|01:19||spectra has left IRC (spectra!~spectra@debian/developer/spectra, *.net *.split)|
|01:19||bigpresh has left IRC (bigpresh!~bigpresh@freenode/staff/bigpresh, *.net *.split)|
|01:22||bigpresh has joined IRC (bigpresh!~bigpresh@freenode/staff/bigpresh)|
|01:29||spectra has joined IRC (spectra!~spectra@debian/developer/spectra)|
|01:34||lucascastro has joined IRC (firstname.lastname@example.org)|
|01:49||markus_e92 has left IRC (markus_e92!~markus_e9@2a01:4f8:120:244a::2, *.net *.split)|
|01:55||markus_e92 has joined IRC (markus_e92!~markus_e9@2a01:4f8:120:244a::2)|
|03:42||lucascastro has left IRC (email@example.com, Remote host closed the connection)|
|03:46||ogra has left IRC (ogra!~ogra_@ubuntu/member/ogra, Ping timeout: 252 seconds)|
|05:23||ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)|
|06:27||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|06:37||Natureshadow has joined IRC (Natureshadowfirstname.lastname@example.org)|
|06:59||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Ping timeout: 240 seconds)|
|07:00||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|07:23||ogra has joined IRC (ogra!~ogra_@ubuntu/member/ogra)|
alkisg, you'll be happy to hear: I might be upgrading my first LTSP installation to fat clients soon.
We're faced with the choice of either replacing our 8 year old thin clients or upgrading our 5 year old beast of a server.
Hyperbyte: haha, well that's not a good thing, I think you were the last person still (reliably) testing thin clients :D
|08:19||* alkisg certainly votes choosing fat clients though|
It's a surprise to me the thin clients have lived this long. We're now looking into some Intel NUC boxes.
It was a good choice back then, because we didn't want computers making noise and collecting dust under every desk. Too much maintenance and too much hassle.
But now that those mini PC's are getting more and more powerful that kinda takes away most benefits of thin clients...
I don't think there's a difference in size/noice between thin/fat clients. Well, at least for some models.
Not anymore, no.
Back in 2011 (when we implemented thin clients) it wasn't feasible to get a powerful processor in a small box without fans. ;-)
I think I've seen fanless fat clients way before then
But not powerful enough for our liking and budget.
More expensive, sure, but you'd need a much less powerful server then
|08:26||enaut[m] has joined IRC (enaut[m]!enautmatri@gateway/shell/matrix.org/x-hksfaoqqsyirpmfd)|
alkisg, we needed the powerful server anyway, because we do all our hosting in-house.
Considering everything, we did make the right choices I think.
Meanwhile. I have two questions.
1. How hard is it to move from LTSP traditional thin clients to LTSP-chrootless?
2. How feasible is it to get NBD to store the entire LTSP fat client image in the RAM instead of on the server's disk?
1) will the server be a "client template", or it will have all those hosting services you need?
I currently dedicate 32GB of RAM to the thin clients, so when that frees up, I could technically store the entire fat client disk there.
2) very easy, but it's usually a bad idea, except for kiosks. What benefits do you _think_ you'll get with that?
Ah sorry misread (2)
Hrm. I don't see much point in it, since linux already caches all of NBD if ram suffices
So yeah same question "What benefits do you _think_ you'll get with that?"
Ah that's good. So my work is done in that regard. :-)
Well I think reading from RAM is faster than reading from disk, for NBD. But if it's loading the images into the RAM already then that's great.
Regarding 1, my LTSP server is virtual. The hosting servers are also virtual, but on seperate machines. So yeah, currently, I have one virtual server only doing thin clients.
So it'll fit you fine then
So how do I go about doing that?
