ltsp-client: attempted to send on closed socket

why are you using xinetd ? by default ltsp-server depends on openbsd-inetd since this is the only inetd tested and shipped in ubuntu ltsp installations, i would assume something in your xinetd setup is missing, please refer to the xientd documentation.

I cannot believe that this is a problem of xinetd. I converted the inet.conf via xconv.pl, it should do exactly what inetd does.

Please refer also to similar bugs and bugfixes, e. g. http://lwn.net/Articles/267685/.

I think, it is a fundamental problem of nbdrootd.

Are you able to reboot the server, perform ltsp-update-image, disconnect and reconnect the network on the client without crashes of the clients? If yes, you can safely close this bug; otherwise, you should not.

Attachment: configuration of inetd and xinetd

"why are you using xinetd ?" -- no idea, I just setup a new server, installed ltsp-server-standalone, followed by ltsp-build-client -- then it complained that I'm using xinetd instead of inetd.

Indeed, no client was able to download i386.img at boot time.

So I typed "apt-get install openbsd-inetd"

Lots of scary warning about missing dependencies. But at the end, inetd was installed -- and the clients, which were stuck at boot up screen; was able to download i386.img and continued the LTSP booting process.

No more problem afterwards. I love happy ending :-)

This seems to be the same problem.

http://<email address hidden>/msg35112.html

Solution (see there):

"
> Hello,
>
> because of importance of these probĺems we have now 3 ways to protect
> the clients from freeze because of loosing connections:
>
> 1. TCP-Keepalive tuning (the cleanest way)
> /proc/sys/net/ipv4/tcp_keepalive_time = 600
> /proc/sys/net/ipv4/tcp_keepalive_intvl = 10
> /proc/sys/net/ipv4/tcp_keepalive_probes = 50
>
> 2. Using 'nbd-client' with '-persist'-Option (helps sometimes when 1. fails)
>
> 3. Using 'cron' script, which checks every minute ...
> if (the connection is lost) {
> if (nobody uses that client){
> reboot / shutdown
> }
> }
> Here you have to remember, that the programs 'reboot/shutdown/poweroff'
> and their libs have to be cached, before the connection breaks
>
> Now it works fine: even if somebody does something stupid like turn off
> a switch or disconnects a cable.
>
> Best regards,
>
> Wojtek
"

So could anyone please fix the problem now!

Thanks in advance!

I would suggest adding the -persist parameter to the initrd. However, I don't have any idea how to do that...

There are two rc-scripts left:

/etc/init.d/nbd-client
/etc/init.d/ltsp-client-setup

I would suggest to add the -persist option here, too.

this is an ltsp specific prob, marking invalid for nbd

this was fixed in ltsp-5.1.25 (upstram commit 879 http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/revision/879)

Sorry, I did not remember that I reported the bug as nbd-bug. Could you please move it to the ltsp section. I don't know how to do that...

nominated and confirmed for intrepid (if someone finds the time for an SRU)

OK. Thank you very much!

nominated and confirmed for hardy

Intrepid Ibex reached end-of-life on 30 April 2010 so I am closing the
report. The bug has been fixed in newer releases of Ubuntu.

Hardy has seen the end of its life and is no longer receiving any updates. Marking the Hardy task for this ticket as "Won't Fix".