1370217675 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:fdd9:1b87:e7f3:facb
1370218121 J * ircuser-1 ~ircuser-1@35.222-62-69.ftth.swbr.surewest.net
1370218158 Q * thierryp Ping timeout: 480 seconds
1370218453 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370221306 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:242d:bbe8:571c:a9a7
1370221788 Q * thierryp Ping timeout: 480 seconds
1370222058 Q * ccxCZ Ping timeout: 480 seconds
1370222770 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370224938 J * thierryp ~thierry@home.parmentelat.net
1370225424 Q * thierryp Ping timeout: 480 seconds
1370226513 Q * ccxCZ Ping timeout: 480 seconds
1370228569 J * thierryp ~thierry@home.parmentelat.net
1370229053 Q * thierryp Ping timeout: 480 seconds
1370229274 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370231976 Q * ccxCZ Ping timeout: 480 seconds
1370232199 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:f9d4:abea:401d:7798
1370232683 Q * thierryp Ping timeout: 480 seconds
1370233422 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370235831 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:ed51:c7de:10a5:9721
1370236075 Q * ccxCZ Ping timeout: 480 seconds
1370236313 Q * thierryp Ping timeout: 480 seconds
1370236780 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:e892:5e53:10b9:8b28
1370237015 Q * thierryp Remote host closed the connection
1370237048 J * thierryp ~thierry@home.parmentelat.net
1370237531 Q * thierryp Ping timeout: 480 seconds
1370237540 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:30d6:2676:1e35:52ea
1370237634 N * Bertl_zZ Bertl
1370237639 M * Bertl morning folks!
1370237839 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370238398 Q * thierryp Remote host closed the connection
1370239204 A * arekm chasing FATAL: arch/x86/crypto/aesni-intel: sizeof(struct x86cpu_device_id)=16 is not a modulo of the size of section __mod_x86cpu_device_table=24.
1370239207 M * arekm Fix definition of struct x86cpu_device_id in mod_devicetable.h
1370239212 M * arekm on i686, x86_64 builds fine
1370239267 M * Bertl hehe, I've seen a number of warnings during compile tests which should be fixed as well (in mainline code)
1370239454 M * arekm the thing is that it built fine and then I changed two things, updated vserver code to 2.3.6.2 and added irq remapping quirk (http://tiny.pl/hsgb3)
1370239506 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:c52b:67f0:a9a5:58e3
1370239962 J * jeroen ~jeroen@mail.nl02.searchtrends.eu
1370239994 N * jeroen Guest769
1370240028 Q * ccxCZ Ping timeout: 480 seconds
1370240545 J * ccxCZ ~ccxCZ@156.200.broadband11.iol.cz
1370241851 Q * thierryp Remote host closed the connection
1370241871 J * thierryp ~thierry@home.parmentelat.net
1370242113 Q * thierryp Read error: Connection reset by peer
1370242121 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:acbe:b78a:aa62:eff4
