=== markthomas is now known as markthomas|away
=== trick is now known as Guest4137
=== JanC_ is now known as JanC
[04:03] <adun153> Hi, can any body help me? I have two nodes, both of them have the same hardware config, having SFP interfaces. Dmesg on both hardware shows that they both detect the hardware, but Node1 detects it properly as a network interface, Node2 does not. I took over administration of both these servers recently, and I just fresh installed Ubuntu 14.04.3 on Node2. Node 1 runs Ubuntu 14.04.3 as well, dist-upgraded from 12.04. Can any body point me in the r
[04:03] <adun153> ight direction?
[04:07] <mybalzitch> node 2 doesn't pick up the SFP at all?
[04:07] <mybalzitch> or does it say no carrier
[04:08] <adun153> It doesn't pick it up as a net interface
[04:08] <adun153> it does list it out in dmesg, though
[04:08] <mybalzitch> does it show up in lspci ?
[04:09] <adun153> yes, it does
[04:09] <mybalzitch> Hmm
[04:09] <mybalzitch> does it show up in ifconfig -a
[04:09] <adun153> /etc/udev/rules.d/70-persistent-net.rules only show the copper interfaces.
[04:09] <adun153> is that p1p1?
[04:09] <adun153> anyway, want me to do a paste of the output?
[04:10] <mybalzitch> paste it to pastebin
[04:10] <adun153> http://pastebin.com/Uix2zCvN
[04:11] <mybalzitch> can you do ethtool eth0
[04:12] <adun153> It's the other copper interface. http://pastebin.com/4aiFntJg
[04:13] <mybalzitch> oh
[04:13] <mybalzitch> I shoulda noticed that on the mac addy
[04:14] <mybalzitch> do you know the mac address of this sfp?
[04:14] <adun153> Doing "dmesg | grep -i sfp" on each  node gives me the following: http://pastebin.com/R5F4ZMYU
[04:14] <adun153> I don't.
[04:14] <adun153> ItBased on the output, I think Node2's kernel doesn't recognize the H/W as net interfaces.
[04:14] <mybalzitch> yeah
[04:14] <adun153> Is this a driver issue?
[04:15] <mybalzitch> it might be
[04:15] <mybalzitch> have you tried compiling an ixgbe driver from intel from source/
[04:15] <adun153> If yes, how do I find out which friver to install?
[04:15] <adun153> No, haven't.
[04:15] <adun153> But I 've compiled Squid before, so if it is at all easier or just about the same level of difficulty, I'm up for it.
[04:15] <adun153> Are there no packages in the main repos for it?
[04:16] <mybalzitch> about the same level
[04:16] <mybalzitch> not that I'm aware of
[04:18] <mybalzitch> http://ask.xmodulo.com/download-install-ixgbe-driver-ubuntu-debian.html
[04:18] <adun153> lspci shows this: Ethernet controller: Intel Corporation 82599ES 10-Gigabit SFI/SFP+ Network Connection
[04:18] <adun153> Intel website for the H/W is this: http://ark.intel.com/products/41282/Intel-82599ES-10-Gigabit-Ethernet-Controller
[04:18] <adun153> I click on "Software Downloads" on the right side
[04:19] <adun153> On the new page, Linux drivers are on the 4th entry on the list
[04:19] <adun153> It's apparently the ixgbe driver.
[04:19] <mybalzitch> yeah
[04:19] <mybalzitch> we're just tryiung a new ixgbe driver
[04:20] <mybalzitch> (newer)
[04:20] <adun153> lsmod shows the ixgbe driver as loaded
[04:20] <adun153> oh, ok
[04:20] <adun153> alight
[04:20] <adun153> **alright
[04:20] <adun153> I'll read the link.
[04:20] <adun153> Thanks for the advice!
