[00:51] estoy vivo? [00:51] soy muerto [00:53] he he he, I thought that it was disconnected. [01:26] jgrimm: fyi, dogtag-pki finally built, it's migrating now, tests are queued, i'll check again in the AM [01:26] nacc, thanks sir [02:18] is there any apps like iostat for cpu and memory usage ? [02:23] fishcooker, top or htop [06:29] good morning [06:30] moin [07:39] Hello :-) [07:40] First time installing ubuntu-server anything you'd suggest [07:41] I mean't is there anything you'd suggest? (any pitfalls to avoid etc) [07:41] if it's going on the live internet please disable password authentication in sshd_config as soon as you can [07:43] sounds important thanks! [07:50] thrasos: sounds like you are playing around wiht linux servers is this right [07:51] that is true [07:54] thrasos: check htop it is really a nice way to monitor the server resources [07:57] thanks for the tip [08:00] thrasos: you can use 'apt-cache search' to search for packages to install; it searches both descriptions and names. you can use apt-cache show to see more information on the package, and apt-get install to install it [08:07] excellent === jamespag` is now known as jamespage [08:10] thrasos: If you need something speciffic just tell us a direction :), a software raid with mdadm. You can mix random disk sizes (with equal partitions) and raid1 is super nice for redundancy ... many here will disagree with my love to software raid but it did gread job for me on many servers [08:11] <3 zfs [08:13] +1 on zfs it is really REALLY nice [08:22] hi [08:49] Good morning. [09:13] ZFS has advantages and disadvantages [09:14] it seems like ZFS is not as predictable as linux software RAID + ext4 when it comes to performance [09:14] Wasn't it supposed to be superior? (Never used it...) [09:15] and you can't expand a ZFS RAID with extra disks, like you can with software RAID [09:16] well, you can, but not in the same way as you can with software RAID (you need to create a new volume with multiple disks) [09:16] OTOH, it has much better integrity checking of course, and features like de-duplication [09:17] and it supports compressed files (although that's not useful if you use it primarily for already compressed files) [09:18] one nice feature is also that you can enforce UTF-8 filenames :) [09:38] Hi guys, there are few iscsi packages floating around and I am wondering which one is the most stable for a home lab NAS, integration with the zfs commands would be a plus === Amgine_ is now known as Amgine [10:28] hi all :) [10:28] Is there a possibility to make a fan profile for a 3 pin fan [10:28] fancontrol does it for a PWM fan, but not a 3 pin fan. can it be done with a 3 pin fan too? [10:39] I have installed a headless virtual box on my server A, this virtual box contains another ubuntu server B. I can ssh from my laptop to server A, and from server A to server B, but how can I ssh directly from my laptop to server B ? [11:00] cpaelzer: I updated the blueprint to find any outstanding merges in there that aren't already in proposed. I have nut, logwatch and python-django. Are you taking care of logwatch today? I'll look at nut (I think it's pending my review from caribou) today. nacc isn't here but I'll ask him about python-django later. [11:00] roaksoax: ^ [11:35] rbasak: yeah logwatch is already in proposed now [11:36] rbasak: IIRC python-django had something about test failures - the log of the IRC meeting has some details, but in any way nacc is the right one to know [11:39] cpaelzer: ah great. Thanks! [11:39] (re: logwatch) [12:10] Hello all! [12:10] Is disabling Ipv6 support a good practice ? why ? [12:11] If you have an ill configured network it can give you grief [12:11] I'd rather fix the network [12:12] Hmm OK, but what is the risk of having ipv6 disabled ? [12:13] That you cant communicate over ipv6 if you need to [12:43] loosing half the internet! [12:46] rbasak: and migrated, I see you already set done - thanks [12:58] Genk1: Nowadays you want to slowly move to ipv6. So do try to leave it enabled. [13:00] lordievader, OK thanks! [13:03] Hello, is nish aravamudan in this chat ? [13:03] zioproto: he's nacc, but not here right now. [13:03] rbasak, thanks, I wanted to ping him about this https://bugs.launchpad.net/ubuntu/+source/puppet/+bug/1570472 [13:03] Launchpad bug 1570472 in puppet (Ubuntu) "Set systemd as default service provider" [Medium,In progress] [13:03] do you know on what time zone is he ? [13:04] zioproto: UTC-8. [13:05] thanks ! === rvba` is now known as rvba [13:42] caribou: around? I'm just writing up review notes from nut now. As we need to upload today for FF, do you have time for some tight review loops, or would you like me to just take care of it? [13:43] Sorry it's been so late coming. It's (almost) one of the last