1498349667 J * fstd_ ~fstd@x4e310358.dyn.telefonica.de
1498349671 Q * fstd Read error: Connection reset by peer
1498349683 N * fstd_ fstd
1498353543 M * Bertl_oO  off to bed now .. have a good one everyone!
1498353545 N * Bertl_oO Bertl_zZ
1498361822 Q * romster Remote host closed the connection
1498361854 J * romster ~romster@158.140.215.184
1498381119 J * bonbons ~bonbons@2001:a18:235:d701:60fe:985a:5075:8ea3
1498381121 M * arekm Bertl_zZ: do you plan to update vserver patch to some maintained longterm kernel in near future?
1498381906 Q * sannes Remote host closed the connection
1498381916 J * sannes ~ace@2a02:fe0:c130:1d90:e54a:32b6:5944:1874
1498382701 J * alumno ~alumno@186.182.15.134
1498383251 Q * l0kit Server closed connection
1498383266 J * l0kit ~1oxT@0001b54e.user.oftc.net
1498383392 Q * alumno Quit: Saliendo
1498383571 Q * yang Server closed connection
1498383574 J * yang yang@pine64.gnu.si
1498383822 Q * guerby Quit: Leaving
1498384210 N * Bertl_zZ Bertl
1498384213 M * Bertl  morning folks!
1498384287 M * Bertl arekm: 4.1.41 is long term and maintained, no?
1498384317 M * Bertl but in case you are asking about a more recent long term kernel, there are plans to support that too
1498385120 M * arekm 4.1 is maintained in theory. It's kind of single drop every month (and very late with fixes; especially when compared to 4.4 or 4.9 series)
1498385123 M * arekm ok
1498385125 M * arekm thanks
1498385654 Q * romster Read error: Connection reset by peer
1498385689 J * romster ~romster@158.140.215.184
1498386249 J * thierryp ~thierry@lns-bzn-40-82-251-147-27.adsl.proxad.net
1498386464 Q * fosco Server closed connection
1498386468 J * fosco fosco@91.208.40.1
1498387173 M * yang So I installed BenG's kernel for Debian stretch, but for some reason I am unable to change and "grub-update" the boot order...also "wget -O - http://repo.psand.net/pubkey.txt | sudo apt-key add -" fails to download the right PGP key..
1498387226 M * yang http://paste.debian.net/plainh/bb0ec776
1498387242 M * yang It keeps defaulting to 4.9.0-3-686-pae
1498387285 M * yang that one is from 0 till 3 in grub.cfg, so I changed to 4 for BenG 4.1.39 but it keeps rebooting into 4.9.0-3-686-pae
1498387339 Q * arekm Server closed connection
1498387341 J * arekm ~arekm@phobos.pld-linux.org
1498387979 Q * sannes Ping timeout: 480 seconds
1498388936 Q * ntrs Server closed connection
1498388938 J * ntrs ~ntrs@vault08.rosehosting.com
1498391027 M * yang ok I figured it out
1498391029 M * yang instead of GRUB_DEFAULT="4", it had to be GRUB_DEFAULT="1>3" because of the submenu's order
1498391088 M * Bertl welcome to the new grub! :)
1498391155 Q * AndrewLee Server closed connection
1498391169 J * AndrewLee ~andrew@210.240.39.201
1498393352 Q * jrklein Server closed connection
1498393483 J * jrklein ~cloud@proxy.dnihost.net
1498394556 M * Bertl  off for now ... bbl
1498394557 N * Bertl Bertl_oO
1498395194 J * guerby ~guerby@ip165.tetaneutral.net
1498398103 Q * Aiken Remote host closed the connection
1498398690 Q * guerby Quit: Leaving
1498398698 J * guerby ~guerby@ip165.tetaneutral.net
1498398966 Q * guerby 
1498398995 J * guerby ~guerby@ip165.tetaneutral.net
1498401979 Q * mnemoc Server closed connection
1498401979 J * mnemoc ~amery@kwa.jpi.io
1498402026 Q * _Shiva__ Server closed connection
1498402027 J * _Shiva_ shiva@whatcha.looking.at
1498402411 Q * DLange Server closed connection
1498402421 J * DLange ~DLange@dlange.user.oftc.net
1498402904 Q * padde Remote host closed the connection
1498403146 J * padde ~padde@patrick-nagel.net
1498403669 Q * thierryp Ping timeout: 480 seconds
1498404855 J * Ghislain ~ghislain@81.56.195.31
1498405755 M * Ghislain i wonder if the ip sharing in the kernel could reprogrammed to use ebpf instead of patching the ip code
1498406030 M * Bertl_oO no idea what you want to tell us here ... :)
1498406099 M * Ghislain you know the new kernels from 4.1 and forward have ebpf a new ssytem writen to include little pseudo coe in the kernel. Seems it is really well suited for networking filtering
