=== Lcawte|Away is now known as Lcawte [00:11] root@Local:/var/www/html# dir shows js.js. /var/www/html# scp root@ForeignIP:/var/www/html/js.js /var/www/html yields scp: /var/www/html/js.js: No such file or directory. Any ideas on how to fix this? [00:11] I thought it'd be really straightforward [00:13] ApplesInArrays: how about "ssh root@foreignip ls -l /var/www/html/js.js" [00:14] the file needs to be appended like so: scp myusername@university_computer:/home/myusername/file.odt homeusername@Felix:Desktop/file.odt [00:16] ApplesInArrays: I don't think that's it: http://paste.ubuntu.com/9758961/ [00:16] sarnold: you're probably right. I was looking at https://help.ubuntu.com/community/SSH/TransferFiles === Lcawte is now known as Lcawte|Away [00:44] Hey hey. I know this isn't exactly an ubuntu issue, but seeing as we are all sysadmins in here maybe someone has seen this before? I am working on a rather old server that's lenny based giving segfaults (http://pastebin.com/raw.php?i=GXiuCQcj) that don't make any sense to me. [00:51] F^3: a few thoughts; most of them happen at the same point in the program, it'd be worth getting core dumps from the thing and the ngetting stack traces [00:51] F^3: it might be bad memory modules, memtest86 might help, but it seems too consistent to me [00:51] I don't even know what app it is. [00:52] sarnold, Maybe a full stick has gone bad? [00:53] F^3: oh? I assumed the 'b' was some anonymization you'd done.. [00:53] sarnold, No :( [00:53] 'b' was taken directly from dmesg [00:55] F^3: you could use a tool like 'execsnoop' here to try to find it: http://www.brendangregg.com/blog/2014-07-28/execsnoop-for-linux.html [00:55] sarnold, Since this is lenny I can't access repo [00:55] F^3: there's a chance the program changes it's name after execution, which might make it harder to spot.. [00:55] 's anymore to install compiling apps. [00:56] F^3: ohhhh. so it might be too old to use the kernel's ftrace tools? [00:58] Possibly. It hasn't been updated in nearly 3 years and this system has been generating abuse reports lately which is why I'm looking into it. All signs point to it not being an intrusion unless they're erasing their specific entries in the logs. [00:58] There is a squid proxy though so I assume they're using that to make the abuse happen. [00:58] if it had a -reasonable- process name I might think it's brute-force exploit testing [00:59] probing for correct offsets to use or something similar [00:59] but 'b' is just so strange. [01:00] intrusion feels plausible for me; you've got t ofigure out what is spawning that 'b' process. That'll answer a lot of questions. [01:01] I'm probably going to just wipe/rebuild the server if it passes memory tests. Else I'll just move the ip to a new box. [01:01] that might be faster :) but my curiousity would eat at me.. [01:02] Since I am a programmer this b proc doesn't intrest me ;P [01:03] The only thing that interests me is getting the proxy back up to keep my inbox from being flooded. [01:03] anyone here help with an LVM question? [01:04] F^3: aha :) then happy fixing :) [01:04] trying to extend an lvm partition and it cant find /dev/sda* [01:04] https://gist.github.com/mikedevita/4ab15b876cab1d69ea79 [01:04] basically / is 100% utilized.. [01:05] On another note why does the default partition setup for ubuntu server only give /boot 250mb. Old kernels are not getting automatically removed so I keep filling it up. [01:07] sarnold, I'm going with the server being rooted. Found this in /etc as 1418130075.pl http://pastebin.com/raw.php?i=wzgBPMY1 (one of about 8 different .pl files) [01:08] F^3: yup. there's no good reason for that to exist. [01:08] F^3: sorry. [01:09] F^3: on the plus side, that means your way forward is clear :) new install from scratch. [01:14] sarnold, But that takes the fun out of upgrading through two eol debian versions. [01:15] F^3: oh man. I guess you can always