[00:00] RoyK: is 24 hours enough? [01:17] DzAirmaX: Any time I've had bad memory, it's taken less than 24 hours to find out. But the main thing is the passes; once memtest has run all its tests more than once or twice you *probably* have the result (which, depending on the hardware in question, might take not much time at all) === JanC is now known as Guest35472 === JanC_ is now known as JanC [01:18] At least with all the RAM issues I've ever had, it's been very reliable. That is to say, the moment even the first test gets to a bad sector the errors start piling up. [02:20] DzAirmaX: normally you just run it until it's done with all passes - older versions of memtest ran in a loop, I don't think the current ones do that anymore [02:21] ok boyz [02:21] its not the ram [02:21] I have 2 passes [02:21] done 2 passes like 2 cycles [02:22] it doesn't loop any more? aww [02:22] all the passes? [02:22] sarnold: last I checked, no [02:22] if you're in a hurry try manually selecting one of the 'random' passes, like #7 or something, I think that one has found most of my memory faults [02:22] Royk: you boot on it and run the test, then it loops [02:24] DzAirmaX: it goes through something like 7 or 8 tests phases, but at least last I ran it, it didn't start over after the last test [02:24] you should at least let it run through the first four stages [02:25] hmm [02:25] I did [02:25] its last hours right? [02:25] 6 hours [02:25] makes sense [02:26] well, probably not memory, then ;) [02:26] is this intel? [02:26] or amd [02:26] yep [02:26] intel [02:26] do you have mcelog installed? [02:26] no I have rsyslog and kdump [02:26] nothing shows [02:26] install mcelog [02:26] machine check exception log [02:27] indicating hardware failure [02:30] RoyK: Family 6 Model 5e CPU: only decoding architectural errors [02:30] Royk: we good? [02:31] it'll log and email you errors if they are discovered [02:31] (that is, email you if you have configured the mta correctly) [02:33] DzAirmaX: postfix or exim or sendmail or qmail or whatever obscurure thing you're using [02:33] yeah [02:33] all configured and working [02:33] * RoyK prefers postfix [02:33] I just checked the nano /etc/mcelog/mcelog.conf [02:34] and its looking good ,) [02:35] ok, then it's just wait and see [02:35] sometimes it's hard to provoke an error [02:36] the old way, which I beleive should work well still, is to recompile a kernel, using all cores [02:36] it should stress pretty much everything on the system [02:36] the system crash after 24 hours, I just have to be patient [02:37] try to compile a kernel [02:37] kernel compiles go too quickly these days [02:37] it'll take some time and if you use make -j , it'll stress pretty much everything [02:38] sarnold: guess that depends on how many modules you're adding [02:38] sarnold: it's still a nice way to stress the system, at least for a wee wile [02:38] sarnold: it's still a nice way to stress the system, at least for a wee while [02:38] RoyK: what is the command for achieving what you are asking for? [02:39] wget https://cdn.kernel.org/pub/linux/kernel/v4.x/testing/linux-4.8-rc6.tar.xz ; wget https://cdn.kernel.org/pub/linux/kernel/v4.x/testing/linux-4.8-rc6.tar.sign ; gpg linux-4.8-rc6.tar.sign && tar xf linux-4.8-rc6.tar.xv [02:39] cd linux-4.8* ; make defconfig ; make -j 60 [02:39] or .xz [02:40] something like that [02:40] sarnold: make oldconfig, perhaps [02:40] sarnold: cp /boot/config-something .config [02:40] make oldconfig [02:40] or make localmodconfig? [02:40] yah [02:45] is it going to impact the actual kernel? [02:45] no [02:45] you're just compiling it :) don't go to any effort to install it, or the modules [02:47] Please enter name of data file: [02:47] after the gpg linux-4.8-rc6.tar.sign && tar xf linux-4.8-rc6.tar.xv [02:48] xz, not xv [02:48] sarnold is probably drunk [02:48] good idea [02:48] tar xaf [02:48] tar xf [02:48] same request [02:49] DzAirmaX: use tab