debian.org/468114
vagrantc: nfs instead of nbd for the resilience part?
If so, I think aoe is much more resilient...
ltsp-config aoe ?
last i looked at it it was some ugly manual setup
No ipappend 3 needed, allows user networking et
vblade is rather easy to set up
Either with a .conf file, or by directly running it
alkisg: so if we add ltsp-config aoe ... then it's an option worth exploring
The swap is still a question, it's much easier with nbd
So, for loopback support, maks was 'ok just fix those things'? nice...
specifying root=/dev/etherd/e0.0 ... that seems a bit hard-coded
As hardcoded as /dev/nbd0 :)
alkisg: i'd also like to add loopback support for the NFS hook
alkisg: yes, but /dev/nbd0 is a local device ... isn't /dev/etherd/e0.0 dependent on how AoE is set up on the server?
We need to start an ltsp-sprint-todo page in the wiki :)
The server can publish different partitions on different clients with different names
alkisg: i wonder if we could get some of the debian-edu folks interested in a debcamp LTSP sprint
We can just tell the sysadmin that we expect it there... e.g. like rootpath=/opt/ltsp/i386 for nbd
i.e. is /dev/etherd/e0.0 a reasonable default?
It's like "eth0"... I'm only worried about the non-routability of aoe, would anyone want ltsp over wan?
Or, I don't know if rate limiting and firewalling are possible in ethernet frames...
as long as NBD still works ...
True, he can just switch to those
right now NFS+squashfs requires some really stupid hacks to work around changes made in initramfs-tools last minute to detect /sbin/init
but those would need to be fixed to properly support loop=
Ah there's also xnbd, this should be quite resilient, it even supports caching
basically, we need to figure out a default rootfs for LTSP in stretch
How long can we wait? They might fix the nfs/overlay issue...
currently, LTSP is broken in sid ... the linux 4.x hasn't migrated to stretch yet ... not sure when it will
i hate to leave LTSP with a broken default in the hopes that it will someday get fixed upstream
About e0.0 in aoe, we can use e.g. e67.0, if we still want "67" to mean "ltsp"...
alkisg: but it has to be an integer or something?
It's "shelf / slot"
Hmm we can even use a volume label
blkid will give us all the devices that the client can see
So, "LTSP", "Swap"...
volume labels sound much better than arbitrary numbers
They could even match the root path or at least its ending part, similar to nbdpath=/opt/ltsp/i386
(so that it can be defined by dhcp, by arch, stuff like that)
thought dhcp wasn't a thing here?
The client is still netbooted though until it gets the linux+initramfs
It can get parameters from pxe...
Hrm flow control might be an issue... it gets solved with tcp automatic rate limiting, I don't know how that will work with ethernet frames
So gigabit server + 100 mbps clients could possibly result in the server behaving like 100mbps
that would be bad
vagrantc: with which kernel can I reproduce the nfs+overlay issue? Will 3.13 have it?
alkisg: not sure if ubuntu's kernels have it patched differently ... debian's using mainline overlay(fs) now
Jessie has 3.16?
jessie still has aufs
So I can't reproduce it with a live cd :-/
|19:00||vmlintu has joined IRC (email@example.com)|
https://lkml.org/lkml/2014/10/30/242 no replies though...
But they did fix a similar problem for cdroms, https://lkml.org/lkml/2014/10/28/930
alkisg: ok, so i should probably file a bug in the debian BTS and mark it as affects ltsp, and sometimes the debian maintainers then know where to apply the right pressure :)
alkisg: file a bug on linux
|19:28||gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving)|
|20:04||Faith has left IRC (Faith!~paty@unaffiliated/faith, Quit: Saindo)|
|20:20||vmlintu has left IRC (firstname.lastname@example.org, Ping timeout: 252 seconds)|
|20:56||cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 244 seconds)|
|20:57||cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg)|
|21:22||ricotz has left IRC (ricotz!~rico@ubuntu/member/ricotz, Quit: Ex-Chat)|
|21:27||awei has joined IRC (awei!185bd65d@gateway/web/freenode/ip.220.127.116.11)|
|21:28||danau11 has joined IRC (email@example.com)|
|21:28||danau11 has left IRC (firstname.lastname@example.org)|
Does the "kiosk" ltsp client handle USB devices differently than the "ldm" one?
i don't think it supports it at all if i remember correctly
oh ok... How would I go about adding support for USB peripheral on a kiosk image?
|21:40||telex has left IRC (email@example.com, Remote host closed the connection)|
|21:42||telex has joined IRC (firstname.lastname@example.org)|
|21:56||awei has left IRC (awei!185bd65d@gateway/web/freenode/ip.18.104.22.168, Ping timeout: 246 seconds)|
|22:11||Nickname123 has joined IRC (Nickname123!~Nickname1@adsl-068-016-236-123.sip.asm.bellsouth.net)|
|22:14||gp has left IRC (gp!~Nickname1@adsl-068-016-236-123.sip.asm.bellsouth.net, Ping timeout: 255 seconds)|
|22:19||vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)|
|22:27||AlexPortable has left IRC (AlexPortable!uid7568@gateway/web/irccloud.com/x-xedqfcplufcnwtea, Quit: Connection closed for inactivity)|