try home dentistry if you really miss it.. [01:17] I'll pass. I don't trust myself using a drill in my mouth using only a mirror to see. === Guest97735 is now known as mfisch === 18VABY2RP is now known as lazyPower === bearface_ is now known as bearface === markthomas is now known as markthomas|away [02:15] anyone else here using apt-mirror ? === zz_DenBeiren is now known as DenBeiren === martinst is now known as martins-afk === Nigel_ is now known as G === Dry_Lips is now known as Gutenberg [05:24] Now i have Ubuntu 13.04 i want to upgrade to Ubuntu 14.04.1 LTS how can i do ? Will it upgrade to Ubuntu 14.04.1 LTS if i use sudo do-release-upgrade or it will upgrade to 14.10 ? [06:46] can i disable gpg checks fo a single repo like i can with yum repos? i dont want to do it per check, i want to do it for all checks on that particular repo. its my private one === PaulW2U_ is now known as PaulW2U === dr0x16_ is now known as dr0x16 [07:39] MACscr: it's probably easier to sign your own repo... [08:25] I have upgraded my samba to version 4.1.6 but Im still getting the talloc memory leak error. Is there any way I can keep samba password sync to user accounts and get rid of this memory error? no talloc stackframe at ../source3/param/loadparm.c:4864, leaking memory [08:27] I have upgraded my samba to version 4.1.6 but Im still getting the talloc memory leak error. Is there any way I can keep samba password sync to user accounts and get rid of this memory error? no talloc stackframe at ../source3/param/loadparm.c:4864, leaking memory === Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away [08:57] Good morning. === lordieva1er is now known as lordievader [08:59] Ciao! [08:59] I've this question for you :D [09:00] bimma: Shoot ;) [09:00] I would mount via fstab a windows share with possibility to change permission, with chmod, of the subfolder... with cifs is not possible... [09:01] windows neetwork share [09:02] Makes sense, your Windows share is likely NTFS. NTFS doesn't support Unix file permissions. [09:02] yeah i know [09:02] but there isn't a workaround? [09:03] or another way to mount? [09:03] :( [09:03] No. The solution would be not using NTFS ;) [09:04] :D [09:04] ok [09:07] I have upgraded my samba to version 4.1.6 but Im still getting the talloc memory leak error. Is there any way I can keep samba password sync to user accounts and get rid of this memory error? no talloc stackframe at ../source3/param/loadparm.c:4864, leaking memory === Lcawte|Away is now known as Lcawte === kickinz1` is now known as kickinz1|afk === kickinz1|afk is now known as kickinz1 [10:00] hallyn, https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1411575 can you give your opinion on that [10:01] hallyn, it would nice to have that in vivid; not sure about the backport :-) === rvba` is now known as rvba === RoyK^ is now known as RoyK [10:42] zul, coreycb: adding qemu and libvirt to the kilo CA for ppc64el support === kickinz1 is now known as kickinz1|afk === kickinz1|afk is now known as kickinz1 [10:46] How can i Control a user not to Access crontab Command in Ubuntu Desktop and server === unreal_ is now known as unreal [11:58] hello [11:58] how do i enable this? [11:58] http://www.ubuntu.com/usn/usn-2455-1/ [11:58] i cant find what the command is [11:59] thanks in advance [12:02] Craig4who: instructions are at https://wiki.ubuntu.com/Security/Upgrades [12:13] So I plan on migrating servers but i'm not sure what software to use for this, I don't really want to pay $100 for a migrate, so is there anything free? [12:21] hi rbasak [12:21] what do i search for on that link? [12:22] cant find anything useful i can follow === Lcawte is now known as Lcawte|Away [12:27] yea i dont want to implement the fix [12:28] i have already done this [12:28] i want to reenable it so it works again [12:28] Shell command execution can be re-enabled using the expandaddr option. [13:05] <_zxq9_> Does