[04:21] <mybalzitch> you could also
[04:21] <mybalzitch> and it might be easier
[04:21] <mybalzitch> just swap in a 4.1.4 kernel
[04:21] <mybalzitch> and see if it gets detected then
[04:22] <mybalzitch> http://www.yourownlinux.com/2015/08/how-to-install-linux-kernel-4-1-4-in-linux.html
[04:22] <mybalzitch> except the url is different I think
[04:22] <mybalzitch> nope
[04:22] <mybalzitch> url works
[04:24] <mybalzitch> it's just very odd, that they both seem to have the same part number but the other one isn't detected
[04:24] <adun153> mybalzitch: How different would this be from stock Ubuntu kernels?
[04:25] <mybalzitch> I've put that kernel in place of the stock ubuntu kernel with seemingly no ill effects
[04:25] <mybalzitch> worst case scenario is it doesn't boot, and you revert to your stock kernel
[04:25] <mybalzitch> have you tried swapping the HW to see if the nic is defective?
[04:26] <adun153> Which do you think I should try first, new driver or new kernel?
[04:26] <adun153> No, I haven't. It's a bit tricky to do that, since all other Nodes are now in prod, and co-located in a remote DC.
[04:26] <mybalzitch> oh
[04:27] <mybalzitch> I'd try kernel first
[04:27] <mybalzitch> its the easiest
[04:27] <mybalzitch> you do have IPMI right?
[04:27] <adun153> I do
[04:27] <mybalzitch> or iLOM2
[04:27] <mybalzitch> okay good
[04:27] <adun153> okay, trying the kernel route. :)
[04:29] <adun153> GRUB should automatically boot the latest kernel, right?
[04:29] <mybalzitch> yes
[04:29] <adun153> (y)
[04:29] <mybalzitch> call update-grub just to be sure
[04:29] <mybalzitch> but I think it gets called during the install process
[04:30] <adun153> Kernel packages intalled, grub updated manually, now rebooting, keeping fingers crossed.
[04:31] <mybalzitch> :D
[04:31]  * mybalzitch waits for smoke
[04:31] <mybalzitch> FIRE!!!
[04:31] <adun153> At least I'll have something to report to the  kernel team mailing list: Installed 4.14, server fried. FIX IT!
[04:32] <mybalzitch> HA
[04:36] <adun153> mybalzitch: Negative :(   http://pastebin.com/VzQHBgNc
[04:36] <mybalzitch> can you dmesg|grep ixgbe
[04:37] <adun153> http://pastebin.com/c0FHcbtH
[04:37] <adun153> wait, so does that mean that the p1* and p3 interfaces are the SFPs?
[04:38] <mybalzitch> looks like it
[04:38] <mybalzitch> when I looked up those p2p interfaces it said they were supermicro, so I assumed they were onboard
[04:38] <adun153> I see
[04:39] <adun153> Check this out: http://pastebin.com/13tyL8LV
[04:39] <mybalzitch> that looks like a 10g card
[04:40] <adun153> Why does it not show up in 70-persistent-net.rules though? http://pastebin.com/i6zX5WDs
[04:40] <mybalzitch> I'm not sure
[04:40] <mybalzitch> wrong mac address maybe?
[04:41] <adun153> Would it helped if I show mac addresses from Node 1's DFP?
[04:41] <adun153> *SFP
[04:41] <mybalzitch> no
[04:41] <mybalzitch> well
[04:41] <mybalzitch> see if its vastly different
[04:44] <adun153> Similar: Node 1- 00:25:90:c3:da:d6  ---- Node 2- 00:25:90:c3:cc:3a
[04:44] <adun153> http://pastebin.com/XASpPEZG
[04:47] <adun153> mybalzitch: opinion/suggestions?
[05:22] <mybalzitch> sorry afk
[05:22] <mybalzitch> adun153: check all those p*p interfaces for one thats got a link
[05:23] <adun153> That's all right. There are no SLA's for IRC. :)              all of them are "Link Detected: no"
[05:24] <adun153> I think that I should try configuring one of them, just to see if the link will go up if I "ifup" it.
[05:27] <mybalzitch> yeah
[05:27] <adun153> mybalzitch: also, I'd like to removev the 4.14 kernel, is it alright to uninstall the 4.14 packages while the 4.14 kernel is running?