completion [02:49] tar xz linux-5 [02:49] (that is *press the tab key*) [02:49] sarnold: asking a newbie to compile a kernel can sometimes be bothersome :P [02:50] RoyK: it doesn't help that I'm a very lazy person and didn't try it first [02:50] well [02:50] I am maybe a newbie [02:50] but there is definitively wrong in the command line [02:54] well, forget about the gpg sign part [02:55] just unpack the tarball and cd into the dir and cp /boot/config-whatever .config and run make oldconfig [02:56] the then run make -j $(lscpu |awk '/^CPU.s.:/ { print $2 }') [02:57] the then run make -j $(lscpu |awk '/^CPU.s.:/ { print $2+1 }') [02:57] perhaps [02:57] just to stress it a bit extra [02:57] oh, that's the trouble. silly kernel devs didn't sign the .tar.xz, but the plain .tar. [02:57] ah [02:57] sarnold: shouldn't matter much, though [02:57] RoyK: btw man 1 nproc [02:58] ah - didn't know that one :) [02:58] RoyK: it depends on how much you like sending untrusted data through decompression programs [02:58] the then run make -j $(( $( nproc ) + 1 )) [02:59] sarnold: satisfied? ;) [02:59] lol [02:59] <3 [02:59] thanks boyz [03:01] noob question again [03:03] There is no such thing as a dumb question, only stupid people ;) [03:03] well [03:03] lol [03:03] RoyK: what are you doing up this early/late anyway? :) [03:03] couldn't sleep [03:03] :( sorry to hear it [03:04] I have a week off and gf is snoring like a small volcano [03:04] week off is nice though :) [03:04] make -j $(( $( nproc ) + 1 )) => I ahve to confirm every line [03:05] which is not intended I guess [03:05] DzAirmaX: make oldconfig first [03:05] DzAirmaX: that is [03:05] first [03:05] cp /boot/config-4.4.0-31-generic (or whatever it's called at your place) to .config [03:05] DzAirmaX: when you're running 'make oldconfig' just hold down enter for a while [03:05] as in cp /boot/config-4.4.0-31-generic .config [03:05] then run make oldconfig [03:05] ok done [03:06] thanks for the compliment by the way, always appreciate the smart comment [03:06] DzAirmaX: the part 'stupid people' goes to both sides, both to whoever asks and who answers ;) [03:06] RoyK: no worries br0 [03:07] just my little pun [03:07] scripts/extract-cert.c:21:25: fatal error: openssl/bio.h: No such file or directory [03:07] I guess this is not my night [03:07] the server is going to crash anyways xD [03:07] stress or not stress [03:07] heh [03:07] apt-get install libssl-dev [03:08] then try again :/ whackamole to find them all.. [03:08] apt-get install build-essential # too [03:08] heh yes get that one out of the way quickly :) [03:08] you were right [03:08] it's a good stress [03:09] we will see how long it can handle it [03:09] {standard input}: Fatal error: can't close drivers/net/ethernet/intel/i40e/.tmp_i40e_main.o: No space left on device [03:09] hah [03:09] hehe [03:10] tune2fs -m0, perhaps ;) === tikun is now known as sikun [03:14] nah, more like "don't build in a tmpfs" [03:14] also [03:14] hehe [03:14] running mosh / tmux / ssh / screen / and -then- running yes "" inside is a good way to piss off a pandaboard [03:14] 100% cpu use, I had to kill mosh-server [03:15] poor panda [03:15] endangered species - be nice [03:20] well [03:21] How long shoud I stress it? [03:22] good question [03:22] normally a few cycles of this will find memory faults, if you have them, and if it uses all your RAM in the process.. [03:23] ok [03:23] thanks for your help boyz [03:23] going to let it work for the night [03:23] ++ [03:24] good plan :) just don't forget to -look- at that terminal in the morning before using it :) [03:24] goodnight :) [04:59] Im running Ubuntu server 14.04, I had about 100days uptime, the server reset itself for some reason, now when I try to run 'zpool status' Im getting Kernel modules are not loaded: http://codepad.org/yoLGvTJU [04:59] uname -r returns: 3.13.0-91-generic [05:00] if I try to start a vagrant box, I also get "WARNING: The