1370245351 J * Ghislain ~aqueos@adsl1.aqueos.com
1370245992 Q * Guest769 Quit: Ex-Chat
1370246808 Q * derjohn_mob Ping timeout: 480 seconds
1370248237 Q * thierryp Remote host closed the connection
1370248258 J * thierryp ~thierry@home.parmentelat.net
1370248743 Q * thierryp Ping timeout: 480 seconds
1370249424 M * Bertl off for now ... bbl
1370249431 N * Bertl Bertl_oO
1370250104 J * derjohn_mob ~aj@ip-81-210-224-92.unitymediagroup.de
1370251506 J * _nono_ ~gomes@licencieux.ircam.fr
1370252015 J * thierryp ~thierry@zebra.inria.fr
1370255883 Q * Romster Ping timeout: 480 seconds
1370256862 J * Romster ~romster@202.168.100.149.dynamic.rev.eftel.com
1370257067 J * BenG ~bengreen@host-92-27-135-217.static.as13285.net
1370257805 Q * ircuser-1 Ping timeout: 480 seconds
1370258424 J * alpha_one_x86 ~kvirc@190.186.23.65
1370258439 M * alpha_one_x86 hello
1370258457 M * alpha_one_x86 any idea about my missing cgroup entry for vserver?
1370258977 Q * thierryp Remote host closed the connection
1370259995 J * thierryp ~thierry@zebra.inria.fr
1370260902 J * ensc|w ~ensc@www.sigma-chemnitz.de
1370261301 J * ircuser-1 ~ircuser-1@35.222-62-69.ftth.swbr.surewest.net
1370261306 Q * Aiken Remote host closed the connection
1370262654 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1370264714 Q * BenG Quit: I Leave
1370264985 Q * hijacker_ Quit: Leaving
1370265970 N * l0kit Guest806
1370265977 J * l0kit ~1oxT@0001b54e.user.oftc.net
1370266373 Q * Guest806 Ping timeout: 480 seconds
1370267313 M * Bertl_oO alpha_one_x86: must be caused by some odd config/mount you do in your setup, I verified it here and as you can see, everything is there (util-vserver does the cgroup mount here)
1370267334 M * Bertl_oO http://paste.linux-vserver.org/23833
1370267364 M * alpha_one_x86 Bertl_oO: I have checked, mount and /proc/mounts seam right
1370267390 M * alpha_one_x86 I use gentoo, cgroup is mounted be openrc
1370267390 M * Bertl_oO vserver /dev/cgroup cgroup rw,relatime,hugetlb,blkio,freezer,devices,memory,cpuacct,cpu,cpuset 0 0
1370267404 M * Bertl_oO this is, how the mount looks here (done by util-vserver)
1370267449 M * alpha_one_x86 http://pastebin.com/xdt6dECC -> mounted by gentoo
1370267458 M * Bertl_oO but I'm not saying that this is the reason for the missing entries, I'm just saying, it works fine here, so until we know what causes/triggers this ... I have no idea how to help you
1370267548 M * alpha_one_x86 with gentoo only, I have well the entry
1370267554 M * Bertl_oO note that you probably need to tell util-vserver where the cgroup filesystem is
1370267651 M * Bertl_oO and I suspect, that util-vserver will not handle separate mounts for the various subsystems properly, but you need to contact daniel_hozac or the gentoo maintainer (for util-vserver) for that
