1287014531 Q * nou Ping timeout: 480 seconds 1287017778 J * kaemka ~Kaemka@kaemka.vth.sgsnet.se 1287020631 Q * derjohn_mob Read error: Connection reset by peer 1287021350 J * derjohn_mob aj@tmo-059-248.customers.d1-online.com 1287027823 J * infowolfe ~infowolfe@c-67-166-127-67.hsd1.ut.comcast.net 1287034099 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287035226 J * ghislain ~AQUEOS@adsl2.aqueos.com 1287035575 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no 1287036336 Q * ghislain Read error: Connection reset by peer 1287037393 Q * petzsch Quit: Leaving. 1287037631 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287037708 Q * petzsch 1287037899 Q * hijacker Remote host closed the connection 1287038001 J * ghislain ~AQUEOS@adsl2.aqueos.com 1287038083 Q * Piet Ping timeout: 480 seconds 1287038302 Q * ntrs Quit: leaving 1287038627 J * Piet ~Piet__@04ZAAAJZB.tor-irc.dnsbl.oftc.net 1287039323 J * petzsch ~markus@dfn451.rz.tu-ilmenau.de 1287039650 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1287039669 Q * BenG 1287039733 Q * Piet Ping timeout: 480 seconds 1287039779 J * hijacker ~hijacker@213.91.163.5 1287040175 M * hijacker morning fellows 1287040190 M * hijacker Bertl_zZ, let me know when you are around and have a spare minute to test the reiserfs stuff 1287040324 J * Piet ~Piet__@04ZAAAJZ4.tor-irc.dnsbl.oftc.net 1287040594 J * micah_ ~micah@micah.riseup.net 1287040594 Q * micah Remote host closed the connection 1287040738 J * ntrs ~ntrs@vault08.rosehosting.com 1287040780 Q * ntrs 1287041006 J * ntrs ~ntrs@vault08.rosehosting.com 1287043080 J * harobed ~harobed@pda57-1-82-231-115-1.fbx.proxad.net 1287043415 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1287044408 N * Bertl_zZ Bertl 1287044413 M * Bertl morning folks! 1287044520 M * Bertl hijacker: should have time in an hour or two 1287044664 Q * petzsch Quit: Leaving. 1287044775 Q * Romster Quit: Geeks shall inherit properties and methods of object earth. 1287045180 J * kir ~kir@swsoft-msk-nat.sw.ru 1287045671 J * Romster ~romster@202.168.100.149.dynamic.rev.eftel.com 1287045672 Q * kaemka Ping timeout: 480 seconds 1287045971 Q * Romster 1287046248 J * thierryp ~thierry@zankai.inria.fr 1287046705 M * hijacker okay 1287046710 M * hijacker drop me a line when you are around 1287047702 J * barismetin ~barismeti@zanzibar.inria.fr 1287051155 Q * harobed Ping timeout: 480 seconds 1287051580 P * kir Leaving. 1287051735 M * Bertl hijacker: okay, I'm ready now 1287051911 M * hijacker hey, okay 1287051915 M * hijacker i have the reiserfs partiotion 1287051932 M * hijacker what script do you want me run over it / 1287051940 M * hijacker i have 1 guest there as well 1287051947 M * Bertl okay, what you actually need first is an empty partition, could be a loop device as well 1287051964 M * hijacker 200MB will do ? 1287051966 M * Bertl and you want to run the latest testfs.sh with that 1287051972 M * hijacker okay 1287051974 M * hijacker hang on 1287051975 M * Bertl yes, should be more than enougn 1287051978 M * Bertl *enough 1287052022 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/testfs.sh-0.23 1287052149 M * Bertl run it like this: 1287052190 M * hijacker root 1287052191 M * Bertl ./testfs.sh -t reiserfs -t -xyz -D -M 1287052196 M * hijacker ups, 1287052198 M * hijacker okay will do 1287052205 M * Bertl hmm, -F instead of -t IIRC 1287052226 M * Bertl yeah, should be 1287052238 M * Bertl ./testfs.sh -F reiserfs -t -xyz -D -M 1287052263 M * Bertl is your partition (note: it will be formatted) 1287052276 M * Bertl is some free mountpoint 1287052277 M * hijacker okay, i just added an empty partition 1287052279 M * hijacker will format it now 1287052283 M * hijacker and run the script on it 1287052288 M * Bertl no need to format it 1287052299 M * hijacker sorry, i meant mkfs.reiserfs 1287052305 M * Bertl no need to do that 1287052312 M * hijacker alright then 1287052316 M * hijacker just run it over 1287052316 M * Bertl the script will do that and even more 1287052321 M * hijacker cool 1287052346 M * Bertl upload the output somewhere, we'll do another run with -v (and probably -vvv) alter 1287052349 M * Bertl *later 1287052661 J * root__ ~root@213.91.163.5 1287052674 M * root__ http://paste.linux-vserver.org/18180 1287052685 M * hijacker Bertl, the url above is the output 1287052771 M * Bertl so, all xattr related tests seem to succeed there 1287052792 M * Bertl nevertheless, you reported that xattrs are missing after remount, yes? 1287052802 M * hijacker the barrier flag is down 1287052807 M * hijacker we have small b 1287052815 M * hijacker as reported by showattr 1287052836 M * Bertl okay, so let's see if we can improve the testfs.sh tests to check for that 1287052884 M * hijacker okay 1287052889 M * hijacker what can i do? 1287052939 M * Bertl hmm, 131 and 132 should actually check for that 1287052961 M * Bertl let's run -xy with -vvv instead of -t 1287053239 J * Hurga ~foest@business-188-111-094-066.static.arcor-ip.net 1287053269 M * root__ http://paste.linux-vserver.org/18181 1287053331 M * hijacker Bertl, have a look at the url above 1287053356 M * Bertl looks fine to me 1287053368 M * Bertl # 1287053368 M * Bertl do_xattr_verify /mnt/test/ showattr /mnt/test//dir_7096 -+(-)-Bui?([c-])- 1287053369 M * Bertl # verify /mnt/test//dir_7096: -+(-)-Bui?([c-])- = ----Buic- 1287053386 M * Bertl so the barrier definitely survived the unmount/mount 1287053619 M * Bertl so, as the script seems to show correct behaviour, I'd suggest to try to create a similar test case which fails 1287053641 M * hijacker okay 1287053644 M * hijacker just did 1287053647 M * hijacker hang on 1287053653 M * hijacker i will try with mount/umount 1287053655 M * hijacker then with reboot 1287053696 M * hijacker ----buicX vservers/ 1287053696 M * hijacker ----buicX vservers/testvm 1287053727 M * hijacker setattr --barrier vservers/ 1287053734 M * hijacker now i have the barrier 1287053742 M * hijacker showattr vservers/ 1287053742 M * hijacker ----BuicX vservers/ 1287053780 M * hijacker umount and then mount with tag 1287053783 M * hijacker and i have: 1287053791 M * hijacker showattr vservers/ 1287053791 M * hijacker ----BuicX vservers/ 1287053808 M * hijacker ups 1287053812 M * hijacker it is small b 1287053824 M * Bertl what is your umount/mount option? 1287053835 M * Bertl i.e. copy/paste the line here 1287053848 M * hijacker umount vservers/ 1287053858 M * hijacker mount vservers/ 1287053861 M * hijacker and the fstab is 1287053876 M * hijacker UUID=53b2f252-081d-46d0-afdb-fb07836421fc /mnt/vservers reiserfs rw,tag 0 0 1287053886 M * Bertl ah, pebkac 1287053910 M * Bertl please add 'attrs' as mount option for reiserfs 1287053928 M * Bertl (otherwise reiserfs does not support xattrs :) 1287054005 M * hijacker i confirm with attrs it now survives the umount / mount 1287054019 M * Bertl another problem solved ... 1287054025 M * hijacker initially it only required the tag option ? 1287054038 M * hijacker yes, solved;-) 1287054040 M * Bertl nope, since the beginning, attrs is needed 1287054086 M * Bertl tvfs somewhat papers over the fact that reiserfs doesn't support xattrs out of the box 1287054100 M * Bertl i.e. the attributes are cached as long as the inode stays in memory 1287054257 M * hijacker okay 1287054276 M * hijacker do we have that necessary option for reiserfs in the wiki ? 1287054295 M * Bertl no idea 1287054468 M * mnemoc hi, do you know of any recent effort of getting vserver support in libvirt? 1287054513 M * Bertl the status I know off was that it should be broken down to minimal changes and resubmitted 1287054522 M * Bertl if so, they will include it asap 1287054545 M * Bertl (but I doubt daniel_hozac had the time and motivation to do so :) 1287054586 M * hijacker Bertl, i'll try to upload it there for future reference 1287054633 M * Bertl please do so, maybe add it to the FAQ section if not already there 1287054954 M * hijacker will do 1287055419 M * mnemoc Bertl: so daniel's patchset is still valid to the current code base of libvirt? 