[03:20] <major_majors> i'm in the middle of an Ubuntu server install. i've set up a software RAID 10 with 3 LVM volumes on top: /, /home, and swap. now i'm at the stage where it's asking me where i'd like to install GRUB and i'm lost. where should GRUB go in this configuration?
[04:15] <nezZario> What kernel should I have running trusty?
[04:15] <nezZario> I have 3.13.0.-55-generic
[04:15] <nezZario> I have no clue what horrific things someone did to this server
[04:27] <sarnold> nezZario: that looks about right, seven hours ago 3.13.0-55.94 was released
[04:28] <sarnold> I'm not sure how the minor versions will appear in uname -a output, probably only dpkg -l output will have the 55.94 bit
[04:45] <nezZario> might make you laugh - very sleep deprived, meant to type "sudo chmod" and somehow typed chudo sumod (seriously)
[04:45] <nezZario> thanks sarnold, i'm going to sleep. =)
[04:45] <sarnold> nezZario: hah, that's awesome ;)
[04:46] <sarnold> nezZario: yeah, time to step away from the keyboard :) have a good night, nice weekend :)
[07:12] <Norbin> hello
[07:12] <Norbin> running latest ubuntu server through hyperv, the terminal window bugs on me, it finishes loading and then nothing responds, i can't type anything
[07:12] <Norbin> i was told to change the resolution via the grub config file which i did - tried different color depths/res codes
[07:13] <Norbin> but same issue persists
[07:13] <Norbin> i even reinstalled
[07:14] <Norbin> http://i.imgur.com/aCqYqhs.png
[07:14] <Norbin> this is what i see
[07:15] <sarnold> Norbin: I wonder if you turn off lightdm service if you'd get to a nice getty prompt or not
[07:24] <Norbin> shall i try it? via recovery?
[07:24] <Norbin> since i can't really do anything on the actual terminal
[07:24] <Norbin> only way i can 'do' something is via recovery / as root
[07:24] <Norbin> i think so at least, not a linux expert :)
[07:34] <Norbin> annnnd it works, did absulotely nothing but reinstalled for the 3rd time
[07:34] <RoyK> Norbin: http://askubuntu.com/questions/384602/ubuntu-hyper-v-guest-display-resolution <-- something like that?
[07:35] <RoyK> Norbin: I used to admin hyper-v and it was a PITA with ubuntu, this was v1, though, so hopefully better now
[09:44] <LeMike> hello. I am doing a rsnapshot and want to have the date in the log. unfortunately the log always shows "/bin/date +%F" instead of the real date. how can I have the real date logged instead of the command? would you just append the output?
[10:09] <lordievader> Hello
[10:22] <LeMike> damn. my rsnapshot commands generate empty directories like "unused0" ... how do I avoid that?
[10:30] <jpds> LeMike: Real date as in "date -R"?
[10:33] <LeMike> any kind of date jpds. even `/bin/date` would be listed in the log but not it's output
[10:36] <jpds> LeMike: Wait, is this a script you wrote around rsnapshot?
[10:37] <LeMike> nope. I do `backup_script  /bin/date +"backup started at %F"` and the command is written in the rsnapshot.log instead of it's output. I like to have the output logged or some kind of "started at 2015-06-20" in the log
[10:38] <jpds> LeMike: So, why do you need that?
[10:40] <LeMike> to put those two lines in another script that will monitor the backups. jpds
[10:40] <jpds> LeMike: You can't read the cron output from syslog?
[10:40] <LeMike> Dunno. Can I?
[10:41] <jpds> LeMike: Well, cron logs every command to /var/log/syslog with a timestamp already.
[10:42] <LeMike> jpds:  `grep rsnapshot /var/log/syslog` is empty :/
[10:42] <jpds> LeMike: Do you run rsnapshot with cron?
[10:43] <jpds> LeMike: Also, you know that you can have the monitor check the timestamp of the last backup directory?
[10:44] <LeMike> yes rsnapshot runs via cron for three days now, jpds. the timestamp would not help, as I need to monitor the duration of every backup.
[10:46] <jpds> LeMike: This is on a centos box but:
[10:46] <jpds> 2015-06-18T21:00:01.174637+00:00 backup CROND[27783]: (backup) CMD (/usr/bin/rsnapshot hourly)
[10:46] <jpds> 2015-06-18T21:00:52.106269+00:00 backup rsnapshot[27830]: /usr/bin/rsnapshot hourly: completed successfully
[10:47] <jpds>  /var/log/cron and /var/log/messages on centos.