[05:27] <mybalzitch> that would be my next suggestion
[05:27] <mybalzitch> yes it is
[05:27] <adun153> ok
[05:28] <adun153> I'll remove the 4.14 kernel first, reboot, then try to get the p1* interfaces working
[05:28] <adun153> ty
[05:28] <mybalzitch> :D
[05:31] <adun153> mybalzitch: Got this: http://pastebin.com/QG27w5pr should I be concerned?
[05:32] <mybalzitch> you can just link /vmlinuz and /initrd.img back to the /boot/ equivilants
[05:37] <adun153> mybalzitch: did I do the first two commands right? http://pastebin.com/D7NHvSza
[05:38] <mybalzitch> looks good to me
[05:39] <adun153> or should I have pointed the vmlinuz symlink to the "vmlinuz-3.19.0-25-generic.efi.signed" instead? Since this is an efi system
[05:41] <mybalzitch> I'm not sure
[05:41] <mybalzitch> grub shouldn't be using the / symlinks anyway
[05:41] <mybalzitch> it should be using /boot/..
[05:48] <adun153> anyway, it rebooted just fine, so no worries, hehe.
[05:48] <mybalzitch> woo
[05:50] <mybalzitch> adun153: any luck on bringing the interfaces up and having a link on them?
[05:51] <adun153> working on that now; if I get them to work, I'm going to rename them back to eth* :p   ---> http://www.hellovinoth.com/ubuntu-14-04-renaming-ethernet-interfaces-from-p1p1%E2%80%B3-to-eth0%E2%80%B3/
[05:52] <mybalzitch> :D
[05:55] <adun153> mybalzitch: Alright!!!! http://pastebin.com/gxWVv9XM
[05:55] <adun153> Thanks so much for helping me diagnose and troubleshoot, etc.
[05:55] <adun153> I learned a lot.
[05:56] <mybalzitch> np
[05:56] <mybalzitch> glad we got it working
[05:57] <mybalzitch> adun153: don't forget to make the changes to the network config files so it persists between reboots
[05:57] <adun153> yes, I'll write it to 70-persistent-net.rules
[05:57] <adun153> Thanks again! :)
[05:57] <mybalzitch> np
[05:57] <mybalzitch> I'm surprised your previous admin never fixed this, lol
[05:58] <mybalzitch> if you ahve any other fun issues, let me know
[05:58] <adun153> alright, I will
[05:58] <adun153> "fun", heh
[05:59] <mybalzitch> XD
[06:26] <lordievader> Good morning.
=== Guest81512 is now known as IdleOne
=== beyondsight is now known as Guest77465
=== beyondsight is now known as Guest42907
[08:46] <Adri2000> hello
[08:47] <Adri2000> any plan to update python-openstackclient to 1.6.0 in wily?
[11:10] <rbasak> frediz: any news on an update kimchi package please? It's feature freeze today, and I'm happy to upload with just the conffile /usr/share/doc symlink issue fixed. After today we'll need to get an exception which will be a pain.
[11:10] <rbasak> (sorry again I know it's me being backlogged that has caused this urgency)
[11:13] <frediz> rbasak: ok I was trying to sort out everything today, but that must be a bad plan :) . I've changed ln to cp , so I'm going to reupload kimchi so that you can upload and I'll deal with the rest and Gianfranco's comments after. Ok ?
[12:18] <rbasak> frediz: sounds good. Thanks!
[12:57] <frediz> rbasak: thanks again Robie; other changes will come next week probably as I won't be able to work on that till then.
[12:58] <rbasak> frediz: OK. BTW, "or" in copyright file as commented in the bug is wrong AFAIK. In this case, the licences are combined and have an "and" relation and I think are compatible so that's fine.
[12:59] <rbasak> It's effectively the same as LGPL but with mandatory copyright and license notices from the other licenses included. So I think what you did in debian/copyright for that is fine.
[12:59] <rbasak> It might be worth getting an opinion from somewhere else though.
[12:59] <rbasak> (before you change it)
[12:59] <rbasak> I am still happy to upload as-is.
