[00:01] could error in /etc/vsftpd.conf cause a problem? [00:04] clayd: prepend your service commands with 'strace' to see what's happening [00:04] (strace sudo service...) [00:05] ha ha. thanks i was so close [00:05] can someone help me get a second bond device to come up on ubuntu 10.10? [00:05] oh great. i am guessing i need to install strace first? [00:09] pm i think we are onto something :) [00:14] well dang :( [00:19] i hate when good theads in a forum get taken ove by wording flame wars and the answers are never found. /end rant [00:29] does anyone know which kernel includes support for ixgbe intel 82598EB 10GB ethernet? [00:32] can someone help explain symbolic links? so vsftpd is linked to a different file. so when i try and open it to read i get the file it is pointed at. is vsftpd an actual file or just a link? [00:33] NginUS: do you have a pci id for that? [00:34] pmatulis : yes, it's there [00:36] NginUS: where? [00:36] pmatulis: : 01:00.0 & 01:00.1 [00:38] NginUS: that's not it [00:38] NginUS: anyway, are you saying that the current stable release does not support this h/w? [00:39] pmatulis: : There's also a 'DISABLED' label on it in lshw's output. That says the bus info is pci@0000:01:00.0 & pci@0000:01:00.1 [00:40] pmatulis: Apparently. It doesn't come up when configured alone or bonded. [00:43] pmatulis: correction- it's listed when I do modprobe -l ixgbe.... [00:43] ...? [00:44] It's loaded, too... [00:46] how important is dbus? i found a site that says to install it for the vsftpd issue but no explanation why or if it works [00:51] clayd: you can always purge it after if it doesnt help [00:51] ya it didnt help :) === jjohansen is now known as jj-afk [01:21] soren: ping. I remember you saying that lucid as a Xen Dom0 was basically not an option. Is that also the case for lucid as an OpenVZ HN? [01:28] HN? [01:30] "hardware node" [01:30] The openvz equivalent of dom0 [01:31] random forum post suggests the official spiel is that LXC replaces OpenVZ. [01:32] https://wiki.ubuntu.com/ContainersSpec [02:26] twb: I'm fairly certain that if you use openvz's kernel it could. There is no supported Dom0 kernel in Ubuntu. LXC is the new container spec baked into the kernel. It's in Lucid as a "tech preview", but a lot of tools are missing. Several blueprints for UDS-N are proposed to address it [02:27] OK, so basically Ubuntu has dropped *its* support for OpenVZ, suggesting I move to LXC (even though it isn't ready yet). [02:28] the openvz kernel was dropped a while back due to lack of upstream support if I recall correctly [02:28] So my options are to either roll my own OpenVZ kernel, put up with missing features in LXC, move my jails to KVM and take the performance hit, or stick with hardy. [02:28] Hm, OK. [02:28] sounds right [02:30] or, move to esx like me [02:31] I don't really want to migrate to a proprietary solution. [02:31] * Patrickdk wonders what solutions above isn't proprietary [02:31] maybe kvm [02:31] Uh, all of them? [02:32] Virtuozzo is proprietary; OpenVZ isn't. [02:34] esx has container technology? [02:34] zoopster: no, it's more like Xen. AIUI esxi is a hypervisor and a linux-based dom0 [02:35] didn't think so [02:35] http://en.wikipedia.org/wiki/ESXi === zz_sailerboy is now known as sailerboy [04:31] Where's sysctl.conf loaded in lucid? [04:31] Ah, /etc/init/procps.conf [04:56] New bug: #660925 in samba (main) "package samba-common-bin 2:3.4.7~dfsg-1ubuntu3.2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/660925 [05:00] <_2F_> hi, i'm creating a raid1 on my server but the howto speak about grub legacy modification and i have grub2, can someone help me to make it please ? http://www.howtoforge.com/software-raid1-grub-boot-debian-etch-p2 [05:08] What packages are included in "Basic Ubuntu server"? (network install) [05:15] ader10: that probably refers to the ubuntu-standard metapackage [05:16] The UI prompting you is called "tasksel"; you could try grabbing it on an existing system and examining its data files to confirm [05:17] Thanks [05:17] I googled tasksel "basic ubuntu server" but nothing turned up [05:26] just upgraded to 10.10, suspend no longer works on a dell latitude d620: PM: Device 00:0a failed to suspend: error -5 [05:26] Any ideas? [05:41] I said "grab it" not "google it" [05:42] dasunsrule32: ACPI suspend? Are you running "pm-suspend"? If not, what are you running? [05:44] twb: not sure just upgraded [05:44] twb: pm-suspend is installed [05:45] dasunsrule32: I mean, how are you initiating the suspend process? [05:45] closing the lid [05:45] You're running Ubuntu Server on your laptop? [05:45] no [05:45] <_2F_> hi, i'm creating a raid1 on my server but the howto speak about grub legacy modification and i have grub2, can someone help me to make it please ? http://www.howtoforge.com/software-raid1-grub-boot-debian-etch-p2 [05:45] Then you should ask #ubuntu. [05:45] twb: desktop, i have been trying to get help there for 25 minutes [05:46] :( [05:46] dasunsrule32: it is not appropriate to ask here just because you can't get help there. Sorry. [05:46] ok, lata [05:46] thx [05:50] come join us at #hackers or ##itt-tech . if you dare that is..... [05:53] <_2F_> ... === ivoks-afk is now known as ivoks === ivoks is now known as ivoks-afk === sailerboy is now known as zz_sailerboy [08:07] twb: Probably even more so. [08:29] soren: sorry, I lost my scrollback. What are you replying to? [08:30] 00:21 < twb> soren: ping. I remember you saying that lucid as a Xen Dom0 was basically not an option. Is that also the case for lucid as an OpenVZ HN? [08:30] I'm assuming HN is the equivalent of a dom0. [08:30] soren: yep, and OK. [08:30] Further down you can see me working it out on my own [08:30] twb: Lucid does LXC just fine, though. [08:31] twb: Ah, I just looked at stuff that mentioned my name. [08:31] I rarely actually catch up on IRC in the morning (I've got 135 windows open right now... I woulnd't be done catching up until late in the afternoon), but I do read stuff that irssi has highlighted overnight. [08:32] No worries [08:32] I only pestered you specifically because I know you have a clue, and I know you were talking about Xen and UEC [08:32] Sure, np. [08:39] Basically my position is that out of KVM and LXC, it's safer for me to pick KVM because AFAICT nobody is really using LXC yet [08:39] At least as far as paying customers go -- my own scratch boxes