[01:49] has anybody noticed random high ports being open (or at least randomly responding) with no tcp listeners? [01:49] $ nmap localhost -p 0-65535 [01:49] 40424/tcp open unknown [01:49] 41820/tcp open unknown [01:50] ... [01:50] 59865/tcp open unknown [01:51] goofy... whatever is causing it (tested on fresh build/air gap) [01:52] and it only happens if a non priviledged user does a port scan [02:00] can anyone else confirm the same behaviour? as a non root user run nmap localhost -p 0-65535 [02:00] several times in a row [02:02] a1fa: any specific OS you want tested, or just any Ubuntu? [02:03] 15.04 or 14.04 [02:03] 14.04.02 [02:03] and 15.04 with 3.19.0-24-generic [02:03] stupid question but [02:03] you wouldn't happen to have dhcp on your network would you? [02:04] (dhclient might be listening on random high ports for DHCP changes and such) [02:06] i do [02:06] i can replicate the behavior on wily, but only insomuchas there's only one open port [02:06] a random high tcp port on my side [02:06] and only one [02:07] multiple UDP ports are 'listened' on by dhclient [02:08] odd [02:08] and i just saw one with a few high number random TCP ports open [02:08] but the next scan they're gone [02:09] it only happens with non-root user [02:09] which makes me think they're destructible connections - since high ports are usually used for client-initiated TCP [02:09] and nmap needs those too iirc [02:09] i did see S/Ack on tcpdump [02:10] so forwhatever reason, something answered, followed by RST [02:10] just tested the same thing on a chromebook, same behaviour [02:11] lemme start up my CentOS box [02:12] and see if it's replicated there too [02:12] my guess is it's standard stuff happening [02:14] doesn't happen for me at all, on 14.04.2 [02:14] you didn't happen to *run out* of available ports did you? [02:14] i'll spin up my ubuntu server vm to test in a moment [02:14] cycling through them too fast, in time-wait [02:15] hmm, atleast with *that* nmap command, they aren't going to time-wait, cause they aren't full tcp session test [02:15] patdk-lap: thats what i was thinking.. but i saw a s/ack [02:16] could be hundreds of reasons [02:16] depending on what you have installed [02:16] base [02:16] I know I only install minimal-installs so [02:16] don't know what base is [02:17] basically nothing besides what comes on a default install [02:17] ya, I never install that *much* [02:17] the odd thing is, its doing it on the chromebook too [02:18] running chromeos+14.04.02 [02:18] about 30, nmap runs now [02:18] and always the same 3 ports, of stuff I have running [02:18] ssh/http/munin [02:18] dhclient gets involved a little at some point [02:18] running as non root user, right? [02:18] centos can't replicate this [02:18] oh non root? that will ahve other things [02:19] what kernel version teward [02:19] a1fa: for...? [02:19] centos [02:19] centos? you're 30 seconds slow i already powered off the VM [02:19] :P [02:19] i'll tell you after i test on my Trusty VM [02:19] k [02:20] ok, as non-root [02:21] it happens, and also happens on rhel7 [02:21] its worth looking into it, and maybe opening a bug [02:21] why? [02:21] so far it's not worth it at all [02:22] it's not supposed to ack back at all [02:22] ran nmap 5 times and it started to show open ports locally [02:22] in rapid succession [02:22] otherwise it wasn't giving any ports (14.04) [02:22] nmap just acts funky when you don't use it as root [02:22] mhm [02:22] cause it can't use raw sockets [02:23] so it attempts to do the best it can [02:23] patdk-lap: but the system responds back with s/ack on those prots [02:23] confirmed with tcpdump. on both loopback and actual interface ip [02:24] both use loopback interface [02:24] what happens if you use a remote ip? [02:24] no ports show up [02:25] i'll do some more work tomorrow to debug it [02:27] thanks for confirming [02:30] on a new install of Vivid server, my `systemd-analyze time` shows too much kernel time (>20s). Anyway to improve this? [02:30] ^ kernel time should be about 5s === markthomas|away is now known as markthomas [05:12] What are people using for centralized update management these days? [05:12] I assume Landscape is an option; are there FOSS things that are worth trying? === frickler_ is now known as frickler === markthomas is now known as markthomas|away [06:56] Good morning. [06:57] o/ [07:04] anyone ? [07:05] mobile3: ? [07:06] Can you