[13:01] <frediz> rbasak: sure ; I'm no licensing expert *at all* and all of this as been done some time ago, so I need first to recheck files, copyright file to be able to understand :) . But I remember having got inspiration from another package already in Debian.
[13:43] <roaksoax> /win/win 9
=== markthomas|away is now known as markthomas
=== soahccc_ is now known as soahccc
[17:02] <bulldawgden> how do you make ubuntu server use IPv4 instead of IPv6?
[17:02] <bulldawgden> I am running it in Virtualbox
[17:02] <jrwren> bulldawgden: IETF says ipv6 should always be prefered in a dual stack configuration.
[17:03] <jrwren> bulldawgden: you can turn off ipv6
[17:04] <bulldawgden> is it easy to turn off?
[17:06] <RoyK> bulldawgden: really, why?
[17:07] <RoyK> bulldawgden: but still, if you want to do this, first hit on google http://www.binarytides.com/disable-ipv6-ubuntu/
[17:07] <jrwren> bulldawgden: I'm curious too, why?
[17:10] <TJ-> bulldawgden: see https://help.ubuntu.com/community/WebBrowsingSlowIPv6IPv4
[17:13] <RoyK> imho if you've got bad hardware not doing IPv6 right, replace the hardware :P
[17:13] <jrwren> RoyK: me too, I love my ipv6 delegated prefix :)
[17:13] <RoyK> IPv4 isn't going anyware
[17:13] <RoyK> https://xkcd.com/865/ :D
[17:17] <tackyone> Hello, I have run into a bit of an issue. I was doing a upgrade-manager -d command, and midway through my network connection cut out. I logged into the machine physically, but didn't see any processes still running that would indicate that the server was still upgrading, so I rebooted the machine, as I couldn't do much with the machine. After restarting the machine, it now kernel panics on bootup, and I can not access the mach
[17:17] <tackyone> Which I am currently in the process of making). Could someone help me get my server back on it's feet? I took a backup using fwbackups of /, so if I need to, I can restore as much as possible for the OS drive with that.
[17:18] <linocisco> hi all
[17:20] <linocisco> I am going to install ubuntu server 14.04-2 64bit on existing Elastix server which is CentOS based and take up the whole HDD when installed, How can I add ubuntu server installation along side of it and make dual boot.?
[17:20] <bulldawgden> thanks
=== t3p0 is now known as crazyagile
[18:35] <linocisco> hi
[19:02] <SCHAAP137> allright, that 13.10 to 14.04.3 server upgrade went exceptionally smooth
[19:07] <RoyK> SCHAAP137: nice
[19:20] <JanC> I still need to upgrade a 12.04 server to 14.04  :-/
[19:21] <SCHAAP137> 12.04, oldschool
[19:22] <JanC> SCHAAP137: it used to be 10.04 before 12.04.01 was around  :)
[19:26] <lordievader> JanC: Is it still running on the original hardware?
[19:26] <RoyK> JanC: I had an 8.04 server around until a few months ago :P
[19:29] <lordievader> RoyK: We still do....
[19:29]  * lordievader runs
[19:33] <JanC> lordievader: it's a VM actually, and not on the same hardware after we moved everything to a different dedi server  :)
[19:33] <RoyK> lordievader: ouch
[19:33] <lordievader> We're working on a replacement, luckily.
[19:34] <lordievader> JanC: Ah, I see.
=== cipi is now known as CiPi
[19:34] <JanC> (and the reason I didn't migrate yet is that I didn't check all software on it for compatibility)
[19:34] <RoyK> lordievader: this one 8.04 machine was badly messed up - stuck at 8.04.1 and unable to upgrade until it was finally killed
[19:35] <JanC> mail server might need some configuration changes etc. (the usual stuff, I guess)
[19:35] <lordievader> We run qmail on it. No one knows how it works and everyone is afraid to touch it.
[19:35] <JanC> *ieuw*
[19:36] <lordievader> Yes.
[19:36] <lordievader> Like I said, we are working on a replacement.