I'm happy to try new things on [08:53] soren: are you gonna give me the LXC salespitch? :-) AFAICT nobody is using it yet. [08:54] Sorry, remember those 135 windows, I mentioned? They basically mean that I don't notice stuff unless you mention my name :) [08:55] Hence my long response time :) [08:55] So, out of kvm and lxc, I'm also much more in the kvm camp. [08:55] They serve different purposes, though. [08:55] Just like KVM and OpenVZ serve different purposes. [08:56] ...and if you're looking to serve the purposes that OpenVZ does, and you can't use OpenVZ for whatever reason, LXC is what you want to look at. [08:56] twb: ^ [08:57] soren: hence the "soren:" at the end :-) [08:57] twb: :) [08:58] soren: and yeah, eventually I'd like to use LXC for lts-in-lts, and kvm only for random stupid things like windows and sco VMs [08:58] But sure, if you have the hardware for KVM and your workload is suitable for KVM, *definitely* go with KVM. [08:58] soren: fortunately for this customer I'm in a position to dictate VT-x CPUs. [08:58] I deal with software that enables random people to run random OS's. A container is of very, very little use to me. [08:59] Both in terms of utility, but also in terms of security. [08:59] soren: for me, I want containers so that webdev idiots can each have their happy little world of specific versions of PHP or whatever inside their jail, and when their crappy PHP code gets compromised, it still doesn't affect me or the other idiot webdevs [08:59] If I'm just running my own stuff, it's probably fine, but I really don't feel comfortable having random people running stuff in my kernel. [09:05] team: great job on triaging new/undecided server bugs: http://webnumbr.com/ubuntu-server-triage [09:09] ttx: neat! [10:26] hello folgs [10:26] folks ^:= [10:32] .... . .-.. .-.. --- === macbookpro is now known as RainKing [11:45] do you guys run two instances of openssh on your servers (password auth for local network and key auth for public network)? [11:47] or does the openssh-server support some kind of rules with authentication types (connections from gw/SNAT would require key, others key or passwd)? [11:47] reisi: I dont mix these zones on one server. [11:48] why not just disable password auth all togeather. Keys only :) [11:48] well, the possibility of losing the keyfile :) [11:50] 1 day of security training, encrypted disk on workstations, and company security policy that forbids loosing key files ;) [11:51] heh === doko_ is now known as doko [12:25] where can I find some documentation on setting up Ubuntu Server 10.04 on a NetApp FAS2040 possibly with vmware in between? [12:27] matt605: What does that have to do with Ubuntu? [12:27] I guess you should be looking for vmware/netapp documentation [12:29] I'm not sure.. should I connect the NAS directly, as iSCSI device to the Ubuntu boxes, should I use NFS, or what's the Best Practice? [12:31] I haven't found any Ubuntu specific docs on netapp/vmware side unfortunately [12:31] matt605: That's not up to Ubuntu. You should look for bestpractises for Vmware ESX and Netapp [12:31] which will probably tell you to setup NFS between ESX and the Netapp [12:33] ok that's what I read till now, thanks Jeeves [12:34] I need to share some storage between the Ubuntu Servers and I think NFS will do it [12:41] Yes, nfs is about your only option [12:41] iscsi is point2point [12:42] hallyn: commented on your BP proposals, ping me when around [12:47] smoser: around ? [12:48] Jeeves_: wouldn't it be somehow possible to mount a volume of an ubuntu server on another one? [12:49] <_ruben> sounds like what nfs would do [12:49] ah yes;-) [12:52] I was thinking about something like OCFS [12:55] Daviey, zul: someone should be filing the 5 UEC blueprints. I can do it, if you prefer. [12:57] sommer: around ? [13:01] Hi, when I boot Ubuntu Server, I get "/dev/mapper/servername-root: contains a file system with errors, check forced." which causes it to scan. At 70% I get "/dev/mapper/servername-root: Duplicate or bad block in use!" and then it just stops. [13:01] Is there any way to perform a scan before loading certain entries? [13:02] Thelz: It's probably asking a question, which you can't see, because some smartasses thought plymouth would be cool. [13:02] [13:03] ? [13:03] ttx: wilco [13:29] Hey guys, I was just wondering if anyone would be able to help me get my webserver setup? [13:29] ttx: i can do it [13:29] zul: talk to Daviey [13:29] ok [13:29] ttx: yo [13:29] sommer: do you plan to file a Server doc blueprint, so that we can schedule a session at UDS ? [13:30] ttx: ya, I was going to work on that today [13:30] something like ubuntutheproject-server-n-documentation ? [13:30] sommer: cool [13:30] hey davi [13:30] grrr..Daviey [13:43] zul: fingers not cooperating this morning? [13:44] zoopster: no not until after a couple of cans of coke :) [13:44] zul: know what you mean! [13:44] zoopster: thank god there is a cvs down the road from the hotel [13:45] zul: o/ [13:46] New bug: #661197 in tomcat6 (main) "Sync tomcat6 6.0.28-7 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/661197 [13:46] zoopster: i will have to break my rule from shopping any store that has an outdated VCS as a name while there ;) [13:47] zul: there is not much around that resort...but there is a gas station across the street [13:47] zul: and I'll have my massive suv available for a taxi [13:47] zoopster: muhahaha [13:50] Daviey: are you going to cry if I requestsync groovy from 1.7.0 to 1.7.4 ? [13:50] ttx: around, but haven't read all email yet [13:50] ttx: we are early enough in the cycle to be able to fix it :) [13:50] hallyn: ping me when you have read it [13:50] so go for it [13:51] * ttx needs to collect the syncs from all that push-patches-to-debian work he did [13:51] Daviey: ok, I'll keep that log for reference, when you start cursing. [13:51] lol [13:52] * ttx checks for potential new bd first === nIMBVS is now known as nIMBVS_ [13:53] ew [13:54] hey guys :D [13:54] * zul remains blissfully ignorant about groovy [13:55] can anyone tell me if an image from VBox is compatible with FVM ?? === sailerboy is now known as zz_sailerboy [13:58] Daviey: so that groovy sync is blocked on two new source packages: jansi, jansi-native. That calls for three MIRs: jansi, jansi-native and hawtjni [13:58] ttx: What does .4 bring? [13:58] Do we need it, do we want it? [13:59] ewww [13:59] Daviey: I'm pretty sure you and me don't need it. But groovy programmers, probably [13:59] * ttx looks up changelog [13:59] ttx: ack [14:00] zul: paying attention? :) [14:00] Daviey: umm no :) [14:01] Daviey: I'll file a placeholder bug, hoping it will prevent someone from just blindly asking for the sync [14:01] ttx: you think that will help?! [14:02] At least it documents what it takes :) [14:02] couldnt you put it on the sync blacklist [14:02] * ttx sobs, looking at changelogs not updated since 0.5 [14:03] Google please help me [14:04] :( [14:04] they have "release notes for 1.7 [14:07] Daviey: http://jira.codehaus.org/secure/ConfigureReleaseNote.jspa?projectId=10242&version=16563 [14:07] I found the mysterious changelog generator ^ [14:08] all those minor versions include a mix of bugfixes, improvements and new features [14:08] and in great Java tradition, probably API changes. [14:14] ttx: \o/ [14:15] ttx: You weren't kidding... A changelog generator! Damn, we need one of them. [14:15] I noticed a new kind of syslog deamon in 10.04. how can I best migrate my old configuration? [14:17] Daviey: i'm dead serious when I do Java merge/syncs. [14:17] ttx: Lets hope it doesn't bring us fun. [14:20] ttx: o/ [14:20] Daviey: hi! [14:20] hey mathiaz [14:20] Daviey: could you attend the community call with jcastro and kim0 in 40 minutes? [14:20] mathiaz: o/ [14:21] mathiaz: Sounds good. [14:21] ttx: I've registered a bunch of BP for UDS [14:21] mathiaz: Just looking through the specs.. it seems we have some overlap... you've raised some UEC specs? [14:21] ttx: what do you think about the plan? have you noticed them? [14:21] mathiaz: cool ! Did you mention them on the IdeaPool yet ? (remove the lines from the brainstorm area that is covered by an already-filed spec) [14:22] Daviey: yes - most of my specs are actually related to deploying UEC on physical hardware [14:22] ttx: I'll update the IdeaPool page [14:22] the idea is to keep in the pool things that are not part of an existing blueprint, to see how much we could still file [14:22] mathiaz: thanks ! [14:22] mathiaz: I currently have a lock on the IdeasPool page [14:23] ttx: I've set the BP approver to you [14:24] mathiaz: rigth, approval is not very important for that step [14:24] mathiaz: what's important is that theuy show up as proposed for uds-n [14:24] ttx: ok - I'll double check that [14:24] and getting accepted for discussion (which can be done by any member of the UDS organizer group) [14:27] ttx: ok - all 5 BP are proposed for UDS-N [14:28] Daviey: no longer editing NattyIdeaPool? [14:28] mathiaz: no [14:28] mathiaz: we need to have a call, after this one. [14:29] ttx: (replied to email) [14:30] Daviey: sure [14:30] Daviey: I think there is some overlap around the installation server [14:30] Daviey: cloud-server-n-uec-deployment-installer [14:31] Daviey: cloud-server-n-install-service [14:32] Daviey: https://bugs.launchpad.net/ubuntu/+source/groovy/+bug/661230 [14:32] New bug: #661225 in openldap (main) "package libldap-2.4-2 2.4.21-0ubuntu5.2 failed to install/upgrade: cannot access archive: No such file or directory" [Undecided,New] https://launchpad.net/bugs/661225 [14:32] Launchpad bug 661230 in groovy "[BLOCKED] Sync groovy 1.7.4-1 (main) from Debian unstable (main)" [Wishlist,Incomplete] [14:32] mathiaz: i said that earlier.. this is why i want to have the call [14:33] mathiaz: Just looking through the specs.. it seems we have some overlap... you've raised some UEC specs? [14:33] Daviey: great - seems that we have the same ideas here [14:34] Daviey: I've just registered 5 BP to cover all the topics you've mentioned in cloud-server-n-uec-deployment-installer BP [14:34] Daviey: I'd rather have small *targeted* sessions to flush out the plan [14:34] mathiaz: slow down sailor! [14:34] ttx, here now. [14:35] mathiaz: lets flesh it out in a call. [14:37] smoser: mumble ? [14:37] ttx, can we schedule something for 3 hours from now ? [14:37] i'm off today (swap'd from monday) [14:37] and am trying to leave [14:38] is it possible to get this bug fixed in 10.04?: https://issues.apache.org/bugzilla/show_bug.cgi?id=48981 [14:38] issues.apache.org bug 48981 in mod_fcgid "FcgidMaxProcesses is not honoured" [Normal,Resolved: fixed] [14:38] Snadder: is there a bug in launchpad for it? [14:39] zul, I have no idea, but the fcgid version in 10.04 is very old. [14:39] Snadder: please open up a bug in launchpad irc is not good for such requests [14:39] open a ppa for it? :) [14:41] New bug: #661230 in groovy (main) "[BLOCKED] Sync groovy 1.7.4-1 (main) from Debian unstable (main)" [Wishlist,Incomplete] https://launchpad.net/bugs/661230 [14:58] smoser: i'll send an email, we can talk on Monday. [15:00] jcastro: kim0: o/ [15:01] mumble? [15:01] mathiaz: o/ [15:01] hey [15:01] Daviey: I guess so [15:01] hi [15:01] which channel? [15:01] Daviey: mumble [15:01] Daviey: mathiaz come join our community channel [15:04] jcastro: Sure? [15:27] ttx: so you're saying call it hardware-server-n-unprivileged-mounts ? [15:27] ttx: or hardware-kernel-n-unprivileged-mounts? === ivoks-afk is now known as ivoks [15:34] hallyn: would that end up being a kernel or a server team work ? [15:34] hallyn: maybe ask jjohansen for advice [15:35] ttx: ok will do, thx [15:38] seems i need to put LVM on top of my RAID1 (/boot) array in order for grub-install to work. is that normal? [15:41] <_ruben> pmatulis: i tend to do the opposite: keep /boot *out* of lvm and on plain raid [15:41] _ruben: yeah, i would like that but that's the only way it works over here (lucid) [15:42] <_ruben> pmatulis: are you trying to install grub on the md device or on the underlying disks? [15:42] <_ruben> you should do the latter [15:42] ttx: hey [15:42] ttx: wanna mumble about the BP? [15:42] _ruben: and even when it does, grub will only install to the disks (/dev/sda /dev/sdb) [15:42] <_ruben> (for the bootloader stuff that is, the actual /boot contents should be on the md) [15:43] I can do that, just a sec [15:43] _ruben: but with just RAID, *nothing* works [15:43] <_ruben> pmatulis: odd, dont recall having done lucid installs on md raid (most my lucid boxes are upgrades) [15:44] <_ruben> could be a regression or grub2 issue [15:44] _ruben: do you know how i can get the partman and syslog files off of the disks for analysis/bug_report? [15:44] _ruben: i'm using KVM btw [15:45] <_ruben> pmatulis: i always do network (pxe+local mirror) installs, so my install env has sshd running in it [15:45] <_ruben> software raid in a vm? odd [15:45] SpamapS: if you file a bash completion spec, maybe we should lump bug 508606 in it [15:45] Launchpad bug 508606 in command-not-found "Recommend tracepath for "traceroute" and "tracert"" [Wishlist,In progress] https://launchpad.net/bugs/508606 [15:45] _ruben: it's for testing [15:45] i.e. command-not-found improvements [15:46] mathiaz: joining [15:46] <_ruben> pmatulis: never messed with kvm myself, so dunno if it'd interfere in any way, i did similar tests on esxi (for documentation purposes), which did work (then again, it was'nt on lucid) [15:47] kirkland: kees: one thing i might want to discuss at uds is the popups for passwords. I seem to get quite a few of them these days (for network, for ubuntu-one, for ssh, etc) and the messages don't seem to always be helpful. just got one now and had no idea for waht (so said no). maybe we can introduce somethign more systematic to make sure user knows who's asking? [15:48] (i added kirkland to that ping bc i was thinking of ecryptfs, though i guess it doesn't even do that :) [16:01] hallyn: in a browser, or on your desktop, or what? [16:02] kirkland: desktop [16:03] hallyn: yeah, i don't like it when i'm prompted for a password and i'm not sure by what either [16:04] hallyn: truth be told, it's usually a debuild i have in the background, that just finished, and gpg is trying to sign the package [16:04] hallyn: but for a long build, it gets me every time [16:07] kirkland: worth a blueprint? (though maybe kees has already discussed this at a past uds) [16:07] hallyn: touch base with kees first [16:08] hallyn: i'd defer to his lead on this [16:09] hallyn: inevitably, someone is going to troll that session for #194472 [16:09] hallyn: go read that bug, when you have a spare 2 hours [16:12] kirkland: heh, ok [16:12] kirkland: put on my lazy-day-toread list [16:12] I was just wondering, how would I go about upgrading from Ubuntu 9.04 to Ubuntu 9.10, I can't go to any version higher than 9.10 though as Xen requires a very specific Kernal version. [16:12] hallyn: ;-) [16:12] hallyn: it's a classic [16:13] SirMeaky: sudo do-release-upgrade [16:13] Scottk: So that wont go any higher than 9.10? [16:14] SirMeaky: Not from 9.04. [16:14] Scotk: Awesome, thanks :) [16:15] Scottk: do-released-upgrade not found apparently [16:15] release* sorry [16:15] SirMeaky: IIRC you'll want sudo apt-get install update-manager-core [16:16] Scottk: Okay then :) [16:17] Scottk: That worked great, thanks :) [16:17] SirMeaky: You're welcome. [16:23] i'd need to run rsync via cron, but i'am unable to setup password-less connection over ssh, what are the alternatives (for dummies)? [16:27] cough [16:29] does anybody know what gzip-rsyncable program is? [16:29] I didn't find it anywhere in the repository [16:35] JamesPage: sorry for the duplicate :) Caught it in my merges review [16:38] nevermind. I found gzip-rsyncable [16:46] brontoeee: Why are you unable to set up a passwordless ssh connection? [16:47] You'll need some way to authenticate the session. [16:54] hello [16:54] i'm using an ubuntu server to export the home directory to several other machines over NFS [16:55] I also mount this nfs home directory on other ubuntu servers and there something sucks [16:55] all files have got the same gid and uid [16:55] -rwx------ 1 4294967294 4294967294 7842 2010-09-17 20:52 OPf Zusammenfassung.odt [16:57] ttx: yeah I think those two would be excellent together [16:57] ttx: bash completion and enhancements to command-not-found [16:57] SpamapS: great ! [16:57] TREllis SpamapS ttx kirkland hggdh hallyn jiboumans http://www.roaksoax.com/2010/10/high-availability-uec-clc-howto === ivoks is now known as ivoks-afk [16:59] ttx: is there already a blueprint for it, or should I make one? [17:00] no there is no blueprint on it yet [17:00] you can file one [17:01] RoAkSoAx, wow, this was faster than I expected :-) [17:01] RoAkSoAx, I will read it later, got to finish preparation for a class now. THANK YOU, BTW [17:01] hggdh: there's still testing to be done but overall should be working ;) [17:03] ttx: this feels like one for other .. [17:05] RoAkSoAx: nice [17:05] RoAkSoAx: did you test failover with active instances? [17:06] hi all [17:07] can somone helpout a newbie [17:07] if they have time ofcourse [17:08] installed ubuntuserver with lamp and ssh, i always use root acount for everything i do here, installed phpmyadmin from terminal, created databses and created another root acount by duplicating with host % so i cna acces phpmyadmin from the outside,anged created a user acount to check username and password for website and gave the database priv only. Next i edited my apache config to disable indexing of my www and changed php.ini m [17:09] SpamapS: yes [17:10] TREllis: not yet... but i don't see why they should fail to run [17:10] actually it could really be packageselection [17:10] which is probably less full than other.. no? === jj-afk is now known as jjohansen [17:11] kroimpa: keep it to a couple of lines at one time [17:11] kroimpa: you cut out at 'changed php.ini' [17:11] ok [17:12] installed ubuntuserver with lamp and ssh, i always use root acount for everything i do here, installed phpmyadmin from terminal, created databses and created another root acount by duplicating with host % so i cna acces phpmyadmin from the outside,anged created a user acount to check username and password for website and gave the database priv only. Next i edited my apache config to disable indexing of my www and changed php.ini [17:12] maxpostsize and max upload size to 6G to allow upload of movie files. i did exactly this nothing more nothing less, never touched anything else on the server. After a reboot i CANT: access phpmyadmin with any acount (it doesnt even give errors like username or password not correct), i cant login on my server. I CAN: acces mysql with root acount(same acount i use on phpmyadmin). tried allot of reinsstals but after a reboot it is alwa [17:13] kroimpa: can't access means.. ??? [17:13] kroimpa: what did you try to do, what did you expect, what did you get? [17:13] after the reboot [17:13] i tried to login to phpmyadmin with root acount and it didnt work [17:13] and i cant loggin to my website anymore [17:14] RoAkSoAx: nice :-) [17:14] but the