1370267680 M * alpha_one_x86 ok, I will contact them
1370267722 M * Bertl_oO off for a nap ... bbl
1370267731 N * Bertl_oO Bertl_zZ
1370267763 M * daniel_hozac you just have to set the per-ss flag for util-vserver
1370267860 M * alpha_one_x86 I presume missing need sysfs
1370267866 M * alpha_one_x86 into gentoo init script
1370267870 M * daniel_hozac huh?
1370267983 M * alpha_one_x86 vserver-utils auto mount the cgroup?
1370268021 M * daniel_hozac yes
1370268042 M * alpha_one_x86 them it's maybe the problem if gentoo do the same too
1370268059 M * daniel_hozac tell util-vserver to use the same ones.
1370268081 M * alpha_one_x86 Then I don't know where come from the problem
1370268120 M * daniel_hozac have you told util-vserver to use them?
1370268253 M * alpha_one_x86 i don't understand
1370268738 Q * kshannon Ping timeout: 480 seconds
1370268750 Q * derjohn_mob Read error: Operation timed out
1370268763 J * derjohn_mob ~aj@ip-81-210-224-92.unitymediagroup.de
1370268783 Q * BlackPanx Read error: Operation timed out
1370268960 J * BlackPanx ~alen@31.15.133.178
1370269302 J * kshannon ~kris@kris.shannon.id.au
1370271509 Q * derjohn_mob Ping timeout: 480 seconds
1370271985 J * derjohn_mob ~aj@ip-81-210-224-92.unitymediagroup.de
1370273124 N * Bertl_zZ Bertl
1370273127 M * Bertl back now ...
1370273861 Q * thierryp Remote host closed the connection
1370274199 M * arekm Bertl: more indicating some change in vserver patch as cause (between .1 and .2)
1370274224 M * Bertl hmm?
1370274229 M * arekm reverted vs patch and kernel built fine
1370274243 M * arekm now going to try .1 again
1370274303 M * Bertl the uidgid strict conversion stuff is still incomplete in .2
1370274324 M * Bertl (so you might hit that, but I guess most people will have that disabled anyway)
1370274984 M * arekm which option disables that?
1370275223 M * Bertl CONFIG_UIDGID_CONVERTED=y
1370275223 M * Bertl CONFIG_UIDGID_STRICT_TYPE_CHECKS=y
1370275246 M * Bertl i.e. setting that to not defined disables it
1370275681 J * bonbons ~bonbons@2001:a18:20a:fa01:2995:4eac:fdbd:1b18
1370276895 Q * derjohn_mob Read error: Operation timed out
1370277092 M * arekm hm, now .1 also failed, makes no sense :(
1370277126 M * Bertl failed how?
1370277402 M * arekm 08:00  * arekm chasing FATAL: arch/x86/crypto/aesni-intel: sizeof(struct x86cpu_device_id)=16 is not a modulo of the size of section  __mod_x86cpu_device_table=24.
1370277406 M * arekm 08:00 < arekm> Fix definition of struct x86cpu_device_id in mod_devicetable.h
1370277548 M * Bertl https://lkml.org/lkml/2012/4/25/480
1370277673 M * Bertl seems there is a problem with modpost and certain tables
1370277703 M * arekm that patch is already in, so something new happens
1370277796 M * Bertl yeah, but very likely it is something related
1370277822 M * Bertl and as we do not even touch the aesni-intel stuff (or anything in that area) I suspect it is still a mainline issue
1370277853 M * Bertl (it might be triggered by the config or accidential reordering or whatever)
1370278132 M * distemper_ may i interrupt? devicetable triggers sth. in my mind... the 3.9-patch wants to create scripts/mod/devicetable-offsets.h, but this file already exits here... so it's maybe sth. that gets generated... 
1370278278 M * distemper_ i also noticed a difference between 3.7 and 3.9... in 3.9 ping isn't working anymore (without NET_RAW caps), but was working in 3.7 out of the box...
1370278299 M * arekm distemper_: thanks, that will be it I guess
1370278385 M * distemper_ did i already say "bertl, thank you for porting to 3.9"? now i did ;)
1370278392 M * arekm Bertl: seems more autogenerated files ended up in diff
1370278467 M * arekm Bertl: arch/um/include/shared/kern_constants.h, arch/um/include/shared/user_constants.h, mentioned scripts/mod/devicetable-offsets.h
1370278528 M * Bertl ah, um again
1370278543 M * Bertl they do not manage to cleanup their junk on 'make mrproper'
1370278555 M * arekm bugreport :-)
1370278573 M * Bertl please file one if you get around
1370278744 M * Bertl actually I already do a mrproper and distclean :)
1370278759 M * Bertl and still um leaves its junk around :)
1370278798 M * Bertl anyway, added those two files for good, they will disappear in the next patch
1370278839 M * arekm and third?
1370279116 M * Bertl I ignored that one :p
1370279148 M * Bertl added now too
1370279513 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1370279594 M * Bertl distemper_: ping is not working? what caps/flags?
1370279606 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:7074:4f53:8cef:28b2