vboxdrv kernel module is not loaded" [05:02] If I try to start these modules with modprobe I get: modprobe: FATAL: Module zfs not found. [05:16] Could this kernel be the problem? Should I try somethign like "sudo apt-get purge linux-image-3.13.0-91-generic" [06:29] Good morning [06:40] jeffrey3234: You have booted on a new kernel and certainly you don't have the zfs module and vboxdrv not compiled for this kernel version [06:54] baptistemm What do you suggest? [06:54] just reinstall these 2 modules? === skylite_ is now known as skylite [08:51] hi, is it possible to create a samba shared folder without editing smb.conf? [08:53] ah, net usershare [09:13] ddellav, coreycb: the ryu update fixed the neutron-gateway n-ovs-agent hang for me [09:13] I've pushed that into newton-proposed in the uCA [09:14] jamespage i figured out my issue was dependency based: neutron-lbaas-agent : Depends: neutron-lbaas-common (= 2:9.0.0~b2-0ubuntu1~cloud0) but it is not going to be installed [09:14] ddellav, erm [09:14] ddellav, yeah - you need to update your neutron-gateway charm from the master branch [09:15] ddellav, we have a slight issue that orphaned binary packages are not removed from the uCA [09:15] so you can still try to install neutron-lbaas-agent [09:15] in the UCA [09:15] but it won't due to mismatching -common [09:16] I'm using the o-c-t latest next.yaml, so i need to pull neutron-gateway manually then [09:20] ddellav, no [09:20] but you must run juju-deployer with the right cli args [09:20] I'm not sure that by default it will automatically pull new charm versions from git repos [09:21] well my normal process is to delete the downloaded local charms before i run the command, then i do juju-deployer --bootstrap -c next.yaml -d xenial-newton [09:21] i'll double check and make sure i deleted the right folder [09:34] ddellav, yeah I do the same [09:37] jamespage ok i re-deleted the xenial directory and ran it again, got farther this time but now It's failing here: http://paste.ubuntu.com/23172840/ [09:37] ddellav, bzr pull on o-c-t [09:38] k [09:40] ddellav, coreycb: I'm happy with ovs+dpdk - those tested ok for me, so I'll backport those to newton-staging for more exposure [09:40] libvirt/qemu looks OK, but undecided as to whether we really need them [09:40] they are always the pain point for security issues [09:41] jamespage: did you test the 16.07+2.6 combo updating your charm? [09:41] cpaelzer, not with dpdk, but vanilla is just fine [09:41] cpaelzer, I still don't have any real hardware to test on [09:41] as what user does Apache2 run? [09:42] jamespage: ok, I postpone running it ovs-dpdk on yakkety on a daily base for the last 8 workday or so :-/ [09:42] jamespage: I hope I can feed back what on the config changed to you so you can adopt in your charm [09:42] whenever I get to it [09:42] sure - we can do that at any point [09:42] cpaelzer, tbh I suspect we don't have fast amounts of users of that feature yet anywya [09:49] What is segments in lvdisplay [10:06] bipul: a segment is a contiguous allocation of space on a pv [10:08] cpaelzer, So if i get segment = 2 that shows what? [10:09] PV physical volume is nothing but collection of P.E which is equal to 4 MB default. P.E= Size of partition in MB/4 [10:11] 1 Physical extent = 4M, And Total P.E = Size of single Partition in MB/ 4* [10:12] cpaelzer, congrats btw [10:13] * jamespage looks forward to no longer uploading cpaelzer work :-) [10:14] hehe, thanks - still I might ask for a review every now and then - so you might never get totally away :-) [10:17] cpaelzer, do we have s390x porters yet btw? [10:17] I wanted to try out the ovs test suite on powerpc and s390x (be's) to see if things have improved [10:17] testing powerpc now [10:19] coreycb, did the announce of b3 go out? [10:19] coreycb, ftr apparently I used to post to my blog as well which is syndicated to planet.openstack.org [10:19] you might want to setup something