1332288000 Q * dowdle 
1332288235 Q * clopez Ping timeout: 480 seconds
1332289196 J * clopez ~clopez@82.25.60.213.dynamic.mundo-r.com
1332289900 J * misc-- ~misc@202.171.160.4
1332289937 M * misc-- hello! can vserver run on top of xen?
1332290077 M * Bertl yup
1332290088 M * misc-- ahhh ok
1332290106 M * misc-- I just have to say what I'm thinking of doing here... it's ok if you're not sure
1332290124 M * misc-- so basically we host four virtual servers inhouse here, with a number of guests no each
1332290155 M * misc-- but seeing I'm leaving the company, we've decided to move it all to a managed hosting environment
1332290230 M * misc-- and seeing that you buy cloud servers which usually run xen, it costs money to buy an extra xen guest. So ok, thought we might as well lump everything on 2 or 3 xen guests but now just looking at all of my guests that I run here, it would be much better to have say 3 xen guests on our managed provider and split that up in to multiple vserver guests
1332290277 M * Bertl works fine
1332290301 M * misc-- not sure if the provider would support managing all of that, but we have guys here that can do basic building/removing guests and all that. 
1332290302 M * misc-- ok cool
1332292122 M * Bertl off to bed now ... have a good one everyone!
1332292127 N * Bertl Bertl_zZ
1332292828 Q * imcsk8 Ping timeout: 480 seconds
1332293786 J * nyerup_ irc@jespernyerup.dk
1332293867 Q * WMP Ping timeout: 480 seconds
1332293889 J * WMP ~oftc@auburn.sored.pl
1332294019 Q * nyerup Ping timeout: 480 seconds
1332294019 N * nyerup_ nyerup
1332300769 J * ghislain ~AQUEOS@adsl2.aqueos.com
1332303333 Q * ghislain Quit: Leaving.
1332306845 Q * Aiken Remote host closed the connection
1332308542 J * derjohn_mob ~aj@88.128.225.200
1332311205 J * voegelas ~voegelas@HSI-KBW-109-192-051-187.hsi6.kabel-badenwuerttemberg.de
1332311444 Q * clopez Ping timeout: 480 seconds
1332312700 Q * derjohn_mob Ping timeout: 480 seconds
1332312798 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no
1332315007 J * arekm matrix157@91.234.176.246
1332315015 M * arekm hi
1332315042 M * arekm Bertl_zZ: did hidepid turn out to be a bigger problem?
1332316086 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1332316091 Q * hijacker_ Read error: Connection reset by peer
1332317439 J * hola c193def5@ircip4.mibbit.com
1332320885 J * ghislain ~AQUEOS@adsl2.aqueos.com
1332323813 M * Bl4ckB1rD did yesterday new kernel come out ?
1332323852 M * Bl4ckB1rD seems like kernel.org is down -.- can't check
1332323865 M * Bl4ckB1rD but i do see new kernel and patch @ vserver site
1332323865 M * arekm 3.3 is released
1332323874 M * Bl4ckB1rD is that stable release ?
1332323915 M * arekm it's "mainline" in kernel.org terms
1332323960 M * Bl4ckB1rD i see
1332323963 M * Bl4ckB1rD good.
1332323966 M * Bl4ckB1rD thanks
1332324933 J * kir ~kir@swsoft-msk-nat.sw.ru
1332325692 N * Bertl_zZ Bertl
1332325702 M * Bertl morning folks!
1332325717 M * Bertl arekm: in what way?
1332325739 M * arekm Bertl: no fix, so far ;))
1332325766 M * Bertl ah, well, I didn't find the time to look into it yet :)
1332325793 M * Bertl but I do not consider it a problem as it should work fine in the 'compatible' mode, no?
1332325828 M * arekm well, I previously used grsecurity for this feature but now switched to using mainline one
1332325878 M * arekm so it is a problem for people like me (in pld it's very commonly used grsec or now mainline based feature)
1332325879 M * Bertl so it's you who is interested in a proper solution then :)
1332325910 M * Bertl anyway, as I said, I don't think it is a big problem per se
1332325942 M * Bertl i.e. the simplest way would be to make that flag per mount, probably mainline will do that anyways for chroot mounts
1332325967 M * arekm flag per mount doesn't solve anything for me at least
1332325982 M * Bertl I presume it isn't per mount atm
1332325999 M * Bl4ckB1rD morning bertl
1332326037 M * Bertl give me a few minutes I take a quick peek what can be done
1332326072 M * arekm cool, thanks!
1332326634 M * Bertl as I suspected, the option is currently attached to the pid namespace
1332326705 M * Bertl the quickest solution would be to override it in guests with a guest specific flag, would that work for you?