1498406120 M * Ghislain i was thinking out loud here
1498406133 M * Bertl_oO we are talking extended berkley packet filter, I got that
1498406154 M * Bertl_oO but what do you want to filter?
1498406171 M * Ghislain to isolate ip to guest
1498406206 M * Ghislain i was woindering if using this framework could permit to prevent patching of the ip code to use this instead
1498406242 M * Bertl_oO first, it wouldn't do anything on the virtualization side
1498406277 M * Bertl_oO secondly, if you are fine with adding overhead, network namespaces are the way to go anyway
1498406317 M * Ghislain ok i see
1498406411 M * Ghislain the issue with namespace is that you cannot share an ip on 2 vps
1498406487 M * Bertl_oO you can share the namespace, no?
1498406559 M * Ghislain for me you are creating a new virtual nic so it does not seems possible to have several vps owning a network card but i never used it in the flesh
1498406574 M * Ghislain i could just be missreaading that
1498406658 M * gnarface i'm only partially following, but doesn't the host own the virtual network devices?
1498406732 M * gnarface secondarily, wouldn't you just make two and bridge them anyway?
1498406745 M * gnarface maybe i don't understand what the goal is
1498406782 M * Ghislain share one ip on multiple guest
1498406813 M * Ghislain i was beleiving the namespace do not allow that
1498406819 M * gnarface hmmm.  i guess i don't know
1498406944 M * gnarface in my use cases, i'm usually going through extra effort to do the opposite
1498406992 M * Ghislain mostly i use one guest on one machine to isolate the service from the host, ease migration and access in case the customer breaks everything
1498407018 M * Ghislain and also they cannot mess with the network
1498407117 M * Ghislain i do not like the idea of creating false switch and network inteface in the kernel just to isolate where a filter seems to be able to do the job 
1498407139 M * Ghislain of course this is weaker as you have not complete separation and cannot use iptables on guests
1498407167 M * Ghislain but i just feel this is better to filter instead of emulate a complete switching archtitecture
1498408493 M * Ghislain well i just wondered if we could use ebpf like we use cgroup to replace cpu limit but in network part, just thinking out loud :)
1498408524 M * Ghislain but yes it will not filter the network name of interface etc..
1498408553 M * gnarface an interesting notion
1498408591 M * gnarface i'm not sure i agree with the part about the emulated network thoguh
1498408614 M * gnarface i suppose it depends on what you're doing with it
1498408631 M * Ghislain well it is not KISS compatible to emulate 2 network card and a switch to limit a guest to an ip for me
1498408648 M * Ghislain seems a little toot complicated for the task
1498408787 M * Ghislain but i agree if you want to emulate real architecture to use tools that use them the emulation is quite legit
1498410811 Q * clopez Server closed connection
1498410828 J * clopez ~tau@neutrino.es
1498413379 Q * webhat Server closed connection
1498413390 J * webhat ~quassel@31.25.99.5
1498417378 J * thierryp ~thierry@lns-bzn-40-82-251-147-27.adsl.proxad.net
1498418954 M * arekm https://lwn.net/Articles/726405/
1498420146 Q * fb Server closed connection
1498420148 J * fb fback@red.fback.net
1498422552 J * Aiken ~Aiken@d63f.h.jbmb.net
1498423885 Q * bonbons Quit: Leaving
1498426027 Q * tokkee Server closed connection
1498426029 J * tokkee tokkee@osprey.tokkee.org
1498431763 Q * Defaultti Server closed connection
1498431789 J * Defaultti defaultti@lakka.kapsi.fi
1498432507 Q * transacid Server closed connection
1498432526 J * transacid ~transacid@transacid.de