1287055431 M * Bertl unlikely 1287055469 M * mnemoc :) 1287055481 M * mnemoc i'll try to invest some hours on that this week 1287055489 M * mnemoc +end 1287055489 M * Bertl sounds good! 1287055575 M * Bertl start with a really primitive version and submit that, chances should be good that they'll include it quite fast 1287055588 M * Bertl then start submitting more 1287055600 M * mnemoc incremental... good idea 1287055800 M * mnemoc where's util-vserver repo? 1287055832 M * mnemoc savannah's? 1287055848 M * mnemoc uh, ancient 1287055874 M * mnemoc found 1287056764 M * mnemoc daniel_hozac: is there anything newer that http://www.redhat.com/archives/libvir-list/2008-January/msg00097.html ? or in not-htmlized format? :) 1287057046 M * mnemoc http://people.linux-vserver.org/~dhozac/p/m/libvirt-0.3.3-vserver.patch 1287057239 Q * BenG Quit: I Leave 1287057499 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1287058990 J * harobed ~harobed@pda57-1-82-231-115-1.fbx.proxad.net 1287058993 Q * BenG Quit: I Leave 1287059345 Q * derjohn_mob Ping timeout: 480 seconds 1287059608 J * xch ~chr@pD95EB97F.dip.t-dialin.net 1287060537 N * micah_ micah 1287060763 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287060955 Q * xch Remote host closed the connection 1287061720 Q * pgmateos Quit: Verlassend 1287062031 J * FlashDeluxe 574f5e1c@ircip2.mibbit.com 1287062159 M * FlashDeluxe Hi! Ive got a question, i have a vserver host with three guests and one of the guest has an ip 192.168.11.11 and this ip does exist on the host too, device is eth1:0 how can i tell the guest to use this ip on its own, so that the host doesnt map the eth1:0 interface anymore? 1287062543 J * balbir_ ~balbir@122.172.15.164 1287062909 Q * petzsch Quit: Leaving. 1287062914 Q * root__ Quit: Leaving 1287062942 Q * balbir_ Read error: Connection reset by peer 1287063136 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287063303 J * theocrit1 ~Hubert@kim.theocrite.org 1287063362 Q * theocrite Read error: Connection reset by peer 1287063641 M * FlashDeluxe or can i add an ip adress to the guest, which is only used by the guest? 1287063648 Q * petzsch Quit: Leaving. 1287063694 J * balbir_ ~balbir@122.167.253.84 1287063716 M * FlashDeluxe and if yes, how do i do that? 1287063815 M * mEDI_S http://lkml.org/lkml/2010/10/12/279 1287063817 M * Bertl Linux-VServer uses something called IP isolation, for performance and simplicity 1287063857 M * Bertl with this, there is no way to have an IP exclusively for the guest 1287063871 M * Bertl (i.e. the host will always be able to use it) 1287063963 M * mEDI_S Bertl: hi, i think this artikel is interesting for u ;) a linux scheduler simulator 1287064018 M * Bertl well, not sure why you think so, but yeah 1287064156 M * FlashDeluxe Bertl: I want my guest to be reachable by an database program, but this program tries to connect to the host and not to the guest and thats a problem 1287064185 M * Bertl which probably means that something on the host already bound the IP 1287064224 M * Bertl i.e. by default, a task binding to 0.0.0.0 will 'grab' all IPs available on host or guest 1287064246 M * FlashDeluxe yes, but how can i find out? or how can i give the guest an own ip? the other guest do habe their own ips :/ 1287064252 M * Bertl while that is fine for the guest (where the number of available IPs is already limited), it will 'grab' everything on the host 1287064281 M * Bertl FlashDeluxe: no, guests do not have their 'own' IPs, all IPs are 'shared' with the host 1287064302 Q * ntrs Ping timeout: 480 seconds 1287064353 M * FlashDeluxe Bertl: and how can i get e.g. via ssh to one of my guests? 1287064392 M * Bertl simple, first, as described in the FAQ and other places, restrict your host's sshd to the 'host only' IPs 1287064412 M * Bertl then restart the guests or at least the sshd service inside the guests 1287064434 M * ghislain FlashDeluxe: if you want a guest to have one ip then just remove the ip from the host's config and use the guest interface config to set it on the guest 1287064436 M * Bertl the sshd inside a guest will only be able to bind those IPs assigned to the guest 1287064500 M * FlashDeluxe mhhh ok you`re right, thats quite simple...but i want a database program to get access to a database over an ip and a port, how can i do that? 1287064540 M * Bertl specify that IP and port for the database connection? 