an LTS release receive updates between Postgres minor versions, or only patch updates? As in, 9.3 will forever be 9.3, but 9.3.X are expected updates. Or would 9.4 or 9.5 ever hit an LTS repo? [13:06] hell [13:06] hello [13:07] anyone there? [13:10] .. [13:11] @ubuntu [13:11] #ubuntu [13:13] <_zxq9_> shaurya: People are here, just probably most are doing Friday things. [13:14] i'm installing the badgekit api from mozilla. anyone know why im getting this error when I run a script? http://pastebin.com/p26qNmVG === the_ktosiek is now known as ktosiek === jrwren_ is now known as jrwren === chmurifree is now known as chmuri [13:40] jamespage: looks trivial enough [13:41] jamespage: also, we have 2.2.0 in a ppa for now, [13:41] i'll merge it into vivid soon, but am hoping it goes from experimental into unstable first [13:41] _zxq9_: postgres would get minor updates only, afaik. [13:42] _zxq9_: there are other ways to get postgresql packages: http://www.postgresql.org/download/linux/ubuntu/ [13:43] <_zxq9_> jrwren: ty [13:44] jamespage: i'll go ahead and push it (well, after testing) to v === Pici is now known as Guest66126 [13:51] hallyn, awesome === Guest66126 is now known as Pici [13:53] jamespage: but i'm having fs corruption issues on my rootfs. grr. which doesnt' allow me to create a lv to build in. this'll slow me down [13:53] breakfast first, then deal with it. bbl === kickinz1 is now known as kickinz1|afk === Guest8667 is now known as balloons === nickande1 is now known as nickander === Lcawte|Away is now known as Lcawte [15:55] hello my dovecot sql roundcube quota is displaying 0MB from unknown [16:35] how easy is it to rebuild a software RAID array in ubuntu using the default software raid setup? [16:37] asking because I am entertaining potential ideas [16:38] sudormrf, https://raid.wiki.kernel.org/index.php/RAID_setup#6._Manage [16:39] ppetraki, ? looked at that wiki entry before coming here. that wiki entry doesn't talk about rebuilding an array when a disk fails [16:41] sudormrf, you just do an add while in manage mode, the man page has the rest. It's all pretty straight forward. like so http://www.thomas-krenn.com/en/wiki/Mdadm_recover_degraded_Array#Re-adding_the_partition_to_the_array [16:42] ppetraki, thanks. I am going to test this out in a VM right now (build the array, simulate a failure) to see what happens. if I like it I will probably use it on my new NAS [16:42] sudormrf, faster to use loopback block devices http://www.thomas-krenn.com/en/wiki/Mdadm_recover_degraded_Array#Re-adding_the_partition_to_the_array [16:43] sudormrf, woops old link http://askubuntu.com/questions/546921/how-to-create-virtual-block-devices === martins-afk is now known as martinst === ^rcaskey is now known as rcaskey === markthomas|away is now known as markthomas [17:25] ppetraki, so here is what I am trying to do -> main FS on SDA, sdb-e will be raided together. do I need to run through the normal install first and then go back and setup RAID or is there a way I can accomplish this through the normal setup? [17:34] sudormrf, If it were me, I would keep the install simple and script the data array creation after the OS is up [17:34] yeah. that seems like the way to go === Lcawte is now known as Lcawte|Away === Guest78540 is now known as hxm === Lcawte|Away is now known as Lcawte [19:10] looking at doing this in cent now. thank you for your help ppetraki [19:10] sudormrf, np [19:11] any idea what this is about? "[504700.275921] IPv4: Oversized IP packet from 192.168.10.254 === mayday_jay is now known as lektrik [19:11] any idea what this is about? "[504700.275921] IPv4: Oversized IP packet from 192.168.10.254" [19:12] the IP address is the address of the machine [19:14] RoyK: look slike that error message comes from IP packet fragment re-assembly. [19:14] RoyK: someone trying to sneak packets through NIDS is my first guess [19:16] NIDS? [19:18] ppetraki, are you