1370279675 A * arekm confirms the ping thing
1370279710 M * arekm socket(PF_INET, SOCK_RAW, IPPROTO_ICMP) = -1 EPERM (Operation not permitted)
1370279710 M * arekm socket(PF_INET, SOCK_RAW, IPPROTO_UDP)  = -1 EPERM (Operation not permitted)
1370279710 M * arekm socket(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6) = 3
1370279710 M * arekm socket(PF_INET6, SOCK_RAW, IPPROTO_UDP) = -1 EPERM (Operation not permitted)
1370279710 M * arekm socket(PF_INET, SOCK_RAW, IPPROTO_ICMP) = -1 EPERM (Operation not permitted)
1370279712 M * arekm write(2, "mtr: unable to get raw sockets.\n", 32mtr: unable to get raw sockets.
1370279731 M * arekm v6 worked but v4 didn't
1370279754 M * Bertl is RAW_ICMP present (ncaps)?
1370279771 M * distemper_ pingand fping (need that for smokeping) used to work without additional caps, now it says "icmp open socket: Operation not permitted" and i need to add NET_RAW to bcapabilities to make it work 
1370279789 M * distemper_ k, arekm was faster :)
1370279819 M * Bertl NET_RAW shouldn't be required, RAW_ICMP should suffice
1370279874 M * Bertl check with 'nattribute --get --nid <guest>
1370279951 M * Bertl hehe, okay, stupid bug (and confirmed)
1370279965 M * distemper_ that says: ncapabilities: raw_icmp nflags: hide_netif,state_admin
1370279979 A * distemper_ is too slow today
1370279999 M * Bertl in net/ipv4/af_inet.c move the 'override:' from line 343
1370280021 M * Bertl to line 347, after the goto out_rcu_unlock
1370280115 M * Bertl distemper_: thanks for reporting!
1370280353 M * distemper_ it's the least i can do ;)
1370280393 M * Bertl and it's very appreciated!
1370280464 Q * thierryp Remote host closed the connection
1370282529 J * imcsk8 ~ichavero@148.229.1.11
1370283230 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:b599:f33b:8846:48e2
1370283694 Q * alpha_one_x86 Quit: KVIrc KVIrc Aria 4.3.1, revision: 6250, sources date: 20120701, built on: 2013-03-26 15:17:55 UTC http://www.kvirc.net/
1370283713 Q * thierryp Ping timeout: 480 seconds
1370284419 J * BenG ~bengreen@cpc35-aztw23-2-0-cust207.18-1.cable.virginmedia.com
1370285379 J * thierryp ~thierry@home.parmentelat.net
1370285497 Q * BenG Quit: I Leave
1370285605 J * BenG ~bengreen@cpc35-aztw23-2-0-cust207.18-1.cable.virginmedia.com
1370285648 Q * BenG 
1370287923 Q * thierryp Remote host closed the connection
1370288673 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f
1370289766 Q * hijacker_ Quit: Leaving
1370290369 J * thierryp ~thierry@home.parmentelat.net
1370291410 Q * thierryp Remote host closed the connection
1370291787 M * Bertl _are__, arekm: updated to vs2.3.6.3 (including the fixes so far)
1370291962 M * _are__ Bertl: thanks, I failed to run our build-environment on 3.9 so far, seems I habve to do it by hand now that I am back from holiday for first tests
1370292111 M * Bertl yeah, I think backporting to 3.8 doesn't really make sense, now that it is already EOL
1370292154 Q * bonbons Quit: Leaving
1370293129 N * _are__ _are_
1370293156 J * alpha_one_x86 ~kvirc@190.186.23.65
1370293198 M * alpha_one_x86 re, cgroup need be mounted before or after vprochide ?
1370293295 M * Bertl IMHO doesn't matter
1370293313 M * Bertl but as daniel_hozac said, you need to tell util-vserver where the subsystems are
1370293346 M * alpha_one_x86 normally gentoo distro do it, how control it?
1370293508 M * Bertl no idea, the documentation (http://www.nongnu.org/util-vserver/doc/conf/configuration.html) doesn't imply that it is possible to handle separate mounts per subsystem
1370293534 M * Bertl but the answer from daniel_hozac somewhat hints that it should be possible (probably with latest util-vserver)
1370293755 M * daniel_hozac touch /etc/vservers/.defaults/cgroup/per-ss
1370293772 M * daniel_hozac i guess the flower page isn't updated, it is in the source.
1370293795 M * alpha_one_x86 and I have well: blkio.reset_stats
1370293801 M * alpha_one_x86 but not the rest
1370293834 M * alpha_one_x86 No such file or directory
1370293845 M * alpha_one_x86 I do the tree
1370293849 M * Bertl yep
1370293880 M * Bertl you probably also want to specify the mount point
1370293960 M * Bertl daniel_hozac: or should the /etc/vservers/.defaults/cgroup/mnt be avoided when something else does the mount?