1332326878 M * arekm and what would it done? I want hidepid to work in guests, too for example. I only miss vserver-stat and vps working
1332327012 M * Bertl okay, so you already _tested_ it inside the guests and it works as expected there?
1332327029 M * Bertl because we could simply disable it in xid=1
1332327059 M * Bertl that should make spectator stuff work regardless of the settings
1332327069 M * arekm almost works in guests - "ps aux" shows only my processes bug pstree  shows nothing
1332327099 M * arekm s/bug/but/
1332327110 M * Bertl is pid 1 visible?
1332327122 M * Bertl (that's usually the problem with pstree :)
1332327178 M * arekm not visible
1332327201 M * Bertl okay, you got some time for testing?
1332327223 M * arekm yes
1332327241 M * Bertl as I said, proper solution will be to adapt 'our' proc changes to the upstream environment
1332327265 M * Bertl but there are a few quick and dirty hacks we can do to make it work for now
1332327327 M * Bertl for example, there is has_pid_permissions() in fs/proc/base.c
1332327379 M * Bertl we might simply skip the two hidepid related checks for xid=1 
1332327510 M * Bertl try to add 
1332327555 M * Bertl if (vx_check(0, VS_WATCH_P))
1332327588 M * Bertl     return true;
1332327607 M * Bertl at the beginning of has_pid_permissions()
1332327666 M * Bertl btw, does pstree work on the host with hidepid on as user?
1332327873 P * kir PING 1332327873
1332327960 M * arekm hah, it doesn't :)
1332328002 M * Bertl yeah, guess it's time to finally fix pstree :)
1332328062 M * Bertl after just 10 years :)
1332328428 M * arekm https://sourceforge.net/tracker/?func=detail&aid=3509672&group_id=15273&atid=115273
1332328462 M * arekm Bertl: would vserver-status also be fixed by if (vx_check(0, VS_WATCH_P)) in has_pid_permissions() ?