[10:47] <LeMike> those files are not there :/
[10:48] <jpds> LeMike: Yeah, not on Ubuntu, they're both in syslog.
[10:50] <LeMike> syslog does not have it. seems like the information will be lost there after some hours/days/lines. I need to have it persistent in the rsnapshot.log (which is in the backup) but with the proper date.
[12:15] <drAvanti> hello all,
[12:16] <drAvanti> can someone pls help me with info: I want to setup and ubuntu box as a file server, however I dont now if there is a gui app that can help manage users, their rights, inheritance and root folders, etc
[12:42] <RoyK> drAvanti: there are some, but generally not very good. you don't need to learn a whole lot to do it from the commandline, and you'll do it better and faster that way, and you'll end up understanding what really happens
[12:42] <RoyK> drAvanti: there are things like webmin (DON'T USE IT) and others, but still, better do it manually
[12:42] <RoyK> !webmin
[12:43] <drAvanti> @Royk thanks for the info. But in a case where I would like a user to administer the server....add new users, create new folders, etc thats why I asked for a good gui
[12:43] <RoyK> drAvanti: how many users?
[12:43] <drAvanti> except if it can be done via a web browser.
[12:43] <drAvanti> like 60users
[12:44] <RoyK> !ebox
[12:44] <RoyK> oh, it's zentyal now...
[12:45] <drAvanti> k, thanks, ill check it out
[12:47] <RoyK> drAvanti: looks like http://ajenti.org/ is mentioned also
[12:48] <drAvanti> this zentyal looks commercial
[12:49] <RoyK> mhm - it was in ubuntu earlier - I don't quite follow these things
[12:50] <RoyK> ajenti is GPL, though, and on github
[12:50] <RoyK> https://www.digitalocean.com/community/tutorials/how-to-install-zentyal-on-ubuntu-14-04
[13:31] <trippeh> 64bit Ubuntu 15.04 boots in <1s and uses a whopping 20MB of RAM with sshd and some ntp running
[13:32] <trippeh> fairly minimal install based on Ubuntu Core (apt-get edition) though, but still impressive.
[13:37] <RoyK> trippeh: systemd is doing its tricks, like with windows bootup, you get the login prompt before other things are started
[13:37] <trippeh> according to logs, its still within 1s
[13:38] <RoyK> you can see that on centos/rhel7 where you may have to wait a minute after the login prompt shows up, until you can ssh in
[13:38] <trippeh> also, not in container, but full vm with vga, bios etc.
[13:38] <RoyK> but then, centls/rhel sucks imho
[13:38] <trippeh> so that 20MB includes kernel
[13:38] <RoyK> I beleive so
[13:39] <RoyK> you'll probably need around 80MB for the whole boot, though, to unpack the initrd
[13:39] <RoyK> or you can strip down the initrd to only modules you use by setting MODULES=dep in /etc/initramfs-tools/initramfs.conf
[13:40] <trippeh>             3400720             8197120  58.5% initrd.img-4.0.0-1
[13:40] <RoyK> that's small
[13:40] <trippeh> oh, yeah, not ubuntu kernel, but I doubt that changes much besides initrd/vmlinuz size
[13:41] <trippeh> most of the initrd and extra vmlinuz stuff is discarded after use, though
[13:41] <RoyK> last I tried to boot debian on low memory, I came down to 56MB - lower than that just crashed
[13:41] <RoyK> yep...
[13:41] <RoyK> so don't use the standard initramfs for small systems ;)
[13:45]  * trippeh tests with modules=dep
[13:46] <trippeh> well that broke boot ;)
[13:46] <RoyK> hehe
[13:47]  * RoyK wonders why
[13:47] <trippeh> no wait, it came up, but something is wonky
[13:50] <trippeh> there we go
[13:50] <trippeh> virt bios is slowing the boot down ;)
[13:50] <RoyK> hehe
[13:50] <RoyK> kvm?
[13:50] <trippeh> yeah
[13:51] <RoyK> setting up 1504 in a kvm vm myself...
[13:51] <RoyK> (storage backend 7x2TB in RAIDz2)
[13:51] <trippeh> getting drm up takes like 0.1s too
[13:51] <RoyK> drm?