1370294087 Q * Ghislain Quit: Leaving.
1370294174 M * daniel_hozac no, that should be set to the right place.
1370294193 M * Bertl wasn't clear for me from the flower page comment
1370294206 M * Bertl thanks for the clarifications
1370294436 M * alpha_one_x86 same with this file
1370295894 J * sarshawn ~sarshawn@09GAADEUC.tor-irc.dnsbl.oftc.net
1370296363 P * sarshawn 
1370296383 J * derjohn_mob ~aj@ip-81-210-237-172.unitymediagroup.de
1370296685 M * Bertl alpha_one_x86: did you build the kernel yourself?
1370296714 M * alpha_one_x86 yes
1370296758 M * Bertl could you upload the .config?
1370296955 M * alpha_one_x86 yes...
1370296960 Q * imcsk8 Ping timeout: 480 seconds
1370296998 M * Bertl okay, then please do that, I'll try to recreate the issue with your config
1370297108 M * alpha_one_x86 http://files.first-world.info/temp/config.xz
1370297194 J * BenG ~bengreen@cpc35-aztw23-2-0-cust207.18-1.cable.virginmedia.com
1370297203 J * thierryp ~thierry@2a01:e35:2e2b:e2c0:1c4a:1f99:297d:41fe
1370297263 M * BenG 3.9 kernels now in repo.psand.net
1370297275 M * BenG Debian wheezy only
1370297322 M * BenG I've not tested much, but they pass the basics
1370297334 M * alpha_one_x86 3.9 is too young for me...
1370297357 M * BenG start testing now and we'll have a stable one for when its mature
1370297388 M * BenG though really I don't think 3.9 is the one we'll settle on for the next stable Debian
1370297410 M * alpha_one_x86 I can't, each kernel need be updated one more than 50+ hosts and can have problem at reboot
1370297501 M * BenG how do you get your kernels then? and how do you test them?
1370297592 M * alpha_one_x86 BenG: vanilla + patch, I wait a stable vanilla before go on it, after I do extensive test
1370297672 M * BenG same pattern as me really, but I put out a few for testing along the way
1370297688 Q * thierryp Ping timeout: 480 seconds
1370297741 M * alpha_one_x86 I switch to new kernel only if is very very interresting for something
1370297742 J * imcsk8 ~ichavero@148.229.1.11
1370297998 M * Bertl alpha_one_x86: from looking at the code, the io_wait_time is specific to the cfq io scheduler, and you have all 3 scheduler enabled in the kernel ... could it be that for whatever reason, your guests uses a different io scheduler?
1370298031 M * alpha_one_x86 I don't think
1370298040 M * alpha_one_x86 how check that's?
1370298139 M * alpha_one_x86 cat /sys/block/sda/queue/scheduler -> cfq
1370298162 M * alpha_one_x86 cat /sys/block/md1/queue/scheduler (/, /home, ...): none
1370298186 M * alpha_one_x86 md: /dev/md3 /home ext4 rw,noatime,nodiratime,nobarrier,data=ordered 0 0
1370298550 Q * BenG Quit: I Leave
1370299113 M * Bertl alpha_one_x86: could you try with CONFIG_SCHEDSTATS=y ?
1370299446 M * alpha_one_x86 I try...
1370300897 M * alpha_one_x86 same
1370300938 M * Bertl okay, no idea then, modified config (adjusted to my test hardware) works fine as well
1370300969 M * alpha_one_x86 you have tested with md device ?
1370300981 M * alpha_one_x86 I think it's linked with that's
1370300981 M * Bertl you might want to try to let util-vserver do the mount, just to verify if it is userspace or kernel related
1370301014 M * Bertl the test system is running with lvm over md
1370301042 M * Bertl so IMHO that shouldn't be the problem
1370301069 M * Bertl anyway, I'm off to bed now ... maybe I get another idea tomorrow ...
1370301088 N * Bertl Bertl_zZ
1370301550 Q * sannes Read error: Operation timed out
1370302123 J * sannes ~ace@cm-84.211.93.82.getinternet.no