similar, or ask me todo that! [10:46] coreycb, posted on javacruft to get the syndication [10:56] keen to know if I can help in anyway to get https://github.com/mitchellh/vagrant/issues/7155 working.. I'm unsure as to why there's so much broken practice there [10:57] jamespage: there were no s390x porterboxes that I'd new of - but there might still be some [10:57] jamespage: if I can test something for you or lend you my machine let me know [10:57] cpaelzer, just drop the s390x exclusion from d/rules, and try build the package! [10:58] hehe, openvswitch as in yakkety? === iberezovskiy|off is now known as iberezovskiy [11:08] cpaelzer, yes [11:08] ddellav, coreycb: uploaded snapshot of ceilometer with fixes for autopkgtest failures... [11:09] I debated doing it on the released b3 version, but as we're coming up to rc I did not see that it would hurt. [11:13] it seems I'm locked out of the vlan :-/ going to is ... [11:14] ah no, I just tried to log on from the wrong system for 3 minutes :-/ [11:21] jamespage: currently building, do I understand it right that you are interested int he test results that will now be executed? [11:21] cpaelzer, yeah [11:22] I had four failures on powerpc [11:22] ok, I'll check the console later and let you know then [11:22] it just started tests, but those usually take a while [11:35] 1108 still no error ... [11:35] which ones failed on powerpc ? [11:45] 2223 just failed when I checked the console, still ongoing === Ussat is now known as _SJW_ [12:12] jamespage: 1159, 1165, 2223, 2237 failing - maybe the same four you had http://paste.ubuntu.com/23173348/ [12:35] cpaelzer, yes identical list [12:36] cpaelzer, I renabled all tests last night - today getting some races and failures [12:36] * jamespage sighs [12:44] :-/ [12:44] jamespage: on x86 you mean the transient ones we already had before? [12:44] or races and failures on powerpc [12:44] cpaelzer, yeah - I have a clean run on i386 armhf amd64 and arm64, so uploaded with just a single tests disabled on ppc64el [12:44] todays backport for the UCA has the old failures again [12:45] and now they showed up again :-/ [12:45] * cpaelzer joins the sigh'ing [12:45] I can only think that they really are racey, and when there is more load on the builders, we see these problems [12:45] yes, but that "just" means you have to continue masking them [12:45] I'm rather happy with just 4 errors on the be architectures [12:54] I somehow like Ubuntu as server OS more than my CentOS [12:55] FManTropyx: you are in the right chan to find a lot of ack's :-) [12:56] hahah, I thought that comment would probably be less successful on #centos [12:56] I just need to learn more, to better manage the system [12:57] FManTropyx: like managing many systems as one as an admin including deployment and such, or just managing a linux system in general? [12:57] just my tiny little VPS [13:19] hey guys I'm trying to install seedbox on my server. Is there any script which can do it fast? [13:22] elKosyniero: https://www.reddit.com/r/seedboxes/comments/3ntadw/best_tutorial_for_manually_setting_up_a_seedbox/ ? [13:26] cpaelzer: thanks. I'm looking for multi-user tutorial but these ones are not ;> [13:26] ;< [14:09] hello, all. i'm having a problem with a remote host accepting ssh connections. it seems like my computers ip got banned somehow and i can't figure out how to unban it [14:09] i thought i whitelisted it but all connections from this one computer are refused [14:13] NOVAtechi: check in iptables, are you using fail2ban or denyhosts? [14:13] #sudo iptables -L -v -n [14:14] if not then in hosts.deny or something. Are you getting anything back (actual rejected connection) or is it a steath drop? [14:14] etc. etc. :) [14:18] using fail2ban, i'll double check iptables. when trying to connect it says 'network error: remote host refuses connection' [14:19] the only thing i can think of is there is a key conflict [14:21] what if i used a