familiar with mdadm at all? [19:18] sudormrf, yeah [19:18] ppetraki, ok. so I have a couple of questions about it, would you mind answering them if you can? [19:19] sudormrf, go ahead [19:19] so to simulate a disk failure, I removed one of the HDDs from the VM [19:19] the system would then not boot (dropped me into maintenance mode)...is this normal behavior? [19:20] first time ever messing with this stuff [19:20] sudormrf: IIRC ubuntu still refuses to boot off a degraded raid by default (which is rubbish) [19:21] sudormrf: this can be changed easily https://help.ubuntu.com/community/Installation/SoftwareRAID#Boot_from_Degraded_Disk [19:21] yeah hang I think I know what you need to boot degraded [19:21] refusing to boot from degraded is rubbish [19:21] the whole point of a raid is to allow the system to work with a dead drive [19:21] in "/etc/initramfs-tools/conf.d/mdadm" set "BOOT_DEGRADED=true" and rebuilt the initrd [19:22] ok so that is normal behavior if you don't change it. that is good to know. [19:22] sudormrf, RoyK +1 [19:23] RoyK: network intrusion detectio nsystem [19:23] I also set "md-mod.start_dirty_degraded=1" in /etc/default/grub just in case [19:23] sarnold: possibly my openvas scans ;) [19:23] RoyK: aha :) [19:23] RoyK: feels likely [19:24] sarnold: thanks [19:25] sarnold: any idea how to disable that for outgoing packages? [19:25] thanks, looking in to this some more [19:25] will be back with more ?'s later I am sure :) [19:26] RoyK: it was probably -incoming- with faked source address [19:26] sarnold: well, it hasn't happened since I paused the scan [19:26] sarnold: and the scan is against some computers at work [19:27] just trying to open some eyes about computer security at work :P [19:28] this guy with a PhD has setup a lot of stuff, but doesn't seem to spend much time patching i [19:28] t [19:29] RoyK: I'm not sure how a packet could be fragmented on the sending side, but .. i'm not -that- familiar with the networking code, it's pretty finely tuned which sometimes means not easy to reason about :) [19:29] RoyK: ugh. [19:33] ok another question...how can I verify that the stuff is actually being mirrored? :X [19:35] you could use hdparm to grab the same sector from each drive an diff it [19:36] or dd [19:36] sudormrf: just run a check [19:36] sudormrf: echo check > /sys/block/mdX/md/sync_action [19:36] or perhaps [19:36] sudormrf: echo repair > /sys/block/mdX/md/sync_action [19:37] sudormrf: mdraid is used in thousands or millions of home NAS systems and in large-scale production systems [19:37] sudormrf: it works [19:38] sudormrf: https://raid.wiki.kernel.org/index.php/RAID_Administration <-- good docs here [19:44] sudormrf, dd if=/dev/foo skip=$(((17408/512) )) bs=512 count=1 | hexdump -C [19:44] * ppetraki took me a min to find that [19:45] where 17408 is your block number [19:47] sudormrf, so if you were really paranoid you would read say 10 blocks from each disk, create an md5sum and then check [19:47] or use cmp to compare them directly [19:48] does the url parameter in kickstart honor the proxy setting? [19:51] rcaskey: if you want to use a proxy in kickstart, you have to specify it. [19:51] * ppetraki forgot about cmp, uses dd rune to debug metadata format [19:51] bekks, I want to use a proxy for fetching packages only [19:51] and attempted to the line I used was... [19:52] url --url=foo --proxy=192.168.1.2:1234/ [19:53] foo! [19:53] err http://192.168.1.2:1234/ (also tried it without the trailing /, although it was present in a lot of examples) [19:54] rcaskey: For me, that works. [19:54] hrmm [19:54] if it fails to access the proxy server does it try directly or just fail? [19:55] rcaskey: Thats something you have to find out, since we cant know it :) [19:55] RoyK, thanks! [19:55] well i'm not seeing it go through the proxy so i'm trying to figure out if it was maybe trying and falling back and I had an issue with the