tables are intact and pici said there were no errors in my mysql file [17:14] RoAkSoAx: i'm going to read it in detail after some meetings [17:15] so the error log in mysql shows nothing, but the apache error log is 300+ mbs big [17:15] XD [17:15] kroimpa: So, what did you get, an error message? A dancing banana picture? [17:16] kroimpa: please understand, its very hard to understand or help with a statement like "I can't" .. we need details. [17:16] i know im sorry, my first week at linuxxing [17:16] :) [17:17] when i try to login with root on phpmyadmin it does nothing just return to the inog page, no error message nothing [17:17] when i try to login on my site using usrname and passwd i just get redirected back to the inlog page [17:18] but the mysql tables and user settings are still the same [17:18] as before the reboot [17:19] so somehow after the reboot, i just acces to phpmyadmin and access to my mysql tables while the user settings are the same as before the reboot [17:20] *i cant get acces [17:21] so the question is what couldv chaned after a reboot? [17:21] before everything ws working fine [17:25] i tied reinstalling like 5 times, but everytime after the first reboot i lose access to phpmyadmin with root acount and i cant login on my site [17:27] should i write in detail how i installed and configured my server? [17:29] how can i see IP address of connected users (ssh)? [17:32] consumerism, netstat -n | grep :22 [17:32] ttx: are you "blessed" with the powers to approve blueprint nominations to UDS-N? [17:34] qman__: thanks, also found a different answer in another channel: 'who' [17:35] ah, didn't realize who showed the connecting address [17:47] all: have a good weekend! [17:53] hi would anyone please take a look at my post :)? [17:53] http://ubuntuforums.org/showthread.php?t=1597577 [17:53] if they have time ofcourse :) [17:57] Hello hello. I have a question about group permissions. I wasn't able to find the right thing on the net (although I'm new so I might have missed it). Right now I set up 2 users and 1 group in my samba file server. I set up a sub-folder for one of the groups. Whenever user 1 creates or modifies a file, I get access denied eventhough we are part of the same group. How would I as user 2 see the file that user 1 saved in m [17:57] thanks btw [17:59] oh... wait... should create mask be 775 vs 770? [18:01] 775 does allow other to write and change the data [18:01] i think your 770 is correct [18:01] ohhh ok.. right that's what I thought because I don't want guests to see it.... [18:01] indeed :) [18:02] im new to ubuntu to lol strugling with it [18:02] :P [18:02] hmmm I am going wrong somewhere [18:02] kroimpa: tell me about it! so far I've been in here oer the past 3 days as i strugle through samba [18:02] hahahaha [18:02] same here man [18:03] started ubuntu a week ago [18:03] nothing but sweat and tears [18:03] :P [18:04] i always try to fint the logic behind problems [18:04] it gets better :) [18:04] find out this week that som probs arent logic [18:04] *found [18:04] :P [18:04] yes the logic helps. I found some knowledge of programming helps pick it up fast. I made a lot of progress over 3 days [18:04] :D [18:04] made a nice php website [18:04] but after i reboot my server [18:05] no access :( [18:05] rstinalled like 5 times [18:05] and evertime after the first reboot i lose access [18:06] reallyy anyoning:P [18:06] so what u using your server for? [18:07] I'm making a samba file server for the company I'm in [18:07] our IT guy is mental [18:08] Is anyone here using NFS4 for /home directories for clients? [18:08] hahaha ok:D [18:10] Was hoping someone who did may have tuning advice. === ivoks-afk is now known as ivoks [18:12] New bug: #661350 in samba (main) "package smbclient 2:3.5.4~dfsg-1ubuntu3 failed to install/upgrade: corrupted filesystem tarfile - corrupted package archive" [Undecided,New] https://launchpad.net/bugs/661350 [18:12] I knwo it's lunch time but is anyone around that might be able to help with my issue stated about about group and user permissions? [18:13] :) [18:14] is there a command/utility to blank gnome, but still enable certain players to do their stuff, like mplayer [18:30] -rwx------ 1 4294967294 4294967294 7842 2010-09-17 20:52 OPf Zusammenfassung.odt [18:30] that's what I get when I mount a NFS from my ubuntu server [18:30] Anyone here? [18:30] all files have got this uid gid [18:31] I've got a question in regards to how the ubuntu official repositorys are setup. What software is in use for management of the repository? [18:31] Just dpkg-dev, or reprepro, or some sort of home brew software? [18:31] v [18:31] ssh says " WARNING: UNPROTECTED PRIVATE KEY FILE! ", but I can't change the permissions because of the fat file system. How do I tell ssh to ignore private key permissions ? [18:47] anyone know how to upload files bigger than 5 GB, because when i change it to 6GB in my php.ini file, i get this error: [18:47] [Fri Oct 15 19:09:40 2010] [error] [client 192.168.1.1] PHP Warning: Unknown: POST Content-Length of 51 bytes exceeds the limit of -2147483648 bytes in Unknown [18:50] looks like the limit setting overflowed [18:50] it probably has a cap below that [18:51] yeah the limit is 5g, cos when i change it 6g all the POST actions fail [18:51] but i like to know how to change the unlying limit [18:51] *underlying [18:51] kinda strange though [18:51] owww [18:51] srry [18:51] didnt read ya second post [18:52] il look in the php.ini file [18:52] the flag is probably a 32-bit number [18:52] and therefore can't go any higher than that [18:52] aaah that must be it [18:52] you'd have to modify the source [18:53] so i have to edit the source code of ubuntu? [18:53] i mean apache2 [18:53] or should i look in the php.ini [18:53] if there might be another limit defined [18:54] no, that limit cannot go higher without modifying the source code of PHP [18:54] and compiling it [18:54] damn:P [18:54] im kinda new [18:54] just using ubuntu for a week [18:54] so ill pass :P [18:55] but thanks allot for the answer:D [18:55] that method really isn't designed to go larger [18:55] if you need to upload larger files there is probably a better approach [18:55] so in the php field there are no options for uploading large 1080p movies [18:55] yeah im using just a standard POST form [18:56] so i had to edit the max_post_size limit too [18:57] is there