1287064551 M * Bertl (not that this would be Linux-VServer related) 1287064626 M * FlashDeluxe i did that, but the database tries to connect with the host and not with the guest and that fails because the port is not open on the host but on the guest 1287064677 M * Bertl you sure about your nomenclature? I mean, why would a 'database' connect to anything? 1287064698 M * Bertl usually an app connects to a database, not the other way round 1287064715 M * Bertl and the database _bind_ certain ports on certain IPs 1287064723 J * ntrs ~ntrs@vault08.rosehosting.com 1287064888 M * FlashDeluxe yes youre right, i try to connect to a database on my guest with a tool on a windows client, but as i said, this fails, although the port is binded with the ip and in listen status, because the tool tries to connect with the host server, but not with the guest 1287064922 M * Bertl okay, so why does the tool try to connect to the host IP instead the guest IP? 1287065026 M * FlashDeluxe because there is a device on the host called eth1:0 which has the same ip as the guests, but if i remove this device from the host, it is also removed from the guest an then the guest has no ip anymore 1287065062 M * FlashDeluxe sorry, if my expression sometimes is not clear 1287065084 M * Bertl how does the tool on the windows machine know about eth1:0 ? 1287065128 M * FlashDeluxe the tool is stupid, that does not know anything, i looked it up myself at the host 1287065204 M * FlashDeluxe perhaps it is clearer if you see it on your own: 1287065205 M * FlashDeluxe http://paste.debian.net/96333/ 1287065513 M * Bertl looks as expected for this kind of setup, still I do not see how that relates to the tool on your windows machine? 1287065758 M * FlashDeluxe ok then i ask an other question :my other guest machines do have an device eth1 with an ip and this device and this ip doesn`t appear at the host, if i execute ifconfig, why? 1287065827 M * Bertl because ifconfig is old and deprecated 1287065843 M * Bertl (and the other guests use secondaries instead of aliases) 1287065857 M * Bertl i.e. check on the host with 'ip a l' 1287065895 M * Bertl you can get the same behaviour for this guest by removing the 'name' entry in the config (which contains '0') 1287065925 M * Bertl (or by simply using secondaries on the host if the IP is not created by util-vserver) 1287065977 M * FlashDeluxe mhhh ok, thanks for the info, i didn´t know that 1287066016 M * Bertl np, common problem for (debian) users 1287067129 M * mnemoc Bertl: what do you think the most primitive vserver support in libvirt should provide? resources? interfaces? 1287067190 M * daniel_hozac patch 01? build/start/stop/delete probably. 1287067943 J * derjohn_mob aj@tmo-085-66.customers.d1-online.com 1287068588 Q * derjohn_mob Ping timeout: 480 seconds 1287069271 J * derjohn_mob aj@88.128.156.243 1287071359 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287071722 Q * tolkor Read error: Connection reset by peer 1287071744 J * tolkor ~rj@tdream.lly.earlham.edu 1287072134 M * ghislain when hashifying do we need twice the disk space ? 1287072151 M * ghislain if i have a disk at 88% full and i hashify the guest 1287072167 M * Bertl no 1287072188 J * bonbons ~bonbons@2001:a18:1:1402:2c0:9fff:fe2d:39d 1287072222 M * ghislain ok thanks bertl, also when i backup my guest i do not need to backup the hashification directory this is transparent for the system it is a normal file 1287072267 M * Bertl no, but you should backup the guests together 1287072290 M * Bertl (btw, the hasification dir is a normal dir with hardlinks to the hashified files) 1287072323 N * Bertl Bertl_oO 1287072329 M * Bertl_oO dinner, bbl 1287072333 M * ghislain ^^ 1287072336 M * ghislain bon appetit 1287072399 M * ghislain i was thinking the other way that the guest had link to the hashification directory but in linux hard links should be "equal" on both side i guess 1287072550 Q * harobed Ping timeout: 480 seconds 1287073205 Q * ncopa Quit: Ex-Chat 1287073601 Q * barismetin Remote host closed the connection 1287073625 Q * derjohn_mob Ping timeout: 480 seconds 1287073723 Q * Hurga Remote host closed the connection 1287073726 J * derjohn_mob aj@88.128.156.243 1287074944 Q * thierryp Remote host closed the connection 1287075131 N * Bertl_oO Bertl 1287075134 M * Bertl back ... 