help me with Webuzo Control Panel on VPS ? [07:06] Ubuntu 14.04 x64 [07:06] Ah, no. But do ask your questions. [07:06] * ObrienDave does not want to know what webuzo is ;P [07:08] ObrienDave: No adventure in the early morning? [07:08] getting close to sleepy time :) [07:08] I have been given 4 IP , while installing webuzo 1 IP got bonded to it... I have placed index.php file in public_html folder... when I am accessing via bonded IP , I am able to get correct index.php file but with all other 3 IP , I am getting different things.. [07:09] What to do so that I can get the same index.php file from all of my IP ?? [07:09] Vhost configuration? [07:10] easy to add an extra ip to your machine, manually .. http://askubuntu.com/a/313887 [07:10] mobile3: Who set up the Web server? [07:11] lordievader: they gave me with Ubuntu 14.04 x6 [07:11] minimal installed [07:11] I changed it to Ubuntu 14.04 x64 [07:12] mobile3: So you control the Web server config? [07:13] ya I can... [07:14] How to solve this problem ? [07:14] mobile3: Check the vhost configuration. [07:14] how ? [07:15] By reading the configuration/ [07:15] mobile3: Step 4 in this tutorial shows how vhosts are configured: https://www.digitalocean.com/community/tutorials/how-to-set-up-apache-virtual-hosts-on-ubuntu-14-04-lts [07:16] In that particular case it is any incoming request on port 80. [07:21] not such file /etc/apache2/sites-available/000-default.conf [07:21] in my server [07:22] mobile3: What is in /etc/apache2/sites-available? [07:22] No apache2 folder [07:25] mobile3: What Web server are you using to serve your control panel? [07:26] Ubuntu 14.04 x64 .... Webuzo Control Panel [07:26] mobile3: That is not a Web server. [07:27] Apache [07:28] lordievader : Apache Version : 2.2.29 [07:28] But you have no /etc/apache2? Where is your Apache coming from? [07:29] I don't know ! [07:30] mobile3: What is the output of 'apt-cache policy apache2'? [07:30] apache2: Installed: (none) Candidate: 2.4.7-1ubuntu4.4 Version table: 2.4.7-1ubuntu4.4 0 500 http://archive.ubuntu.com/ubuntu/ trusty-updates/main amd64 Packages 500 http://security.ubuntu.com/ubuntu/ trusty-security/main amd64 Packages 2.4.7-1ubuntu4 0 500 http://archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages [07:31] For in the future [07:31] !paste | mobile3 [07:31] mobile3: For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://imgur.com/ !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [07:32] mobile3: That is not installed... Is Apache coming from outside of the repo? [07:33] I don't know... [07:34] mobile3: What is the output of 'whereis apache2'? [07:34] whereis apache2 apache2: [07:37] Huh? How did you just get the version information? [07:44] lordievader :http://www.awesomescreenshot.com/image/436540/dcbc63865c203be5ee4bce22fa8caea7 [07:45] Ugh, does that thing do its own software installation? [07:45] I am upgrading 10.04 to 12.04 but middle of upgrade my system reboots.. [07:46] When checked system logs it states that "locale: /lib/libc.so.6: version `GLIBC_2.15' not found (required by locale)" [07:46] My present glibc version is 2.11.. [07:46] any suggetions? [07:47] lordievader : may be but I don't know [07:50] neonixcoder: verify versions of libc6, libc-bin and locales packages. Make sure they match by installing and/or configuring them manually if needed, then finish your release upgrade. [07:51] neonixcoder: newer versions of .deb package files will have been downloaded into /var/cache/apt/archives/ [07:52] ok? [07:52] Thanks for your reply.. [07:54] jelly: when I checked with apt-cache policy for libc6 and libc-bin it shows installed is 2.11 and candidate is 2.15. But for locales it shows installed and candidate as 2.11 [07:54] Any thoughts on this? [07:54] lordievader : are you there ? [07:55] Hey guys, if I nslookup my zone 'test.co.uk' and it returns the server as a loopback does this mean I've configured something wrong? [07:55] neonixcoder: both of those are unusual for 12.04. My 12.04 installation has locales: Installed: 2.13+git20120306-3 Candidate: 2.13+git20120306-3 [07:55] mobile3: Still here, just busy. Anyhow I get the feeling it does install its own packages. I have no idea how, why or where. So I really cannot help you. [07:56] jelly: I am still on 10.04 machine.. [07:56] lordievader : okay [07:56] I am upgrading from 10.04 to 12.04 [07:57] neonixcoder: Restart the upgrade? [07:59] neonixcoder: pastebin the output of "apt-cache policy" and "apt-cache policy libc6 libc-bin locales" please [07:59] lordievader and jelly: Let me give some background here.. [08:01] I have some 200 machines need to upgrade from 10.04 to 12.04(Intention is to upgrade to 14.04). I have some test machines where I can upgrade multiple times if something went wrong.. [08:01] I am going to follow http://askubuntu.com/questions/125742/why-did-my-upgrade-to-12-04-fail-with-glibc-not-found-or-libc6-or-requires [08:02] from there I install that PPA which is giving me 2.15 versions for libc6 and libc-bin [08:02] neonixcoder: oh dear [08:03] Somewhere that sounds like a bad idea. [08:03] ........PPA for libc6? *grabs popcorn* [08:03] neonixcoder: well, good luck with that, I won't be able to help with that mess [08:04] Ok.. Let me remove that ppa from list then.. [08:04] I did not try to upgrade it with that PPA, before doing it I want your valuable advice.. [08:05] neonixcoder: Did you install anything from the ppa? [08:05] right now no.. [08:05] Just added it.. [08:05] Phew, doged a bulled. [08:05] bullet* [08:05] * jelly would prefer to the valuable current state of the system before dispensing potentially harmful advice [08:06] to see* [08:16] jelly: My present policy list is http://pastebin.com/Lugf3j7N [08:16] any suggetions? [08:28] jelly:Any thoughts on how to proceed with upgrade? [08:33] Any suggestions guys? [08:33] How can I upgrade with out glibc issue? [08:49] Quick recap.. I removed ppa and given my apt-get cache stuff.. [08:52] Or can I know which repo provides glibc? [08:58] going home, ping you people tomorrow === [BNC]Savemech is now known as Savemech === vassie_ is now known as vassie === OliPicard_ is now known as OliPicard === justizin_ is now known as justizin === mwhudson_ is now known as mwhudson === yofel_ is now known as yofel [11:23] Hey guys - What's the command to read my bind9 error logs, can't seem to find it anyway? [11:23] Amillo: tail -f /var/log/bind9/* ? [11:24] Amillo: tail -f /var/log/syslog ? [11:24] Amillo: Same place most of the other logs are [11:25] my bind9 is failing to restart but my error message doesn't make sense to me [11:26] isc_stdio_option '/var/log/query.log' failed: permission denied [11:26] but I've changed the ownership of that file to bind [11:26] Also has configuring logging: permission denied [11:46] '/var/log/query.log' sounds wrong to me. [11:55] momomo: if all you want are security updates, remove the lines in /etc/apt/sources.list that have '-updates' and then do 'apt-get update'. you will only get updates from the security pocket === Dan_ is now known as Guest86431 [13:09] Hey [13:12] I'm having what I believe might be an issue? - When I do an nslookup from both my slave and master of my zones they are turning loopback addresses as the server - http://i.imgur.com/UZq0Mk9.jpg is this normal? (sorry for blocking stuff out - on a work network don't wanna mess anything up) === rcj` is now known as rcj [13:43] {stick around next time} === admcleodafk is now known as admcleod [14:47] Wow, tis weechat client integrates screenshot links. [14:47] *this [14:48] So, I'm wondering, what are people using for centralized update management these days? (It looks like Landscape is an option. but O [14:48] *I'm hoping for FOSS, and ideally a tool that'll work on deb & rpm distros [14:49] automatic updates [14:51] ^ that [14:51] but if you want centralized management of Ubuntu servers, it's Landscape (but that can be expensive) [14:51] or puppet or other multisystem management systems [14:52] Puppet is rather nice. [14:53] lordievader: when configured right, yes. [14:53] True, it can be painfull too ;) [14:54] landscape is up to 10 servers free, no ? [14:55] OerHeks: when you set it up on your own system, 10 physical machines, 10 virtual [14:55] OerHeks: i use it to manage my VPSes (which get listed on the Virtual category) [14:56] problem is you actually have to set it up, as the 'cloud' one from Canonical needs the Advantage licenses [14:56] Nice, it is indeed good usable for small business [14:56] OerHeks: provided there's only 10 physical and 10 virtual machines [14:56] Guys, how long should it take to wipe and reinstall a fresh Ubuntu server, only running Ubuntu 14.04 x64 on a machine with 16GB RAM? [14:58] garethdaine: Mostly depends on the storage and set of installed packages [14:58] garethdaine: writeback