1332328503 M * Bertl I presume that all spectator related commands/tools should work with that
1332328512 M * Bertl it won't fix the pstree issues though
1332328609 M * arekm ok, thesting then
1332329335 Q * brambles Remote host closed the connection
1332331839 M * arekm Bertl: hmm, doesn't work. vserver-stat sees nothing, vps only two processes like before
1332331935 M * arekm too bad that I didn't add some printk there
1332331943 M * daniel_hozac vserver-stat runs in the host.
1332331983 M * arekm so?
1332331992 M * arekm hidepid affects host too
1332331994 M * daniel_hozac VS_WATCH_P won't be enough
1332332131 M * Bertl so, can vserver-stat work without modification at all when hidepid is enabled (on the host)?
1332332239 M * daniel_hozac i would guess not.
1332332320 J * clopez ~clopez@82.25.60.213.dynamic.mundo-r.com
1332332364 M * Bertl okay, arekm please try with VS_ADMIN | VS_WATCH instead then
1332332393 M * daniel_hozac vserver-stat collects VSZ data by checking on the processes.
1332332417 M * Bertl but it could do that from the spectator context, no?
1332332460 M * arekm vcontext --migrate --xid 1 -- ps aux
1332332465 M * arekm returns only own process
1332332543 M * arekm Bertl: without _P?
1332332559 M * Bertl well, doesn't matter if you have guest privacy disabled
1332332587 M * Bertl if you have guest privacy enabled, it will override that
1332332607 M * arekm I have it enabled
1332332623 M * Bertl that's why you don't get anything from the ps
1332332716 M * arekm ok, checking VS_ADMIN + VS_WATCH then
1332332743 M * Bertl but does vserver-stat work with guest privacy enabled at all?
1332332752 M * arekm yes it does. vps aux also sees everything
1332332763 M * arekm without hidepid of course
1332332769 M * Bertl sounds like a bug :)
1332333137 M * arekm it would have to be multi-years bug then
1332333168 M * Bertl yup, like the one in pstree :)
1332333623 M * arekm this VSERVER_PRIVACY is supposed to be all about host-guest or guest-guest too?
1332333648 M * Bertl it basically protects guests from the host admin (as far as possible)
1332333659 M * arekm so I'll disable it here 8)
1332336384 Q * hola Quit: http://www.mibbit.com ajax IRC Client
1332339130 M * Bl4ckB1rD i copied vserver from my centos6 server to my ubuntu host and trying to start it i get: vc_set_sched(): Function not implemented, is this kernel related (i recently built it), or is it vserver related and i should build one from repository?
1332339141 M * Bl4ckB1rD An error occured while executing the vserver startup sequence; when there are no other messages, it is very likely that the init-script (/etc/rc.d/rc 3) failed.
1332339145 M * Bl4ckB1rD that's the error msg
1332339161 M * Bl4ckB1rD but adding: return true or echo true in that rc file doesnt do any good
1332339320 M * Bl4ckB1rD Versions:                    Kernel: 3.0.24-vs2.3.2.3-black                    VS-API: 0x00020308              util-vserver: 0.30.215; Aug  7 2009, 02:57:43
1332339343 M * daniel_hozac you probably have an ancient util-vserver.
1332339400 M * Bertl 0.30.215 is what? 4 years old or so?
1332339406 M * Bl4ckB1rD 2009 :)
1332339413 M * Bl4ckB1rD i'll download latest :)
1332339421 M * Bl4ckB1rD daniel i have your's from repository.
1332339434 M * Bl4ckB1rD i think at least
1332339461 M * Bl4ckB1rD ah nevermind
1332339463 M * Bl4ckB1rD nope.
1332339467 M * Bl4ckB1rD it's one from debian repo
1332339473 M * Bl4ckB1rD it's probably old one :)
1332339477 M * Bl4ckB1rD i'll complie newest
1332339480 M * Bl4ckB1rD and hopefully it works
1332339482 M * Bertl for debian 'anicent' I presume :)
1332339489 M * Bl4ckB1rD lol
1332339490 M * Bl4ckB1rD :D
1332339545 Q * clopez Ping timeout: 480 seconds
1332339645 M * ghislain bertl: you told here that you heard 3.2 was not performing well. On my side i do not feel any performance issue. The only thing i see if better tcp/ip stack performance, better scheduler etc nothing about performance loss
1332339683 M * Bertl I said that there were reports of issues, I didn't say that there actually were any issues
1332339717 M * ghislain ah ok, you frightened me ! i am upgrading the kernel everywhere i can with 3.2
1332339733 M * Bl4ckB1rD works now :)
1332339738 M * ghislain i found this : http://openbenchmarking.org/result/1201066-BY-LINUX32KE16 seems good and a lot better tant 2.6.39
1332339741 M * Bl4ckB1rD when i recompiled latest utils
1332339786 M * ghislain you need fresh utils, they loose vitamins when they are too old
1332339976 M * arekm Bertl: privacy disabled,  if (vx_check(0, VS_ADMIN_P|VS_WATCH_P)) and works now (both, vserver-stat and vps)
1332340019 M * Bertl might be interesting if it work with just
1332340030 M * Bertl vx_check(0, VS_WATCH_P) 
1332340038 M * Bertl when guest privacy is disabled
1332340116 M * arekm well, I'll test it
1332340236 M * Bl4ckB1rD ghislain haha, okay :) i infused it with new version and it works perfectly. thanx.
1332341015 M * ghislain bertl: from my search 3.2.6 and lower had some issue in various part but all seems ok with the older ones. Will tell you if i find any problems with them. For now it works liek a charm and blkio limits are really a blessing for me :)
1332341056 M * ghislain rules of thumb one: allways use the latest utils, after you had them only think about the kernel :)
1332341077 M * ghislain i seems case where i loosed completly all isolation by using old tools so now i know :p
1332341498 J * dowdle ~dowdle@scott.coe.montana.edu
1332341660 M * Bl4ckB1rD i see
1332341862 M * arekm Bertl: works, so watch check is enough
1332341872 M * arekm Bertl: watch_p
1332341890 M * Bertl excellent! thanks!
1332342085 Q * arekm Quit: leaving
1332344579 J * brambles brambles@79.133.200.49
1332344601 M * Bl4ckB1rD cool daniel already included debian directory for building .deb package out of latest vserver-utils.