1287076470 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net 1287077176 J * dowdle ~dowdle@scott.coe.montana.edu 1287077295 Q * infowolfe Quit: Leaving 1287077315 J * infowolfe ~infowolfe@c-67-166-127-67.hsd1.ut.comcast.net 1287077425 M * ntrs excellent 1287077460 M * Bertl yes indeed :) 1287080675 Q * derjohn_mob Ping timeout: 480 seconds 1287082909 M * Bertl off to bed now ... have a good one everyone! 1287082946 N * Bertl Bertl_zZ 1287083908 M * petzsch intresting article on cgroups... maybe we wan't to link to it from the wiki? http://www.pro-linux.de/artikel/2/1464/ressourcen-verwaltung-mit-control-groups-cgroups.html 1287084222 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1287084230 Q * BenG 1287084663 M * Marillion petzsch: English is your native language? 1287084734 M * petzsch Marillion: sorry, i didn't write the article, i know it is german but maybe helpful to many 1287084881 M * Marillion i believe that many peaple find the article ;) 1287085383 M * Marillion it is better, some peaple translate the article in French, Spanish and most peaple have something like this :) 1287085412 M * Marillion and English ;) 1287085874 M * Marillion harry: still around? I'm really looking forward to your Patchset :) 1287085909 Q * petzsch Quit: Leaving. 1287086435 J * petzsch ~markus@dslb-088-075-162-254.pools.arcor-ip.net 1287086543 M * harry yeah 1287086553 M * harry which one? 1287087334 Q * ghislain Quit: Leaving. 1287088597 Q * bonbons Quit: Leaving 1287088870 M * Marillion lastest grsec+vserver patch :)) 1287088960 M * Marillion harry: spender said, he change the stable tree 1287089398 M * harry ah, nice ;) 1287089440 M * Marillion yeah, the power of glory *g* 1287089911 J * kaemka ~Kaemka@kaemka.vth.sgsnet.se 1287090396 A * harry will not be able to fix it before the weekend... 1287090402 M * harry too much work on it it seems... 1287090406 M * harry will be for next week 1287090408 M * harry sorry 1287090466 M * Marillion no problem, you are the master 1287090477 M * harry lol... not really 1287090487 M * harry Bertl and spender/pipacs/... are :0 1287090494 M * harry i just use their work for personal glory :p 1287090504 M * Marillion ok, but more than i 1287090508 M * Marillion :) 1287090528 M * harry ;) 1287090573 A * harry off now... cya'll later! 1287090943 Q * Piet Ping timeout: 480 seconds 1287090952 Q * harobed Ping timeout: 480 seconds 1287091543 J * Piet ~Piet__@04ZAAAKO2.tor-irc.dnsbl.oftc.net 1287091643 Q * petzsch Quit: Leaving. 1287093171 A * rootard is scratching his head over unification disk space results 1287093214 M * rootard I started out with 5 vserver filesystems almost completely the same except for about 20MB's of difference between them 1287093272 M * rootard after unification the hash directory is much smaller than excepted (by just over 1GB) and each vserver directory is using 1.2G according to df 1287093285 M * rootard *expected 1287093298 M * daniel_hozac df? 1287093308 M * daniel_hozac df only checks the filesystem, not directories. 1287093318 M * daniel_hozac try du, across /vservers and /vservers/.hash 1287093351 M * rootard I did 1287093371 A * rootard ran: du -hcs hash nodes/* 1287093382 M * rootard and yes, my hash directory is named hash, not .hash 1287093418 M * daniel_hozac and the results? 1287093451 M * rootard http://pastebin.com/vqC8mxB3 1287093570 M * rootard another fun part o this is that when I "find . -type f -links 2 | xargs du -hcs" inside of the hash directory I am only given 7.8M as the total 1287093639 M * daniel_hozac looks like your steps are way off. 1287093647 M * daniel_hozac how did you hashify? 1287093711 M * rootard initially I did: for n in cnode{09,10,11,12}; do vserver $n hashify & echo $n; done 1287093752 M * daniel_hozac and how did you setup unification? because it's obviously not a standard setup by any means. 