cache on the storage is a factor 10-100 [14:59] compared to slow spinning disks without [15:00] But it shouldn;t take 4 hours should it? [15:00] Being installed by professionals [15:00] I’ve asked our dedicated host to wipe the server and install only a fresh copy of Ubuntu 14.04 x64 and they are saying it’s 4 hours chargable work [15:01] garethdaine: they probably have to do things at the datacenter to make the revisions - network reconfiguration, configuration for networking, etc. [15:01] depending on what all they ahve to do over at the datacenter it could very well be 4 hours of work for them [15:03] teward: assuming that it had a previous version of Ubuntu server on there, what else would they need to reconfigure? [15:03] Hmmm, really? Thought it took like 30 minutes to install [15:03] 30 minutes within an 80% confidence interval maybe [15:04] (if the hardware isn't Ubuntu certified, etc) [15:04] If quoting in advance it makes sense to add contingency time, for example if there are driver issues. [15:04] Or if the hardware turns out to be faulty. [15:04] Etc. [15:04] rbasak: true, I suppose there are a lot of unknowns here [15:06] even with a dedicated host you should be able to put an image back yourself. [15:08] rbasak: Shouldn’t have faulty hardware, it’s current live server that’s just getting a reinstall [15:11] Disks can have latent errors that only show up when doing big things to them. [15:12] This is why double RAID failures happen. Because only on attempt to reconstruct does the second disk fail. [15:17] OK, thanks for the help folks. Should I question the quote I have received and ask them to detail what is taking the 4 hours, or should I just accept that it can take 4 hours? [15:18] garethdaine: i would, but i'm anal about making sure they're not doing extra work [15:19] I would consider what your service costs you overall and what you get for that money, as compared to competitors. [15:19] right [15:19] garethdaine: although, if they have to redo the RAID to reinstall that could potentially explain a large time for completion [15:19] (build the array, verify the array's integrity, etc.) [15:19] but i woudl do that comparison rbasak says as well [15:20] AND ask what the 4 hours entails [15:23] Yes, I will do that teward [15:25] Thanks for the help === markthomas|away is now known as markthomas [16:24] exit [16:24] exit [17:25] hey [17:26] just have some probs with my dns server [17:26] Garogat: 'some probs' is not descriptive [17:26] sry ;) [17:26] i got a domain [17:26] slucky.de [17:27] and i wanna use my own dns server [17:27] *servers [17:27] lets not forget that enter is not punctuation... ;) [17:27] Garogat: i'm followin you so far, so, what's the 'problems' [17:28] Garogat: https://help.ubuntu.com/lts/serverguide/dns.html may help? [17:28] ^ that too [17:30] fsg-preetz.net and c.ns.buddynd.com, but when im trying to test them with nast from denic its always gving me an error. It's says: inconsistent set if NS rps (IP, Ns host names) === ashleyd is now known as ashd [17:48] new vivd server & boot hangs for 15s before systemd starts: http://paste.ubuntu.net/11949526/ === Garogat_ is now known as Pommesgabel [18:43] I have issues with disks on a Marvell 88SE9230 controller, under load, especially when writing I suppose, failures occur [18:44] are there known issues with 88SE9230 and LTS kernel 3.13? [18:44] Seeing issues seems to be the norm with that controller, I experience similar ones [18:45] too bad... I had not used it so far, but I just recently added two disks [18:46] my board is a Supermicro X10SBA, could have just used a X10SBA-L I suppose [18:46] I'll try to find out if more recent kernels have patches [19:25] Uh, I've pushed myself into a corner here. My /boot is at 97%. I want to clean it up with apt-get autoremove, but I can't because a few packages (linux-image-extra-3.13.0-55-generic and -59-generic, as well as linux-image-generic) miss some dependencies (linux-image-3.13.0-55-generic and for the latter two -59-generic) and I am recommended to run apt-get -f install to fix these dependency issues. So I do this, and I am told that I cannot [19:25] install linux-image-3.13.0-59-generic_3.13.0-59.98_amd64.deb because the device is full. No space left on device. I then try apt-get autoremove to free some space, but I get the same errors as before. Running Ubuntu Server 14.04.2 LTS on amd64. Any