another way to upload files using php that doesnt use the html POST form? [18:58] cos the POST limit is the problem here i think [18:59] not that I'm aware of [18:59] the only way to upload things over http, and pass it to php is using post [19:00] aah damn [19:00] what I meant was, you'll probably have to implement something in java or flash or python or some other approach to manage it [19:00] ahh ok il look into that [19:00] php was designed to render web pages, and though it extends quite nicely, 1080p video was never in the road map [19:00] hahaha i figured:P [19:00] are you using x86 or x64? [19:01] x86 on an old p3:P [19:01] that might be the issue [19:01] I wonder if it would *just work* on a x64 system [19:01] mmm [19:01] I have only ever uploaded 300megs with php though [19:01] yeah, 64-bit versions might have a higher limit [19:01] il set up virtual machine on my laptop installing Ubnut server x64 [19:02] and ill try out [19:02] but is its a 32bit limit like u say [19:02] *if [19:02] shouldnt the 5gb limit also not work [19:02] cos 32bit is limited to 4 gigs right> [19:02] the question is if they typecasted it to 32bit, and it won't work on 64bit also, or will it? :) [19:02] well, the registers are [19:03] ow damn ur right [19:03] nothing like uploading a dvd iso to my cellphone over ssh [19:04] LOL [19:04] 20min till complete :( [19:04] u can do that [19:04] WOW [19:04] easily [19:04] and u can mount it too? [19:04] I dunno [19:04] what would be so awsome [19:04] *that [19:04] I just want to take it with me to load it up at the server [19:04] :D [19:05] u have a movie sharing website too? [19:05] im building one for me and my friends [19:05] nope [19:06] just new ubuntu dvd, just updating them while I have a chance [19:06] owwww the new 10.10 version [19:07] is it wise to autoupdate my server?> [19:07] or just doing in manually [19:07] it really depends on you :) [19:07] in my case, I would never trust autoupdate [19:07] but I very rarely see an issue where update messes something up [19:08] hahaha ok [19:08] my redhat systems are screwed after every update, I hate it :( [19:08] :( [19:08] that sux man [19:08] things like that arent easy fixable [19:08] patdk-wk: re redhat updates, are you serious or joking? [19:08] well, it's easy to fix, but it's just a pain [19:08] serious, rhel5 [19:09] hahahaha [19:09] I have grown to love ubuntu, cause the updates don't screw up the server :) [19:09] though, I'm still not trusting enough to go autoupdate :) [19:09] i cant judge im just new to linux and ubuntu [19:09] started a week ago [19:09] but disabled auto upadtes [19:10] I started on slackware 2.x I think, using a 1.2 kernel [19:10] windows update scared me to enable it [19:10] I rarely have an issue with windows updates [19:10] mainly only xp sp3 had issues [19:10] 1 word, genuine [19:10] ;) :P [19:10] well, that is your own fault [19:10] it is [19:11] i do blame myself [19:11] i have 2 w7 home keys lying around [19:11] but i just have to install an illegal w7 ultimate [19:11] call it stupid [19:11] cos that what it is [19:12] but hey! [19:12] we all have things like that! [19:12] how can i see a list of modules that are currently loaded for iptables? [19:13] is there an webplayer other than divxwebplayer that can play mkv in HD res? [19:13] i guess most playters wont do [19:14] EvilPhoenix: http://tinyurl.com/39fcndu [19:15] lol pma [19:16] so what u guys doing tonight? [19:16] weekend started here in the ntherlands [19:16] gonna grab som beers with my mates [19:17] kroimpa: please keep to the topic of this channel [19:17] hey guys, anyone here familiar enough with BTRFS to field a question on it? [19:18] maybe [19:18] ha [19:18] ah ok guntbert [19:18] * patdk-wk looks around for the topic, I think it ran away [19:18] okay, would it be possible to use BTRFS as the storage solution for a MySQL database, duplicating all writes to EXT4 on another drive? [19:19] since BTRFS doesn't have an fsck utility, it would be nice to keep a "safe" copy of data, but use BTRFS for it's snapshot support, backups, etc [19:20] heh? [19:20] * patdk-wk wonders how you would duplicate it on write [19:20] 1s, finding link [19:20] and why btrfs itself isn't safe enough [19:20] well, assuming it's stable [19:20] The B-tree file system is a “copy-on-write (COW)” check-summed file system with multi-device support and writable snapshots. [19:20] you don't understand what that means :) [19:20] that means, when you write to the drive [19:20] if you write to the same file, in the same place [19:21] ext2/3/4 writes to the same place on the drive [19:21] btrfs doesn't [19:21] it doesn't, so that it can do versioning(snapshotting) [19:21] k, nice [19:21] well, the only issue i'm seeing with BTRFS, is that it doesn't have fsck [19:22] so once you write data to btrfs, it's never overritten (unless it runs out of empty space) [19:22] which means you can have your fs corrupt and not be recoverable [19:22] If I remember correctly [19:22] fsck isn't needed, cause of the snapshots [19:23] in ext, if the metadata is out of sync with the file data, your screwed [19:23] in btrfs, it doesn't really happen [19:23] Btrfs does not yet have a fsck tool that can fix errors. While Btrfs is stable, it is currently possible to corrupt a filesystem irrecoverably if your machine crashes or loses power. http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg05749.html [19:23] cause you write the data to a new location, then update the metadata for that [19:23] so if it's corrupted, you will just see an older version, but not a corrupted version [19:24] ya, I haven't payed too much attention [19:24] it works for me, and haven't had an issue :) [19:24] and done some reading up on it [19:25] heh, check this out [19:25] Q: Is btrfs with 2.6.36 really rockstable and ready to use in productive environments? [19:25] A: No, certainly not until there's a working fsck tool -- at the moment it's rather easy to kill a btrfs by just losing power. [19:25] which, that is a concern. sure, have a backup power supply, but still [19:27] Q: I was under the impression that with atomic writes it's impossible to mess up a file system? [19:27] A: Yes, we're not seeing data corruption, we're correctly reporting that the transid of the data block doesn't match the transid in the parent node's pointer, which means that some writes went missing. Then we're hitting