In a non-destructive way, because for a while I'd like to run both fat and thin clients. I can of course always create two virtual machines (I might do that anyway and let the DHCP server point to the new one)
But let's say I make a copy of my thin client server and then I want to switch it over to ltps-chrootless. Is it a matter of apt-get'ing the right packages and running ltsp-build-image / ?
|09:05||hardbidouille67 has joined IRC (email@example.com)|
Hyperbyte: you can use the chrootless image to boot thin clients too, and just point them to a different LDM_SERVER via lts.conf
|10:31||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|10:59||hardbidouille67 has left IRC (firstname.lastname@example.org, Ping timeout: 240 seconds)|
|11:05||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|11:09||hardbidouille67 has joined IRC (email@example.com)|
|11:09||alkisg has joined IRC (alkisg!822b5e11@ubuntu/member/alkisg)|
|11:17||hardbidouille67 has left IRC (firstname.lastname@example.org, Ping timeout: 252 seconds)|
|11:30||hardbidouille67 has joined IRC (email@example.com)|
alkisg: I think the launchpad is back up today! I was gonna ask for an alternative Greek Schools repo.
|11:40||Faith has joined IRC (Faith!~Paty_@unaffiliated/faith)|
Oh wait... I think it's gonna time out again
I can connect to everything fine, but this Greek Schools launchpad PPA
Interesting, my LTSP server can't ping it, but my other clients can...
Helenah: for launchpad questions, ask in #launchpad. It works fine here. For networking issues, in #networking.
This isn't a general support channel, it's strictly for ltsp issues
Wait... there is #launchpad? I didn't know!
But if you ask in #launchpad about your networking issues, of course you won't get help...
It's only for actual launchpad problems, not support for networking issues for any user
I was asking in Ubuntu yestaday with no response, but I'm not thinking it's a network issue on my behalf. I'm probably going to ask in #ubuntu for support on that.
I'm thinking it might be an ipv6 issue with your networking or isp
alkisg: Thanks for clarifying that you can connect to launchpad.
While clients may be using ipv4
I don't have IPv6 yet
All default installations have ipv6
I didn't see the point when my ISP doesn't support it
And `host launchpad.net` will tell you; but since it's offtopic, I'll stop at that :)
Yes, ofc, my apologies
btw, there is https://twitter.com/launchpadstatus
|12:01||lucascastro has joined IRC (firstname.lastname@example.org)|
|12:07||adrianorg has left IRC (email@example.com, Ping timeout: 245 seconds)|
|12:21||yopla has joined IRC (yopla!25479532@gateway/web/freenode/ip.126.96.36.199)|
hello, I've got trouble implementing ltsp on ubunto 18.04. at boot time client says "unknown error in reply to nbd_opt_go". I'm stuck in initramfs.
both client end server are 3.16
yopla: did you follow the official guide
install: http://wiki.ltsp.org/wiki/Installation/Ubuntu for Ubuntu, or http://wiki.ltsp.org/wiki/Installation for other distributions
humm, thank you. this seems repository not mounted in chroot
The stock ubuntu version has known issues, don't use it
humm, thank you. I have done it on server side only
by the way my last NUC hardware forced me to upgrade to 18.04 cause of buggy realtek
strange, it should be possible to just upgrade the kerenl
But that's a question for #ubuntu...
Well, I haven't tried updgrade the kernel only
Rebuild in progress. I'll check if it will work. Thank you
|12:40||adrianorg has joined IRC (firstname.lastname@example.org)|
still not working but I have other error. Seems rootpath not set correclty
With the correct root-path I get the same error :/
sorry, I've got better output syslog server side
"client is noit authorize to access"
I removed /etc/ltsp/nbd-server.allow
yopla: this file doesn't exist in normal ltsp installations
Did you manually create it for some reason?
Yes, it was an attempt to have nbd working
OK, it should be working without it, and broken with it...
without it it didn't work. But the root problem was bad root-path in DHCP I think. Then bad nbd-server.allow. lol
Another problem is this hardware is pxe efi only. No way to revert to legacy PXE in bios.
should consider implementing efi in ltsp-update-kernel
alkisg: Okay I was advised to wget the timing out packages and place them into the APT cache within the chroot, however I now need to force ltsp-build-client to continue on an existing chroot.