[19:36] <JanC> do you know what it does, at least?  :)
[19:36] <lordievader> That we do, that is why no one wants to touch it ;)
[19:36] <JanC> well, it's also a lot easier to replace when you do know exactly what it does
=== markthomas is now known as markthomas|away
[21:16] <arooni-mobile> hey folks.  is there a compelling reason to move from 12.04 to 14.04 LTS?
[21:17] <RoyK> arooni-mobile: just if you need the new shit
[21:17] <arooni-mobile> i think id prefer doing that... so that way if packages work on my dev box they can work on my production server
[21:17] <arooni-mobile> will i have to reconfigure all kinds of packages, ruby, etc, on the server
[21:18] <arooni-mobile> if moving to 14.04 lts from 12.04?
[21:20] <teward> arooni-mobile: you *shouldn't* have to, however, take a backup of your system in its entirety on 12.04 so you can restore if it fails
[21:20] <teward> (there's an upgrade path for packages, etc.)
[21:21] <teward> (you may need to install one or two but...)
[21:21] <arooni-mobile> good advice
[21:21] <arooni-mobile> what method would you recommend for taking a backup
[21:21] <teward> external USB drive, flash drive with clonezilla, clone/image the entire server if possible
[21:22] <teward> if it's a VPS, contact the host to see if they can take a backup and keep it for a few days
[21:23] <arooni-mobile> its a vps;  i'll ask them ; good thought
[21:34] <teward> arooni-mobile: if they can't do the backup, or won't, then you may want to just not upgrade
[21:34] <teward> or backup critical data
[21:34] <teward> and get a list of installed packages
[21:34] <teward> and then reinstall the VPS with 12.04 if necessary
[21:34] <arooni-mobile> im checking what i already have backed up
[21:35] <arooni-mobile> also is there a way to make running & installing security packages an automatic part
[21:35] <arooni-mobile> automatic task, via the command line
=== markthomas|away is now known as markthomas
[22:20] <AEL-H> Is there a way I can prevent a user for snooping in other users files, even with ls -all?
=== medz is now known as samsn
[22:51] <tonyyarusso> AEL-H: The "execute" (x) permission is used in the case of directories for listing rights.
[23:11] <xedniv> is it possible to install/use parted in the ubuntu server cd installer shell?
[23:11] <mybalzitch> parted doesn't ship with the cd?
[23:12] <xedniv> nope it doesnt seem like it
[23:12] <xedniv> i need to create an encrypted lvm and wish to do so manually, so i tried the shell and parted isnt there
[23:12] <xedniv> the installer has never worked well for me to create an encrypted LVM system
[23:13] <xedniv> ex. /boot not encrypted + LVM volume encrypted, with /, etc inside
[23:36] <JanC> xedniv: I'm pretty sure parted is on the server CD, but in any case: parted is a partitioning tool, not a volume management tool...
[23:37] <xedniv> JanC: not, it is not. im not making some wild claim, heh.
[23:37] <xedniv> easy steps: grab server cd, enable expert mode, jump into shell after configuring keyboard layout, host and user.
[23:38] <JanC> it certainly was on the server images in the past
[23:38] <xedniv> classic response, no offense. well, pilgrim, im talking present time, not past tense.
[23:38] <JanC> but like I said: it's not useful if you want to manage volumes...
[23:39] <xedniv> ' i need to create an encrypted lvm and wish to do so manually, so i tried the shell and
[23:39] <xedniv> etc
[23:39] <xedniv> and then i explicitly mentioned the manual process.
[23:39] <xedniv> so, what was the part you were helping with? sure, maybe lvm utilities are there
[23:39] <xedniv> but i cant partition with those.
[23:40] <JanC> if parted isn't there, then sure fdisk or such will be there
[23:42] <xedniv> fdisk doesnt like gpt
[23:43] <JanC> GNU fdisk uses libparted
[23:43] <JanC> so GPT should be fine with it
[23:43] <JanC> (and parted is a dependency of 'ubuntu-standard', so I'd be surprised if it's not on the Server image...)
[23:44] <xedniv> i will check. and no man, stop insisting im pulling that claim out of my arse and go verify it already
[23:44] <xedniv> im happy to tell people how to build a static parted actually