a BUG() as a result, which hangs. [19:29] my machines can't loose power :) [19:30] ha, but worst case, if power does get lost, don't want unrecoverable data corruption [19:30] ya, would be annoying [19:30] robbiew: yes [19:30] good thing I have a 8hour ups + autoshutdown [19:30] ttx: ok..I can certainly approve BPs as well [19:31] just didn't know if you were expecting me to [19:47] kirkland: hm, guess with natty being open we should be pushing new versions of kvm and libvirt [19:47] hallyn: yup [19:47] hallyn: i usually upload one early one (around UDS) [19:47] hallyn: and then check the qemu-kvm and libvirt schedules [19:47] hallyn: and start thinking about the one we'll want to target to have in by FF/Beta [19:50] well it ain't gonna be today :( [20:12] Is there an official AMI for running Maverick on an EC2 t1.micro instance? (couldn't find one from the main pages, and some of the ones listed in public ec2 image searches look sketchy [1 week old accounts claiming to be "canonical"]) [20:13] kirkland: can you check out the current n-kvm-hypervisor blueprint and tell me whether you still think lxc+uec should still be merged with that? [20:13] hallyn: can you paste the direct url? [20:14] kirkland: https://blueprints.edge.launchpad.net/ubuntu/+spec/cloud-server-n-kvm-and-hypervisor === ivoks is now known as ivoks-afk [20:18] Hobart: the official 32 bit ami's will run as a micro instance afaik [20:25] no matter what i try, i can't get vmbuilder to add the virtual machine i'm creating to my virtualization environment via the --libvirt option [20:25] I tried "--libvirt=qemu:///system" and "--libvirt qemu:///system" [20:28] zoopster -> I'm trying to, but even though I'm using an elastic storage backed AMI, it doesn't have the 'micro' option in the dropdown :( [20:30] Is there a channel for general linux-sysadmin issues? [20:32] Hobart: ami id? [20:32] nevermind, i wasn't seeing it because it was inactive [20:32] Hobart: I only looked at two of them but they both had a micro option [20:33] zoopster -> I clicked the [20:34] ami-508c7839 ... it took me to a page that didn't have those on it, trying again [20:34] that's the one I'm running right now as a micro [20:34] AH! If I paste that in it works [20:35] <_Techie_> i have 2 problems with my server setup, problem one is dhcp3-server drops every now and again [20:36] <_Techie_> and the other problem, is that it fails to come out of sleep mode if i leave it too long [20:36] zoopster -> Many thanks! :) [20:36] Hobart: sure thing [20:53] zoopster -> Do you know if there's a Maverick JeOS build that would work on AWS? [20:54] Hobart: you are pretty much running it...the only other option is the minimal install option with the cd afaik [20:55] I noticed 'screen' and some other stuff was in there, was wondering if there were any smaller bases to start from :) [20:57] you can always roll your own using the minimal option, but those are the only images that canonical provides...removing screen/byobu saves very little [20:57] yep [20:58] hey folks quick question (I hope) is there any reason why someone would create a user as root? and then remove all other users from the machine from the sudoers file? someone set up a machine here and doing a sudo -i gives me a weird error but I suspect it's because something got messed up when a root user was formed. it's for a samba fileshare [21:02] like as a result sudo anything from another user other than the user named root gives a sodoer error and I can't find documentation on why someone would do this === _Techie_ is now known as _TechAway_ [21:18] zoopster -> as far as you're aware, is the kernel in the Canonical provided AMIs already tweaked to be a light Xen client? Or is it the full stock Maverick kernel? [21:19] afaik it's the virtual kernel with the support needed for ec2 including pv-ops support [21:38] hi anybody around know a good place to learn the philosophy behind edge/origin concept, i'm looking for the actual changes i should think of in my code, the data flow, best practices, etc [21:38] i need to get a big application clustered by december and i'm a programmer more then a server admin so i better start reading [21:39] repost from tomcat sry i'm eager for information [21:41] why does ubuntu keep every single kernel installed when it upgrades [21:42] i have about 10 of them left over what is the deal [21:42] diabolical_: you can apt-get remove them if you dont' want them. [21:43] diabolical_: It's a sensible tradeoff, since the newest is the default (with their grub setup), and it makes it easy to switch to an older one if there seems to be an issue. [21:43] well what are the pluses and minuses beside when you reboot it goes into the grub option to boot from which kernel [21:43] but every single kernel back to the first kernel with 10.04 [21:43] http://pastebin.com/3WgQqbAf have a look [21:43] diabolical_: Even if you remove all but the current one, it still goes into grub and gives yo ua few seconds before booting. [21:43] * Kyle__ nods [21:44] I know. The only reson I remove the old ones is because the ssd on my laptop is small. Other than that I usually leave them around. [21:44] yeah i mean it is a headless unit but i have a spare monitor next to it to see if there is any boot issues [21:45] * Kyle__ nods [21:45] kyle on my linux laptop it has a small hd also it is a laptop i saved from the trash but it works for simple stuff [21:45] Eh, I got a good, but smallish ssd for mine. Which made the upgrade to 10.10 all the sweeter. [21:46] New bug: #661452 in tomcat6 (main) "tomcat6-instance-create provides incomplete setup" [Undecided,New] https://launchpad.net/bugs/661452 [21:47] Ok another issue. this was so long ago that i dont remember i set my server to drop all inbound packets besides local. But I can not connect to my mysql server Via anymeans with the exception of phpmyadmin, for instance heidisql, netbeans etc... tried every method I did create a user and assigned it to the local network think iptables is blocking it [21:47] New bug: #661453 in dovecot (main) "dovecot.conf always shows as having been locally modified on update" [Undecided,New] https://launchpad.net/bugs/661453 [21:47] meh, i am getting a netbook soon and upgrading it and putting osx on it :) [21:48] haha [21:48] humm. [21:48] diabolical_: thats a good one. ;) [21:49] why you say that ? [21:49] netbook with os x. Awesome. [21:49] Well