proxy or if i don't have the config quite right [19:56] RoyK, yeah, I believe you :). this is my first time playing with it so I am trying to wrap my head around it so I understand it before I put it to use with my files :D [19:58] basically right now I have 2 NASs + 2 servers [19:58] one of the server is a file server [19:58] the other is just doing stuff [19:58] well the file server is doing more than file serving [19:58] one of the disks on my WD EX2 failed [19:59] sudormrf: replace it ;) [19:59] thinking about replacing that with something more beefy and pulling the EX2 + my cloud NAS out of production and putting a 4 disk RAID 1 setup in place. these things are primarily used as backup devices. [19:59] so I go from 2 plugged in devices down to 1 [20:00] sudormrf: investigate zfs; this series of blog posts are a nice fast read and will give you a nice overview of how it works: https://pthree.org/2012/04/17/install-zfs-on-debian-gnulinux/ [20:02] sarnold, yeah I have. been toying with the idea of running BSD on this thing with ZFS [20:02] sudormrf, http://www.supermicro.com/products/motherboard/Xeon/C600/X9DRX_-F.cfm beefy? [20:02] but ZFS will require a larger investment for the hardware [20:03] sudormrf: don't believe the crap about it -requiring- ECC memory.. [20:03] sudormrf: ecc just tends to align well with the paranoia levels of the people who like to deploy zfs :) [20:04] ppetraki, don't tempt me :P. I don't have space for it [20:04] sarnold, has nothing to do with that [20:04] ahd everything to do with ZFS being very RAM hungry [20:04] sudormrf: ahhh [20:04] sudormrf, get you some NVDIMMs and run your whole NAS out of ram [20:04] although, it is not off the table, just something I have to consider [20:04] sudormrf: whatever you do, do not use de-dupe [20:04] trying to build something under the price of the synology thingy [20:04] that'll help memory needs immensely :) [20:05] use dd for what? [20:05] ppetraki, lol [20:05] sudormrf, what kind of workload you got? [20:06] rcaskey, this is a home device that will primarily be used for backing up the main file server and acting as a file server for some other data (which will get sent to the main file server as necessary) [20:06] nothing too crazy [20:06] depending on your definition of crazy, I suppose [20:06] although the WD EX2 is a bit slow for my tastes [20:06] sudormrf, how much space you need? [20:07] rcaskey, this will be an 8TB RAID1 setup when done. 4 disks, raid 1, all 4 disks being 4TB drives [20:07] sudormrf, i'd think basically you can grab anything cheap and do a software raid controller without zfs and be very happy [20:08] software raid controller? software raid needs a controller? [20:08] a raid needs a controller [20:08] that might just be in software [20:09] I was working under the pretext that software raid needs no controller as it it software [20:09] ah [20:09] gotcha [20:10] let me show you what I was thinking for the build [20:11] also i'd strongly consider at least some ssd storage + bcache [20:13] how can i change the locale globally? [20:13] http://paste.ubuntu.com/9763496/ [20:13] also [20:13] +SSD for system drive [20:13] i set the local yesterday with update-locale LANG=... but after a reboot, locale still show's no LANG [20:13] 120gb ssd would do nicely [20:15] sarnold: ZFS doesn't require ECC [20:15] RoyK: that's what I told sudormrf :) [20:15] sudormrf, Micron M500/600 are good [20:16] that little norco is cute :) [20:18] ppetraki, I like samsung drives meself [20:19] sarnold, it's a pretty good deal for the money. hard to find many cases like that. although I got this link from faug in another channel: http://www.mini-itx.com/store/nas [20:19] sudormrf, we kinda destroy drives a living around here and have found the Micron ones to be the most resilient [20:19] ppetraki, heh noted :) [20:20] sudormrf, low write amp, decent garbage collection [20:20] don't know the