I do not know about 'local', but I do know about these similar topics: 'install-localapp', 'localdev', 'RDP-localdev', 'LocaldevCommonGroupWorkaround', 'localxterm', 'local-disks', 'local-boot', 'localapps'
local-boot: If you want LTSP fat clients on a low-speed network, you can put i386.img on e.g. C:\Boot\LTSP\i386.img and use this command line in pxelinux.cfg: APPEND ro initrd=ltsp/i386/initrd.img init=/sbin/init-ltsp root=/dev/sda1 rootflags=ro loop=/Boot/LTSP/i386.img; IPAPPEND 3
I do not know about 'kernel', but I do know about these similar topics: 'ltsp-update-kernels'
Anyway there's an easy way to put the kernel locally, and load the os from the network, it works on uefi too
Currently ltsp doesn't support uefi
Helenah: debootstrap doesn't support continuing. You can only do ltsp-build-client --mount-package-cache, to remember the downloaded packages while restarting from scratch
I'll give that a try.
I had better success in efi pxe boot with ipxe http://ipxe.org/howto/chainloading#uefi. Not much to do with config files.
alkisg: and where is the package cache mount/
Helenah: google 'ubuntu package cache directory'
yopla: ipxe works in some cases. If you have local storage, or if you can configure undi.
grub works in more cases, pxelinux.efi in a few others
Hopefully undi is supported on my hardware. Goal is not to have local storage in the boxes.
boot is much slower than legacy pxe though :/
You mean kernel loading? Because after that, linux runs, so it makes no difference if it's uefi or not
yes kernel loading
Eh, so it's only a couple of seconds difference in boot times? Not significant then...
you're right :)
|14:02||* alkisg waves, later...|
|14:02||alkisg has left IRC (alkisg!822b5e11@ubuntu/member/alkisg, Quit: Page closed)|
|14:28||yopla has left IRC (yopla!25479532@gateway/web/freenode/ip.188.8.131.52, Ping timeout: 252 seconds)|
|14:33||hardbidouille67 has left IRC (email@example.com, Ping timeout: 252 seconds)|
|15:10||alkisg has joined IRC (alkisg!822b5e11@ubuntu/member/alkisg)|
|15:30||GodFather has left IRC (GodFatherfirstname.lastname@example.org, Read error: Connection reset by peer)|
|15:46||pppingme has left IRC (pppingme!~pppingme@unaffiliated/pppingme, Read error: Connection reset by peer)|
|15:47||alkisg has left IRC (alkisg!822b5e11@ubuntu/member/alkisg, Quit: Page closed)|
|15:50||GodFather has joined IRC (GodFatheremail@example.com)|
|16:12||vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)|
|17:27||lucascastro has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
|17:45||lucascastro has joined IRC (email@example.com)|
|18:00||lucas_ has joined IRC (firstname.lastname@example.org)|
|18:04||lucascastro has left IRC (email@example.com, Ping timeout: 246 seconds)|
|18:07||lucas__ has joined IRC (firstname.lastname@example.org)|
|18:10||lucas_ has left IRC (email@example.com, Ping timeout: 240 seconds)|
|18:20||adrianorg has left IRC (firstname.lastname@example.org, Ping timeout: 240 seconds)|
|18:20||adrianorg has joined IRC (email@example.com)|
So I just need to copy the packages which time out to /var/cache/apt/archive/?
I did this, ltsp-build-client still says the repo has timed out and fails. I don't think I'm understanding how --mount-package-cache works.
highvoltage: Yo, I think I ran into your blog today.
I haven't blogged in a while :)
|20:25||Faith has left IRC (Faith!~Paty_@unaffiliated/faith, Quit: Leaving)|
|20:38||ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)|
|22:35||pppingme has joined IRC (pppingme!~pppingme@unaffiliated/pppingme)|
|22:37||pppingme has joined IRC (pppingme!~pppingme@unaffiliated/pppingme)|
|22:39||pppingme has left IRC (pppingme!~pppingme@unaffiliated/pppingme, Remote host closed the connection)|
|22:40||pppingme has joined IRC (pppingme!~pppingme@unaffiliated/pppingme)|
|22:42||pppingme has left IRC (pppingme!~pppingme@unaffiliated/pppingme, Read error: Connection reset by peer)|
|23:21||vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)|