i have enough Windows, Linux machines here and there are 2 Macs just not mine [21:50] About your inbound packet dropping.. it sounds to me like what you really want to do is allow ESTABLISHED state packets to come in, but not NEW or INVALID [21:50] SpamapS: so check the firewall logs to see exactly what it says on dropped packets ? and then open accordling ? [21:51] diabolical_: or just iptables -L -n -v [21:51] diabolical_: that will give you counts of matched rules [21:51] I mean I try ssh tunneling, socket etc. whatever the 5 options Every mysql gui frontend has for remote connection without luck [21:53] Ok i will look into it more.... There is so much in the rules that it over runs the screen buffer and cant even see all the rules [21:54] diabolical_: do you have something automatically adding rules? [21:54] no [21:54] diabolical_: then that doesn't make much sense. [21:55] unless you made hundreds of rules yourself. ;) [21:55] I will check but I don't think anything is auto added I got rid of CSF long ago [22:13] crap, somehow I can't start idmapd [22:14] service idmapd start [22:14] idmapd stop/pre-start, process 1574 [22:14] but then it's not running [22:22] is there a way to see what iptables modules are loaded by the system? [22:22] if so, how [22:36] What the hell happened to the sieve package? [22:41] I start hateing NFS [22:41] allways problems... [22:54] hi there [22:55] hope someone can help me, i have 4 harddisk on my linux machine which i want to be shared and managed from my windows computer, don't know if this is the accurate channel, thanks [22:56] what do you mean with "managed" [22:57] of course you can create a fileserver with your linux machine and then let the windows client connect [22:58] Easiest way to share With Windows is by using Samba [22:58] Aison i meant, sorry for my english, i am swiss, i want to access datas from windows machine via the network and shortcuts to the disks physically on linux machine [22:58] yes i have samba and smb.conf edited but this is not working [22:59] hey there. https://bugs.launchpad.net/ubumirror/+bug/628340 [22:59] Launchpad bug 628340 in ubumirror "documentation improvements such as manpages for ubumirror and/or ubumirror.conf" [Undecided,New] [22:59] i paste some stuff here :http://paste.ubuntu.com/514156/ [22:59] anybody know how I might go about excluding old distros if I'm using ubumirror? [22:59] http://paste.ubuntu.com/514156/ [23:00] pimpytux you can use this, just skip first section about installing vmware server [23:00] http://2tap.com/2007/04/22/sharing-files-between-a-windows-guest-and-ubuntu-host-using-vmware-and-samba/ [23:00] vmware-player* [23:00] Black_Prince i will check your link thanks [23:01] pimpytux, did you already add users to samba? [23:01] Aison no guest only i suppose [23:01] i am in local don't need users [23:01] just access to my files [23:01] well, to access your files, you have to create a username and password. then you can connect with your windows machine [23:02] ok this was working good yesterday, and that's why i am lost [23:02] without any user account created [23:02] it's easy to create a useraccount [23:02] simply take your existing one you already use [23:03] then do [23:03] smbpasswd -a yourllinuxusername [23:03] then enter the password [23:03] then you can connect with your linux username and password [23:03] ok will try this one second [23:03] Easiest way to share all hard drives is to add seperate user and mount all four in /home/user/hd1 /home/user/hd2 etc [23:04] it's better to disable guest stuff [23:04] so i type on a terminal : smbpasswd -a pimpy [23:04] or sudo ? [23:04] use sudo [23:04] ok [23:04] so i type on a terminal : sudo smbpasswd -a pimpy [23:04] yes, then enter your password for sudo [23:04] and then enter your password again [23:05] Aison it says this is the same pwd [23:05] as i user the same as for ubuntu login [23:05] then your user already exists in samba [23:05] pwd not changed [23:05] ok good but samba is having issues [23:06] here it's working very good ;) [23:06] :( [23:06] i can't see on the network the drives from linux, i only see my pc on the network [23:06] disable the guest stuff on samba, restart [23:07] and on windows, connect directly [23:07] ok editing the file and restarting smbd [23:07] i remove the lines guest on right ? [23:07] eg. you can create a new link on your desktop with \\server\MP3 and then open the link [23:07] yes, or set it to = no [23:08] ok thanks 1 sec [23:08] on my machine, NFS Server SUCKS [23:08] :( [23:08] smbd restart says rejected send message [23:09] ? [23:09] when i try to restart samba [23:10] service smbd restart [23:11] sudo is better [23:11] ok done [23:11] can't add a shortcut, under network i don't see the pimpytux machine [23:11] ;) [23:11] do i reboot ? [23:11] New bug: #661505 in munin (main) "Please merge munin 1.4.5-3 (main) from Debian unstable (main)" [Undecided,In progress] https://launchpad.net/bugs/661505 [23:12] no [23:12] ok lol [23:24] anybody an idea, why NFSv3 is working, but not NFSv4? [23:25] failing means, I can connect with NFSv4, but the gid/uid are completly crap because idmapd is not able to start [23:29] i am lost [23:29] thanks for your help Aison i have posted on ubuntu.com [23:29] ;-) [23:29] will go to sleep see you another, time 00.30 here [23:29] normally it's really simple to setup samba ;) [23:30] Aison yes i had this working fine on 10.04 [23:30] here i've got a quite big setup, with over 10TB of memory and around 100 users ,) [23:30] but not on 10.10 [23:30] yes, on 10.10 a lot of things suxxx [23:30] had several problems also, but now almost everything is working again [23:30] lol [23:30] n8 [23:31] i could install 10.10 i had to put 10.04 and upgrade [23:31] as i hade grub rescue message after install 10.10 [23:31] never mind [23:31] *couldn't install [23:32] I just connected a drive someone gave me, it has a Linux Raid Autodetect partition, how do I mount that? [23:32] good night [23:43] Skaag: just one? [23:44] Skaag: you'll need to assemble it into a raid volume using mdadm [23:45] Skaag: you can maybe try mdadm --examine /dev/sdX# or whatever the path to the partition is [23:46] I managed to do this!! indeed with assemble [23:46] I ran mdadm --assemble /dev/md0 /dev/sdc3 [23:47] then I insisted to run it with --run [23:47] now it's accessible, excellent :) [23:47] thanks