ones left :-/ [13:45] caribou: also I'll suggest tweaking the changelog a bit. I can propose the change, but would like an ack for you before keeping your name on it. [13:48] I am trying to understand in LP if there is a list of merge requests against the puppet package [13:48] I arrived here https://code.launchpad.net/ubuntu/+source/puppet/ [14:07] rbasak, reading all the story of LP bug #1570472 probably I could ask also to you [14:07] Launchpad bug 1570472 in puppet (Ubuntu) "Set systemd as default service provider" [Medium,In progress] https://launchpad.net/bugs/1570472 === Guest55639 is now known as IdleOne [14:12] zioproto: I'm not really current with what the issue exactly is. Could you perhaps summarise in the bug, effectively rewriting the bug report including all the information that others have gathered in comments, and identify any inconsistencies? [14:13] sure, I'll try my best [14:13] Thanks! [14:13] should I then make also a merge request on launchpad like I do with the openstack packages ? [14:14] We don't have an official repo branch for puppet packaging in Ubuntu. Normally attaching a debdiff to the bug is fine. [14:14] What is this then ? ~usd-import-team/ubuntu/+source/puppet [14:14] I was just looking that up for you :) [14:14] It's our experimental work on git branches for Ubuntu packages. [14:15] If it's current, feel free to file a merge proposal against the appropriate branch in there. [14:15] If it's not current, we can bring it up to current manually on request. [14:18] but how is it possible that this bug is open since April 2016 ? [14:19] every puppet user is still on trusty ? :) [14:19] hmm, it's only since april [14:19] it would be more concerning if it was april 2012 [14:19] No idea. We'd be quite happy to fix it if we were clear on the details of the problem. [14:20] Also, there's a trivial workaround posted in the bug. [14:21] the workaround does not work if you dont have a site.pp [14:21] we use foreman [14:21] I suspect there are far fewer users in that situation. [14:22] soon tons of people are going to be forced to do the trusty xenial upgrade [14:22] ...who don't have a site.pp and use foreman? [14:25] I just mean that if you have a puppet master with external_nodes you cannot apply the workaround [14:25] * rbasak shrugs [14:25] the packages from the PPA work okay, and also other users reported they work [14:25] Like I say, we'd be happy to fix it if someone who cares comes along and tell us what to do. [14:25] I dont understand what is the blocking factor [14:26] We need to know that it's the minimal fix and there's no regression risk. [14:26] I think we need to wait for nacc [14:26] Since the set of people who test the PPA are a biased sample representing only one use case. [14:26] To understand the regression risk, we need to understand what scenarios are affected, how and why. [14:26] do you know if the tree where the PPA is buillt from is published somewhere ? [14:27] Yes. See the PPA page. You can get the sources from there. [14:27] "and there's no regression risk" -> well, there's never *no* regression risk, but we do want to minimise it, and for that we need a good understanding of the problem and the fix. Saying "the patch works" isn't good enough for a proposed change to a stable release. [14:27] https://launchpad.net/~nacc/+archive/ubuntu/lp1570472 [14:28] if I click on code [14:28] "View package details" [14:28] Open the arrow against the version/series you want. [14:28] You can download the source package files from there. [14:30] ok [14:30] zioproto: thank you for driving this. We appreciate your help. [14:35] rbasak, I understand how to get diff from 3.8.5-2ubuntu1~ppa7 to 3.8.5-2ubuntu1~ppa8 [14:35] but this means it is the version number 8 of the PPA ? [14:36] I would like to diff against che Xenial package [14:36] I mean the buggy one that is delivered to the world :) [14:36] zioproto: you might find https://git.launchpad.net/usd-importer/tree/git-dsc-commit useful [14:37] It imports a Debian source package into a fresh git commit. Then you can use git to diff as normal. [14:37] zioproto: check out the correct parent commit first, eg. ubuntu/xenial-devel if the PPA package is based on that. [14:37] okay, so I use it with https://launchpad.net/~nacc/+archive/ubuntu/lp1570472/+files/puppet_3.8.5-2ubuntu1~ppa8.dsc [14:38] zioproto: yes, but you also need the associated files. You can use dget to pull them from the dsc URL automatically. [14:52] zul, backports should be back to normal soon for ocata [14:53] ack [14:55] I am trying to run a php based application on my LAMP server which is reporting error 500 can any one help me. I have a a scripts that run well on LAMP server with php6 but not working with PHP7 [14:56] Have you checked the logs? [14:57] where