proxy to reroute the traffic so the remote host thinks i'm coming from somewhere else? [14:21] it's a dirty workaround but i think that will work for now [14:27] proxy didn't work, this host is downstairs so i'll be a minute [14:40] not know why, always always need type: sudo ifdown eth0 && sudo ifup eth0 for start work with network card? any advice? thanks [14:48] so apparently my fail2ban configs got all crazy [14:48] reloaded from last weeks backups and that fixed the problem [14:49] any idea how that could've happened? [14:49] my fail2ban and ssh configs were both read/write to everyone [14:51] when to check them and the whole thing was blown away, completely empty. [14:52] when I'm sending files from server to server then one file is sent with speed 20mb/s and next 500kb/s [14:52] how can I check what it is so slow?? [14:52] CPU usage is low, memory as well [15:17] anyone know anything about xenstore ? [15:18] on a vm rackspace guest, i expected i'd be able to see some things in xenstore [15:18] # xenstore-ls [15:18] local: [15:18] xenstore-ls: xs_directory (/local): Permission denied [15:42] what can i possibly do with ubuntu server that i am using in my own home LAN? [15:44] craptalk: not sure I understand the question? or maybe it's too broad? you can do whatever you want with your ubuntu server [15:44] nacc, is it hard to use it? lets say for file server or something ? [15:45] nacc, what do you want to suggest for such local network usage? [15:45] craptalk: it totally depends on your needs [15:45] nacc, since i am asking you, what do you think it is cool to do with it? [15:45] if you don't have needs for a server, don't run a server? :) [15:45] i need learn how to manage it [15:46] craptalk: why? [15:46] nacc, i am asked to handle server with ubuntu in it [15:48] craptalk: by who? what will that server be doing? [15:49] nacc, lol, it is not important man, i just need your suggestion [15:50] nacc, i want to do it within the local network and play with what ubuntu server can do [15:50] file server, managing user control to access the files, maybe some network additional usage [15:50] craptalk: ok, do those things? [15:50] nacc, [15:51] nacc, yes [15:57] jamespage, yes b3 announce went out last week. I'll be sure to blog it in the future as well. [15:58] craptalk: to be clear, i meant those are all things you can do with ubuntu server, so do them. === JanC is now known as Guest6029 === JanC_ is now known as JanC [16:25] I would be very thankful if someone help me to understand the back end process of how temporary vg is creating and data is being migrated from /dev/sda1 PV to /dev/sda3 PV http://paste.ubuntu.net/23174271/ [16:25] jgrimm: would you like to take bug 1611923? [16:25] bug 1611923 in python-django (Ubuntu) "http.request does not support ipv6-formatted ipv4 addresses" [Medium,Triaged] https://launchpad.net/bugs/1611923 [16:26] rbasak, ack [16:26] rbasak: would you have time for a quick HO today? [16:27] nacc: briefly, give me ten minutes? [16:27] rbasak: sure, that's fine with me [16:27] Sorry I know I didn't get back to you yesterday [16:27] rbasak: totally fine! [16:27] rbasak: we can also talk tmrw AM my time, if that'd work better for you [16:31] powersj: how about bug 1448500 and bug 1583503? For the latter, determining a Won't Fix may be fine if nobody is actually affected any more (due to the workaround in neutron). [16:31] bug 1448500 in tftp-hpa (Ubuntu) "tftp-hpa doesn't accept IPv4 connections by default" [High,Triaged] https://launchpad.net/bugs/1448500 [16:31] bug 1583503 in keepalived (Ubuntu Xenial) "keepalived fails to start when PID file is empty" [Medium,Triaged] https://launchpad.net/bugs/1583503 [16:32] rbasak, ok! I'll add to sheet [16:32] Thank you! [16:40] nacc: rebooting now. I'll send you an invite in 2-3 minutes probably. [16:42] rbasak: thanks [16:44] jgrimm: hi! I have this MIR bug which seems to be about packages that would be