help? [19:26] bilde2910: use apt-get purge to manually delete specific packages [19:27] Just tried that too, and nope, I can't because aforementioned dependencies are not met [19:27] bilde2910: start deleting the dependencies, too [19:27] Should I try using -f? Or is that dangerous? [19:27] bilde2910: you can always re-install the metapackages once you're done [19:28] I don't really have much experience with Linux and Ubuntu and servers, so I'm not exactly sure which packages should be present and which shouldn't [19:39] Is it safe to manually delete some of the older ones of these files? http://paste.ubuntu.com/11950111/ [19:43] bilde2910: yeah; if you truncate them instead of delete them, it'll make some deleting steps easier.. [19:44] bilde2910: be sure to keep at least two -- whatever you're currently running, and the newest. if you're running the newest, then keep the next-oldest.. [19:44] or next-newest [19:44] * sarnold stops thinking about that before his head hurts [19:45] anyone know if there's a way to assemble mdadm raid arrays in a particular order? [19:45] i have a raid60 made up of 2 raid6 arrays [19:46] anytime that i reboot this box, i need to manually assemble the raid0 array (the underlying raid6 arrays assemble just fine on boot without any intervention) [19:46] i could always throw something in /etc/rc.local or such, but i'm wondering if there is a cleaner way? [19:51] hexaclock: the arrays are assembled via udevd triggers, so how about a custom rule that assembles the RAiD0 once the 2 underlying arrays are ready [19:53] hexaclock: along the lines of: when a RAID6 is added: 1) if /run/raid6 already exists then assemble the RAID0 and delete /run/raid6, otherwise "touch /run/raid6" [19:54] awesome, i'll look into it [19:54] thanks! [19:56] hexaclock: "/lib/udev/rules.d/64-md-raid.rules" might give you some clues on how to recognise the mdadm array is ready; your rule will want to be in "/etc/udev/rules.d/" and have a different number/name than the system mdadm rule to avoid over-riding it [20:05] sarnold, another question: How do I know which of the files I use? E.g. I have 8 of initrd.img-$version-generic, how do I know which one is the one my system uses? [20:05] bilde2910: uname -a output [20:06] Ah, thank you :) [20:26] I have Trusty server running as a headless VM (on another Trusty server machine). A few times now, the VM has spontaneously booted into memtest86+. [20:27] Any idea why? === kickinz1 is now known as kickinz1|afk [20:41] tanuki: You need to investigate the logs in /var/log/ [20:48] Paradoxically, the thing that pisses me off is when things *don't* fail (when I want them to). [20:49] I dont want things to fail :) [20:49] tanuki: Is there anything non-standard about the server installation, any custom boot-loader configuration, anything unusual about the disk images being provided to the guest by the host? [20:50] TJ-: Not to my knowledge, with the exception of choosing "minimal virtual machine" instead of "minimal system". [20:50] bekks: I want things to fail when I need to know why they fail. [20:52] tanuki: there were 2 things I would focus on... is the root file-system image separate to the /boot/ image and if so is the rootfs image present and ready, and if they are, then has grub set a different 'default' boot entry that happens to be the memtest86+ entry in some circumstances.... the only other option I can think is something running in the guest that actively causes a reboot to memtest86 [20:54] This server isn't used for anything, really (basically, it's a development box for things), so I'm tempted to just set up a cron job on the host that checks that sshd is running on the guest and fires off an email otherwise [20:54] I'm more concerned with power consumption than anything else. === markthomas is now known as markthomas|away === markthomas|away is now known as markthomas [22:02] how to change the linux hostname? [22:07] jak2000: edit /etc/hostname and /etc/hosts, then reboot. can avoid a reboot with 'sudo sysctl -w kernel.hostname=' [22:08] the only place where a domain name enters into the mix is in /etc/hosts. everywhere else hostname is just the first part of the FQDN (up to the first dot) [22:09] ok thanks === spinza- is now known as spinza [23:09] Good day team.. [23:24] neonixcoder you worked with tomcat? [23:25] jak2000: Yes, not deep in to it.. [23:27] What's up? [23:29] check pm friend [23:29] pls