quality, but this case looks nice: Chenbro ES34069. if it is anything like the quality of lian-li, I am not going to touch it. [20:20] sudormrf: I've been looking at "larger" systems myself, I'd like to build something with three 3-way mirrors and room to grow by adding another 3 or 6 drives.. and it'd be nice to have space for some SSDs for l2arc and OS [20:21] sarnold, if I had the space I would soooo do that [20:21] basically trying to come in with better specs for cheaper than this: http://www.newegg.com/Product/Product.aspx?Item=N82E16822108183&cm_re=synology-_-22-108-183-_-Product [20:22] I think I have done so. not sure how the i3 in my parts list compares to the atom in the synology, but typically core i > atom [20:22] sudormrf: it's impressive, isn't it? it's just cheap enough to tempt you to do better, but ... :) [20:22] yeah it is heh. one thing about the synology is there is an app for my phone that would allow me to access the files on my iPhone. that is a really nice feature to have [20:23] don't think there is any way I can do that with a home built system (not fond of owncloud) [20:23] nginx? :) [20:24] I think the problem would be that there needs to be an app for the iPhone to act as the client [20:24] I don't blame you about owncloud, I've gotten bad vibes about them.. [20:24] I have used it and it just feels clunky on anything but a computer [20:24] and it's written in PHP [20:24] yeah, bad vibe #1 :) [20:24] I don't have the development ability to help the project, unfortunately. [20:25] even if you did, it's in PHP :-p, that stuff just repels developers [20:25] there was some other thing similar to owncloud, but it was primarily developed in china, so....pass. [20:26] based on the parts list, do you think I need a beefier CPU? [20:26] I could upgrade the CPU in my main file server to an i7 and pull the i5 that is in that and repurpose [20:26] ;D [20:27] overkill I am sure, but someone was saying that software RAID is pretty CPU intensive. [20:27] depends on the services you're going to run; if it's mostly backup and over-the-internet mobile-based accesses, i3 ought to do alright [20:27] sudormrf, probably not... but more cores are better, especially for distributing interrupts [20:27] yeah [20:27] I want to figure out the over the internet mobile based access part :S [20:27] I am sure there is some sort of app that would let me access samba shares somehow [20:28] https://itunes.apple.com/us/app/fileexplorer-free/id510282524?mt=8 [20:28] derp [20:28] haha [20:28] sudormrf, if you could get more memory, do that, and add a cache, if read speed matters that is, but it doesn't sound like it. So just pocket the difference and buy yourself an even bigger monitor [20:28] first google hit [20:28] sudormrf: Yeah, the samba client [20:28] how to use the locale-command to change the LANG-variable permanently? [20:28] I'd be reluctant to expose samba to the whole internet. [20:28] ppetraki, I have no need for a monitor :D. I don't have a desktop in my place (no space for it). mobo only has 4 sata ports, so I might be having to find a new mobo. [20:28] sarnold, I wouldn't. I would vpn back home [20:28] sudormrf: good good [20:29] sudormrf, well there is that super micro... [20:29] as long as I can get to it, that is what matters :D. ppetraki is that supermicro mini-itx? [20:29] has to fit in that norco case [20:29] oh no, there's nothing micro about what I proposed, let me check your link again [20:29] I know of a board that is mini itx with 8 sata ports and dual nics...but it forces you to use an atom CPU [20:31] if my tax return is as big as it is looking right now (still waiting on second w2), then this is going to be built pretty soon [20:31] lol [20:31] sudormrf, what you have looks fine honestly [20:32] sudormrf, you got a 3x PCI slot, you could get a LSI card and do HW RAID [20:32] ppetraki, don't know if that would fit in the norco case. looking