[13:51] <trippeh> I'm running off some NVMe storage. 450k iops :-)
[13:52] <trippeh> "graphics"
[13:52] <RoyK> ah
[13:52] <RoyK> I have a home server I use for this sort of things ;)
[13:52] <RoyK> that is - for everything
[13:52] <RoyK> cheap thing with 16 gigs of RAM and a bunch of drives
[13:53] <trippeh> could kill the graphics, hmm. kvm supports serial console ;)
[13:53] <RoyK> hehe
[13:53] <RoyK> we've been looking at Natinux at work for some stuff
[13:54] <RoyK> seems rather impressive
[13:55] <trippeh> link?
[13:56] <trippeh> I made systemd verbose but the frame rate is too low to show any of it, not even a flicker usually
[13:58] <RoyK> trippeh: typo http://www.nutanix.com/
[14:04] <RoyK> trippeh: after booting up, +/- buffers/cache tells me 56MB is used, only sshd installed
[14:05] <RoyK> and bootup is about the same as debian jessie
[14:05] <RoyK> bootup time
[14:05] <RoyK> pretty much the same, though, both systemd etc
[14:11] <trippeh> http://pastebin.com/raw.php?i=MnG5rPHs
[14:13] <RoyK> nice
[14:14] <RoyK> I'll have to test that when our openstack cluster comes up
[14:14] <trippeh> with this much ram, 64bit is important! ;)
[14:14] <RoyK> I need about 1k VMs for stresstesting things :)
[14:14] <RoyK> HAHA
[14:19] <trippeh> I feel very retro adding 64MB of swap
[14:19] <RoyK> :)
[14:22] <trippeh> the 1+ GB/s disk I/O is very not retro though
[14:23] <RoyK> hehe
[14:24] <RoyK> I remember once I started playing with linux back in '94, I was amazed it could handle 1MB/s if I setup DMA correctly
[14:33] <trippeh> seems systemd likes to have at least 64MB system memory, or else journald gets sad
[14:33] <trippeh> bloat!!
[14:33] <trippeh> ;)
[14:33] <trippeh> even though its not really using it
[14:55] <jelly> not knowing what happened when your system ran out of RAM is more problematic than allocating a bit in advance
[14:56] <trippeh> jelly: sure, I'm just goofing around
[14:59] <RoyK> jelly: not knowing how much memory your system really needs is worse ;)
[15:09] <RoyK> jelly: and since most things are virtualised these days, memory is usually the bottleneck on the host machines, not CPU, so you really *do* want to know if this or that VM can run on 256MB or if it needs 4GB
[18:12] <trippeh> [    1.243431] tsc: Refined TSC clocksource calibration: 3499.940 MHz
[18:12] <trippeh> [ 8783.366555] random: nonblocking pool is initialized
[18:12] <trippeh> that took a while :P
[18:37] <trippeh> RoyK: bootloader handoff to login in 0.4s :)
[18:38] <trippeh> didnt need all those "usb controllers" ;)
[18:41] <RoyK> hehe
[18:42] <trippeh> now if I could just boot KVM without BIOS emulation
[18:42] <trippeh> spends like half a second in there ;)
[18:43] <trippeh> hmm. maybe ovmf is faster
[18:43] <trippeh> (uefi firmware for kvm)
[18:46] <RoyK> I beleive you can - there are choices for direct boot
[18:46] <RoyK> but then you can't upgrade the kernel from the guest
[18:50] <trippeh> or the initramfs, looks like
[18:51] <trippeh> may be possible to load grub directly perhaps, like grub2 pv xen mode
[18:57] <trippeh> ah, it is. bunch of ceveats tho
[19:33] <RoyK> trippeh: guess you'll have to live with that awful 400ms delay - sorry about it
[19:33] <RoyK> trippeh: imagine the productivity cost, worldwide, if every reboot was 400ms shorter!
[19:34] <trippeh> ;-)
[19:34] <RoyK> thinking like an economist on bad drugs is sometimes rather fun ;)
[19:34] <trippeh> I'm going to need a separate non-virtio /boot looks like
[19:35] <RoyK> that shouldn't be a problem, though
[19:35] <RoyK> just an ext2 fs or something
[19:36] <trippeh> thats fine for this "experiment", but bit of a kludge otherwise
[19:38] <RoyK> perhaps a new standard comes up one day to do it better - but again - but you know how it goes ... https://xkcd.com/927/
[19:38] <trippeh> plz someone just commit virtio support to grub2 ;)
[19:38] <trippeh> it supports "everything" already so whats one more thing! ;)
[19:45] <RoyK> trippeh: perhaps try #grub ;)
[20:18] <LeMike> err. do I need to do a daily backup before rsnapshot weekly ? I mean on the same day
[20:45] <Edtoast_46> hi