to find? [14:58] Logs usually go to /var/log [14:58] If you are using Apache with mod_php you want to look into the error log of Apache. [15:08] Thank you lordievader. Trying to figure out the issue. [15:09] What is the error? [15:19] unable to figure out. accidently deleted the error log. [15:20] now nothing is being added to it. [15:20] Restart apache ;) [15:26] Thanks dear. [15:26] [Thu Feb 16 20:53:51.821767 2017] [:error] [pid 22108] [client 127.0.0.1:59574] PHP Warning: require(include/RequirementsValidation.php): failed to open stream: Permission denied in /home/isthakur/data/isthakur/Sites/yetiforce/index.php on line 17 [15:26] [Thu Feb 16 20:53:51.821822 2017] [:error] [pid 22108] [client 127.0.0.1:59574] PHP Fatal error: require(): Failed opening required 'include/RequirementsValidation.php' (include_path='.:/usr/share/php') in /home/isthakur/data/isthakur/Sites/yetiforce/index.php on line 17 [15:26] permision is fine i.e. 644 on index.php [15:30] permission on folder was not given to other so corrected that but now there is a huge list of errors. I will try to fix it and then let you know. Thank you very much for showing me the route to troubleshoot the problem [15:30] Can you browse to that file and read it as the apache user (www-data)? [15:31] now yes [15:31] I have to leave right now. for half an hour. catch you later. [15:31] and once again thank you very much. [15:32] No problem === din0 is now known as dino82 [16:01] rbasak, the git-dsc-commit too is perfect :) I was able to diff the ppa code from the current stable. The patch LGTM. What is the workflow to propose the patch for the proposed Xenial packages ? Should not nacc do it ? He is the original authro [16:05] jamespage: on your work snapping qemu/libvirt did you create an interface that covers /var/run/libvirt/libvirt-sock-ro ? [16:06] or the non read-only version [16:07] hmm I found a libvirt interface, checking definition [16:09] zioproto: anyone can do it. See https://wiki.ubuntu.com/StableReleaseUpdates#Procedure, and just attach something usable to the bug (that includes packaging changes such as the version in debian/changelog) and subscribe ~ubuntu-sponsors to the bug. [16:20] Question: When there is network links from my IP to an IP blocked via iptables, that likely means my server is compromised, yes? [16:22] Amgine, what do you mean with 'network links' [16:22] ? [16:22] using iftop shows data connection, outbound. [16:24] zioproto: https://dpaste.de/HJUh/raw === runelind_q_ is now known as runelind_q [16:26] this is the output of a command like ss ? [16:26] sudo iftop, it is like top only for network traffic. [16:26] try ss -tp [16:28] do you see a connection ESTABILISHED with that host ? [16:29] No. Nor do I see a connection for the steady streams of spiders, bots, and browsers. [16:29] But I think I understand what you're suggesting. [16:29] Amgine, do you see more stuff with ss -tpa ? [16:30] Just me. [16:30] It could be that the traffic you see with iftop is just ICMP replies that your port is closed. Or maybe you are sending TCP RST [16:32] kk. Just being paranoid, I suppose. [16:36] nacc, welcome in ! :) I was waiting for you :D [16:37] nacc, I have a production openstack cloud where the Trusty to Xenial upgrade is blocked by this bug: https://bugs.launchpad.net/ubuntu/+source/puppet/+bug/1570472 [16:37] Launchpad bug 1570472 in puppet (Ubuntu) "Set systemd as default service provider" [Medium,In progress] [16:38] nacc, I looked at the diff between the package shipped in Xenial and the ppa you published on launchpad. Changes look ok. There is a reason why this patch was not yet proposed to Xenial ? [16:38] zioproto: mostly business :) i have been waiting for upstream feedback on it [16:39] *busyness [16:39] zioproto: let me prepare an SRU today for it [16:39] yahoo :) [16:39] if I can help tell me what I have to do [16:39] zioproto: thanks! [16:39] usually for openstack packages I make a merge request of my git tree on launchpad, but I guess SRU has a bit more of burocracy [16:40] zioproto: yeah, but in this case, pretty easy to argue for [16:42] nacc, leaving the office soon, I sent you my mail in pvt [16:42] of you need any testing to be done write to me [16:42] zioproto: thanks again [16:42] also, if we go to production using your PPA on launchpad, can we rely on that PPA to be available ? [16:43] zioproto: i won't remove it, and it should get trumped by the SRU [16:43] ok great [16:43] the problem is that Openstack MItaka is the only release support both on trusty and xenial [16:44] so we have to upgrade to xenial to move on to openstack newton [16:44] so I guess a lot of openstack people will soon move to Xenial [16:44] better get this puppet thing fixed :D [16:44] I have to go ! thanks for the help ! [18:17] coreycb: do we need a blanket FFE? [18:18] zul, i don't think we do this cycle. it might be worth emailing the release team though just to keep them in the loop. === miczac\away is now known as miczac === miczac\away is now known as miczac [20:34] rbasak: another 'merge' update was pushed by me just now, it's a minor minor packaging strings change for debian/control but it syncs us up with Debian except for our delta that we add. [20:34] sarnold: jgrimm: ^ just an FYI [20:34] (for nginx) [20:34] teward, ack [20:36] hopefully that sneaks in right before FF goes into effect [20:36] otherwise, it probably can be FFe'd without issue [20:39] yup! :) [20:40] whoopsies I left junk .save files stupid editor crashes [20:41] nginx -1ubuntu2 inbound heh. [20:41] someone please shoot me for my screw up [20:41] the taser is over on the desk :P [20:41] `git status` ftw [20:41] nacc: not git tracked [20:41] *manual* merge [20:42] or `git status --ignored` [20:42] nacc: ^ not git tracked. Yet. [20:42] bah, you can still do it in a git repo locally :) [20:42] nacc: I *could* except that these're minor fixes. [20:42] In any case, it's uploaded/fixed finally hopefully in time for FFe. [20:42] I need coffee, really. [20:42] Release team is probably going to slap me for so many uploads today xD [20:42] :) [20:43] nacc: it'll be tracked *soon* on my own gitlab. I couldn't reliably use any git workflows because of some very evil merge conflicts [20:43] that basically needed start-over-from-scratch for the delta [20:43] and about a month of headbanging against fPIE/fPIC [20:43] teward: oh i get that, i just meant, it's trivial to just do a `git init` locally when you start, then you know what you are changing from the debian version, at least -- even if manual [20:44] teward: i've found my own mistakes in trivial merges by doing that [20:44] teward: but on to bigger and better things any ways :) [20:44] nacc: true, but i also have an automated testbuild workflow that would kill that :P [20:44] so blah [20:44] in any case, it's *now* git-worthy since the dynamic module stuff is in [20:44] :) [20:45] I should go get coffee though now heh [20:45] * genii 's ears perk up for a second at the mention of coffee [20:46] genii: give. [20:46] :P [20:46] wow Chrome crashed lol [20:47] * genii covertly meets Juan Valdez at the Columbian border and obtains coffee for teward [20:49] * genii hand-roasts the beans, grinds them in an antique brass grinder, prepares enough for 1 large mug in the French press. [20:49] * genii slides teward the mug [20:50] * teward takes, and disappears gradually (with the coffee and the mug) into the shadowy mists that seemed to just 'appear' in the periphery of the channel. [20:59] Hi guys, there are few iscsi packages floating around and I am wondering which one is the most stable for a home lab NAS [21:19] heh, depends on what your doing [21:19] for me, all of them are stable [21:19] it's more in what you want to interface with it [21:20] lio seems to be so dymaically changing, that if you want to script it (use pacemaker with it), it keeps changing and it's hard to use [21:28] Its mainly for my lab ESXi boxes, I am migrating to ubuntu from freenas [21:29] for that you may be just as well off with NFS if you don't find a satisfactory answer for iscsi [21:31] wedgie: Yeah I am thinking of moving them to NFS but at the moment I have a collection of them on an iscsi volume. If I upgrade to 10G interfaces then NFS is apparently the way to go anyway [21:32] PhoenixMage: you're moving from freenas... will you be using zfs on the ubuntu system as well? If so that's a double good reason to go NFS. Much more graceful handling of compression [21:33] unless your real issue is, your stuck on 1gbit network, and attempting to balance over multible [21:33] but then make sure your iscsi service can make use of multipathing [21:33] but nfs would be perferred though, as you don't have to worry about empty space reclaiming [21:34] indeed. One glorious day NFS will support multipathing but it's not quite standard yet, as I understand it. Haven't looked into it for about a year though [21:34] wedgie: yeah zfs on ubuntu, just improted all my zpools [21:38] huh, maybe nfs4.1 with multipathing *is* a thing already: http://packetpushers.net/multipathing-nfs4-1-kvm/ [21:39] oh, wait, requires recompliling the kernel. So yeah, not quite prime-time just yet [21:40] I am running a lagg across 4 interfaces and 3 hosts, not ideal [21:40] Anyway, seems I might just go with any of the iscsi implementations and migrate to nfs === madsa is now known as Guest13012 === madsa_ is now known as Guest87296 === keithzg_ is now known as keithzg === miczac is now known as miczac\away === Darkman802_ is now known as Darkman802 === miczac\away is now known as miczac