maintained by your team, I think [16:44] https://bugs.launchpad.net/ubuntu/+source/python-requests-kerberos/+bug/1620293 [16:44] Launchpad bug 1620293 in python-requests-kerberos (Ubuntu) "[MIR] python-requests-kerberos and pykerberos (deps of python-keystoneauth1)" [Undecided,Incomplete] [16:45] for the MIR bug I would need to know if keystoneauth1 must require kerberos and checking that there is sign-off on committing to maintain these packages === InfoTest1 is now known as InfoTest [16:59] cyphermox, looking [16:59] cyphermox, tho i'm guessing openstack team unless its generally being used by other packages [17:08] cyphermox, yeah, we've split out openstack only dependencies to "Ubuntu OpenStack" team. Those 2 packages mentioned in the MIR would seem to apply. === _SJW_ is now known as Ussat === iberezovskiy is now known as iberezovskiy|off [18:03] could someone take a look at https://bugs.launchpad.net/nova/+bug/1621257 ? this seems to affect mitaka as well as master, so it might well be caused by some package below nova [18:03] Launchpad bug 1621257 in OpenStack Compute (nova) "VNC console keeps reporting "setkeycodes 00" exception" [Undecided,Invalid] [18:16] jgrimm: alright, thanks [18:42] Hi guys, I'm currently busy following this tutorial on how to setup a secure LEMP server with specific users / groups for each individual website. [18:42] https://www.digitalocean.com/community/tutorials/how-to-host-multiple-websites-securely-with-nginx-and-php-fpm-on-ubuntu-14-04 [18:43] However, I'm stuck since the tutorial was written for PHP5 FPM, I'm using PHP 7. [18:43] SupaYoshi: are you on 14.04? [18:43] Now in the instructions it tells, me to put this into my php 7 conf. [18:43] listen = /var/run/php5-fpm-site1.sock [18:43] Yes I am. [18:44] SupaYoshi: so you're using a PPA? php7 isn't supported officially in 14.04 [18:44] Yeah I am using the PPA. [18:44] I believe. [18:44] SupaYoshi: as much help as I'm willing to give, you'd probably follow the guide at: https://www.digitalocean.com/community/tutorials/how-to-install-linux-nginx-mysql-php-lemp-stack-in-ubuntu-16-04 [18:44] SupaYoshi: which is updated for php7 [18:44] Ah I see. [18:44] SupaYoshi: but regardless, not supported here, really :) [18:44] yeah, when I installed 14.04 that was the latest LTS. [18:45] :P Okay. [18:45] Well. Thanks for the guide. [18:45] Do you suggest me upgrading the system to 16.04? [18:45] regardless of LTS, using a PPA takes you out of support in this case (php7.0 not present at all in official 14.04) [18:45] I;ve never really done a release-upgrade before on a server. I've a homeserver with 12.04. [18:45] SupaYoshi: that's totally up to you, i'd say [18:45] I'm asking what you would do, and why yes or no. [18:46] Trying to learn a bit from experienced people ;) [18:46] I don't administer any servers, so I'm probably not the right person to ask :) [18:47] hehe. Okay! No bother. [19:04] Well.. the tuturoail you gave me isn't for securing users in their own jail. / website. [19:04] So I'm going to go to #php and try there ;) [19:04] Thanks though! [19:05] SupaYoshi: i meant just compare the sections about php [19:05] SupaYoshi: it will require *some* thought on your part :) [19:07] Thanks ;) === IdleOne is now known as Guest54953 === Seveas is now known as Guest88242 === patsToms is now known as Guest52985 === mwhudson is now known as Guest72628 === spammy is now known as Guest67489 === Guest88242 is now known as Seveas === Guest5508 is now known as jelly === Guest4608 is now known as pcw === pcw is now known as PaulW2U === TodPunk_ is now known as TodPunk === smoser` is now known as smoser === tsimonq2alt is now known as tsimonq2 === Guest96370 is now known as Adri2000 === Guest72628 is now known as mwhudson === IdleOne is now known as Guest56430 === masACC is now known as maswan [22:01] xnox, what's happening with qemu 2.6.1? === imr is now known as i- === Guest67489 is now known as spammy