for another mobo with 6 sata ports. just thought about it since I need 1 sata port for the system drive. [20:32] sudormrf, I mean 16x, v3.0 [20:33] sudormrf, they have dimensions maybe not on their site, but I'm sure if you google it you'll find somebody posted it [20:33] like this: http://www.newegg.com/Product/Product.aspx?Item=N82E16813128720&cm_re=mini%2bitx%2bmotherboard-_-13-128-720-_-Product but without all the garbage on it (wifi, bluetooth, etc). that wouuld be nice. [20:33] ppetraki, curious what language you'd prefer owncloud to use? i would assume anything but php, heh? [20:34] dual nic is nice [20:34] vonsyd0w, basically [20:34] sudormrf, have you considered the intel avoton boards such as this one? http://www.newegg.com/Product/Product.aspx?Item=N82E16813132230R&cm_re=gigabyte_avoton-_-13-132-230R-_-Product [20:34] vonsyd0w, "go is the future" I suppose [20:34] thats what I'm looking at to replace my HP N54L [20:35] ppetraki, i'm seeing a lot of projects starting to use "go" [20:35] like this new ansible gui i saw on reddit earlier this week uses go [20:35] vonsyd0w, python community isn't doing itself any favors. That whole annotation thing that just came out... [20:35] annotation thing? i'm not familiar with that? [20:35] link? [20:36] vonsyd0w, price too much out of range [20:36] vonsyd0w, it was on lwn recently [20:36] but nice board :) [20:36] sudormrf, it includes the CPU remember that [20:36] vonsyd0w, http://lwn.net/Articles/627418/ [20:36] thx [20:37] oooohhhh that was one of the ones I think I was looking at before [20:37] I mean I use python everyday but if we were going to build something large in userspace I would probably push Go, especially vs C++ which is the other prevailing camp here [20:37] but has atom [20:37] :S === wendar_ is now known as wendar [20:40] ppetraki: honestly, if python had -started- with type checking with a better non-hacky syntax from the start, I might have enjoyed using it ;) [20:40] ppetraki: but bolting on a bunch of stuff to the side .. sigh [20:41] sarnold, whats your preferred language? [20:41] sarnold, right, that's the thing, and in python's defense it's been around a loooong time [20:42] vonsyd0w: I haven't done more than toys in Rust, but it looks very promising. [20:42] Weren't there like exclamation points in the Rust syntax? [20:42] sudormrf, there are a bunch of boards with the intel avoton chip to consider as well. give that a search and see if it fits your budget. Here is a case I was looking to get w/ an avoton board: http://www.u-nas.com/cases.html [20:44] if I remove dual NIC from the picture my options are not as limited [20:44] vonsyd0w, but how does that perform vs an i3? [20:49] well aren't all i3's dual cores? The intel avoton is a quad or octa-core board. It really depends on your work load. If its a storage box, then the avoton will do just fine. Here is a cpuboss.com comparison between the two that are similarly clocked: http://cpuboss.com/cpus/Intel-Core-i3-3110M-vs-Intel-Atom-C2750 [20:49] ppetraki: yeah, to differentiate macros from non-macros. I LIKE IT BETTER THAN C AND C++'s WAY OF INDICATING MACROS :) [20:51] anyone here familiar with debmirror [20:57] sarnold, heh [21:38] Hello [21:38] hi! [21:38] I've just upgraded the packages on a 14.04 server and it hangs with this warning "W: mdadm: /etc/mdadm/mdadm.conf defines no arrays" [21:39] are you using a software raid? [21:39] I have a RAID 10 array configured as /dev/md127 that still works but I'm afraid of issuing a reboot as I might lose it. [21:40] Should it be present in mdadm.conf? [21:40] it looks like something to be concerned about though i'm not familiar with its functionality [21:41] mdadm --detail --scan shows the array [21:42] should I add it to mdadm.conf? [21:42] seems like a good idea [21:43] cmircea, theres a copy of it in your ramdisk [21:43] cmircea, mkdir /tmp/foo; cd /tmp/foo; zcat /boot/initrd-BLAH | cpio -id; cat etc/mdadm/mdadm.conf [21:43] ppetraki, uh, what ramdisk? [21:45] cmircea, initial ramdisk, Linux needs an early copy of it to assemble your array at boot to form things like booting to an MD [21:46] ppetraki, I see. I have several initrd images, one ending in .new which I assume is the one being installed. [21:46] Should I zcat the latest non-new? [21:46] cmircea, yup! [21:47] any of the old ones should do, its not something that changes often [21:47] ppetraki, looks like the array wasn't mentioned in mdadm.conf [21:47] ppetraki, root is not on the array [21:48] cmircea, well, here's what a working config looks like, http://pastebin.ubuntu.com/9763976/ [21:49] cmircea, just get your UUIDs from mdadm and you're good. [21:50] ppetraki, thanks, I've added mine to mdadm.conf [21:50] cmircea, \o/ [21:51] * ppetraki it's so friday [21:51] What should I do about the hanged SSH connection from which I ran aptitude upgrade? [21:51] do any of you guys have any experience with time machine on ubuntu server? setting it up and what not [21:51] ppetraki, last thing is said was that mdadm defines no arrays. [21:51] wondering if this is still valid: https://www.64bit.co.uk/ubuntu-as-a-osx-time-machine/ [21:52] cmircea, kill it, run it again but this time from within a tmux session but not before you copy your new shiny config file in place [21:52] cmircea, shift ~ . [21:52] cmircea: start running long-running commands in screen :P [21:52] or tmux ^ [21:54] sudormrf, don't know much about macs anymore but everything in that article looks reasonable. The mac itself is the biggest variable [22:01] yeah. I have an OSX vm here I can test with [22:02] whats the prefered way to set the LANG-variable globally on ubuntu server 10.04? [22:02] ppetraki, for the mdadm setup, would you say the steps listed here: https://www.grumpyland.com/blog/183/installing-software-raid-on-centos-567-via-ssh/ will work in ubuntu server (changing the obvious things like yum)? [22:02] thor77, to wait 3 more months, when it goes unsupported, so you don't have to? [22:03] Patrickdk: no good solution [22:03] just saying, support ends in 3 months, you should be moving off it as quick as possible [22:04] uh. wtf [22:04] i meeant 14.04 [22:04] sry [22:04] :) [22:04] ppetraki, SSH connection refused :< [22:04] my hoster doesnt even provide a 10.04 image^^ [22:05] i tried to set it with update-locale, but after a reboot it was gone [22:05] thor77, edit /etc/default/locale [22:06] update-locate updates the *available* locate, it doesn't make one a default [22:06] http://paste.ubuntu.com/9764044/ [22:06] ^ my /etc/default/locale [22:06] hmm, maybe it does [22:06] but after a reboot "locale" -> http://paste.ubuntu.com/9764047/ [22:06] everything's gone [22:07] i think should set it for every user, right? [22:07] no [22:07] uh. [22:07] that only sets the DEFAULT for system services [22:07] how can i set it for every user? [22:07] in /etc/profile.d ? [22:08] what should i insert there? [22:09] I would imagine it would inherit it though [22:10] looks like the users doesnt take the LANG-var from /etc/default/locale [22:10] it should actually [22:10] but it doesnt [22:11] what does, locale -a, show? [22:11] http://paste.ubuntu.com/9764072/ [22:11] its rly weird [22:11] there's de_DE.utf8 but locale-gen said de_DE.UTF-8 [22:11] dont know whats the right name [22:12] looks right [22:12] ? [22:13] should i set the default to de_DE.utf8 ? [22:13] no [22:14] what should i do then? [22:14] I am not sure [22:15] cause everything you did is right, not sure why it didn't work [22:15] and it should never be empty [22:15] but it is [22:16] yes [22:16] im very confused [22:31] sudormrf, looks right, doesn't talk about updating the initrd though, which should copy the config there too [22:32] sudormrf, which I guess would be dracut [22:33] sudormrf, ... and that's the end of my day, have a good weekend! [22:41]