1332344610 M * Bl4ckB1rD awesome
1332345981 J * clopez ~clopez@155.99.117.91.static.mundo-r.com
1332346485 Q * Bl4ckB1rD Quit: http://www.mibbit.com ajax IRC Client
1332346705 J * derjohn_mob ~aj@87.253.171.221
1332347157 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com
1332348925 Q * nicholi Read error: Connection reset by peer
1332348976 J * nicholi ~nicholi@rrcs-76-79-196-34.west.biz.rr.com
1332349603 J * bonbons ~bonbons@2001:960:7ab:0:7526:6a88:3359:4aae
1332350108 J * arekm ~arekm@89-69-21-174.dynamic.chello.pl
1332350131 M * arekm Bertl: pstree workaround - https://sourceforge.net/tracker/download.php?group_id=15273&atid=115273&file_id=438847&aid=3509672
1332350194 M * arekm oh crap, doesn't really work 8)
1332350213 Q * ensc|w Remote host closed the connection
1332350222 J * ensc|w ~ensc@www.sigma-chemnitz.de
1332351029 Q * voegelas Quit: Leaving.
1332351331 Q * derjohn_mob Ping timeout: 480 seconds
1332351375 Q * clopez Ping timeout: 480 seconds
1332352066 J * clopez ~clopez@155.99.117.91.static.mundo-r.com
1332352148 Q * BenG Remote host closed the connection
1332355155 M * arekm https://sourceforge.net/tracker/download.php?group_id=15273&atid=115273&file_id=438855&aid=3509672
1332355161 M * arekm this one works in both cases
1332356103 Q * arekm Quit: leaving
1332356570 Q * clopez Ping timeout: 480 seconds
1332357245 J * clopez ~clopez@155.99.117.91.static.mundo-r.com
1332360135 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1332360355 Q * clopez Ping timeout: 480 seconds
1332361027 J * clopez ~clopez@155.99.117.91.static.mundo-r.com
1332361607 Q * hijacker_ Quit: Leaving
1332362809 J * derjohn_mob ~aj@tmo-089-64.customers.d1-online.com
1332363248 J * vltko ~blagoj@89.205.105.151
1332363258 M * vltko Hi guys.
1332363269 J * cuba33ci_ ~cuba33ci@114-36-240-66.dynamic.hinet.net
1332363285 M * vltko I have a serevr and I want to pu 2-3 virtual serevrs on it, 1 mail sever and 1 web etc....
1332363294 M * vltko Which kernel will you recommend me?
1332363301 Q * cuba33ci Read error: Operation timed out
1332363304 M * vltko I do not want to use 2.6.22 because it is just too old.
1332363307 M * Bertl 3.x should be fine
1332363318 N * cuba33ci_ cuba33ci
1332363328 M * vltko anyone particular?
1332363346 M * vltko 3.1, 3.2, 3.3 ...?
1332363362 M * Bertl I'd avoid 3.1, it's not actively maintained
1332363380 M * Bertl 3.3 is quite new, so not much feedback yet
1332363389 M * mnemoc only 3.0 will be maintained longer
1332363390 M * vltko so... 3.2..?\:)
1332363396 M * vltko aahhh ok
1332363402 M * vltko I will than use 3.0
1332363403 M * Bertl 3.0 or 3.2, depending on your preference
1332363404 M * vltko Thank you.
1332363415 M * vltko No no, I will liten your advice and use 3.0
1332363420 M * vltko listen*
1332363566 M * mnemoc according to http://www.kroah.com/log/linux/stable-status-01-2012.html 3.2 is EOL with the release of 3.3
1332363627 Q * clopez Ping timeout: 480 seconds
1332363631 P * vltko 
1332365406 M * Wonka they still released 3.2.12 though, but that's gonna be the last one then
1332365849 Q * bonbons Quit: Leaving
1332368010 Q * ghislain Quit: Leaving.
1332370930 Q * dowdle