1287093755 M * rootard this had some wonky results in having identical files with multiple links in the hash directory so I removed everything with just two links and then re-ran hashify serially 1287093893 M * rootard ah, the requirements afaict is that all of the vserver roots + hash directory are on the same filesystem (ext3 in my case) so I modified the symlinks to point to this other location and then made the /etc/vservers/$n/apps/vunify directory 1287093945 M * daniel_hozac which symlinks? 1287093967 Q * ntrs Ping timeout: 480 seconds 1287093980 M * rootard http://pastebin.com/LxkpXL6s 1287094011 M * rootard so the modified links are inside of .default ... everything else should be vanilla 1287094111 M * daniel_hozac have you modified the exclude lists? 1287094121 M * rootard nope 1287094132 M * daniel_hozac so /etc is still excluded then. 1287094139 M * daniel_hozac how big is that in your setup? 1287094194 M * rootard 111M according to du inside of a vserver root directory 1287094284 M * rootard another note is that I had to modify /usr/lib/util-vserver/vpkg to recognize "suse" as a pkg type 1287094296 M * daniel_hozac your guests are SuSE? 1287094302 M * rootard yeah 1287094312 M * daniel_hozac well... 1287094319 M * daniel_hozac then you are definitely in unchartered territory. 1287094324 M * rootard :) 1287094420 J * ntrs ~ntrs@vault08.rosehosting.com 1287094508 M * daniel_hozac so figure out which files aren't being hashified. 1287094578 A * rootard nods 1287094595 M * daniel_hozac find cnode09 -type f -links 1 1287094648 M * rootard yeah ... a whole lot of stuff (~27k files) 1287094803 M * rootard it appears that the majority (in number) is in usr: http://pastebin.com/qAT1D7rB 1287094816 Q * kaemka Read error: Connection reset by peer 1287094823 M * rootard any way to tell hashify to just hashify everything? 1287094992 M * rootard in any case, thanks for helping me get here :) 1287095017 M * daniel_hozac what does rpm -qac list? any of that? 1287095109 M * rootard "rpm -qac | wc -l" returns 1952 ... 1287095123 M * rootard only some of which is in /usr 1287095130 M * daniel_hozac have you tried running hashify with -vvvvv? 1287095150 M * daniel_hozac it'll list the reasons for excluding everything. 1287095163 M * rootard sweet! 1287095597 J * kaemka ~Kaemka@kaemka.vth.sgsnet.se 1287096375 M * rootard it appears that some files have no excuse even with the -vvvvv 1287096713 M * rootard looking closely at the list makes sense. There is a lot of crud in there. Thanks again! 1287097680 M * daniel_hozac files like what? 1287097771 M * rootard cached information in var for eg 1287098022 M * daniel_hozac sure, but what files aren't making sense? 1287098704 M * rootard to me, all of them ... they are all the same 1287098771 M * daniel_hozac what? 1287098927 M * rootard ie: between my 5 vservers on this host, all of them have exactly the same content for those files... 1287098957 M * rootard so I could easily save 5GB by hardlinking them with the iunlink attribute. 1287098987 M * daniel_hozac do you have a vhashify log? 1287099008 M * daniel_hozac a find -ls? 1287099029 M * rootard but I do understand that files used for things like timestamps might be a bad idea to hardlink in this way ... and there may be no way to know programmatically what files are needed in this capacity. 1287099046 M * rootard nah, I don't care about log files but I am not looking at 1.2GB of log differences 1287099116 M * daniel_hozac i meant a log file from running vhashify with debugging enabled. 1287099117 M * rootard however, things like /var/cache for diferent programs makes sense 1287099129 Q * infowolfe Quit: Leaving 1287099134 M * daniel_hozac as well as the output of find -ls 1287099138 M * rootard oh, sorry ... it's getting late in my day 1287099155 J * infowolfe ~infowolfe@c-67-166-127-67.hsd1.ut.comcast.net 1287099405 M * rootard I'll save us both some time, /var /sys /home /dev /mnt /tmp /proc are all excluded. If you really want the full log I can send it. 1287099443 Q * infowolfe 1287099473 M * daniel_hozac that works. 1287099477 M * rootard I'm inclined to call it a day though unless there is some other intrinsic value of the log to the vserver project in general 1287099569 J * infowolfe ~infowolfe@c-67-166-127-67.hsd1.ut.comcast.net