| === EP0D is now known as Epod | ||
| === Epod is now known as EP0D | ||
| HardcoreKrypton | anyone knows how to use cloud-init | 10:09 |
|---|---|---|
| rbasak | Yes. | 10:13 |
| HardcoreKrypton | can you tell me ? | 10:13 |
| rbasak | http://cloudinit.readthedocs.io/en/latest/ | 10:16 |
| rbasak | If you have a specific question, please ask it. | 10:16 |
| HardcoreKrypton | thank you ! | 10:16 |
| HardcoreKrypton | okk | 10:16 |
| rbasak | There's also #cloud-init | 10:16 |
| ahasenack | rbasak: hi, is there a way to tell git ubuntu merge to use debian/experimental instead of sid? | 13:11 |
| rbasak | ahasenack: I'm not sure, sorry. I've not been much involved in that subcommand. | 13:21 |
| ahasenack | ok | 13:22 |
| ahasenack | oh | 13:22 |
| ahasenack | I guess I should have read --help more carefully | 13:22 |
| ahasenack | "onto A reference to a commit whose corresponding version to prepare to merge with. If not specified, debian/sid is used. " | 13:23 |
| ahasenack | :) | 13:23 |
| coreycb | jamespage: qemu and libvirt updates are in kilo-staging and ocata-staging | 15:43 |
| jamespage | coreycb: hurrah! | 16:00 |
| rbasak | ahasenack: looking at https://code.launchpad.net/~ahasenack/ubuntu/+source/ubuntu-advantage-tools/+git/ubuntu-advantage-tools/+merge/337213 | 17:06 |
| ahasenack | rbasak: ok | 17:06 |
| rbasak | I'm not very keen on stuff using bash in /etc/update-motd.d/. Debian and Ubuntu moved away from bash due to startup times, and update-motd.d is in the critical path for login time. | 17:06 |
| rbasak | What actually puts files into /etc/update-motd.d/ please? I don't see it in the packaging. | 17:07 |
| ahasenack | rbasak: debian/install | 17:07 |
| rbasak | Ah. Thanks | 17:08 |
| ahasenack | rbasak: I continued using bash (the main ubuntu-advantage script already uses bash) because it's not a new dependency | 17:08 |
| ahasenack | I didn't think about startup times | 17:08 |
| rbasak | I think it'd be fine for the ubuntu-advantage script itself since it isn't in any critcal path. | 17:09 |
| rbasak | But update-motd really needs to be kept quick. | 17:09 |
| rbasak | (especially for stuff present by default on all systems) | 17:09 |
| ahasenack | rbasak: well, the motd script will call ubuntu-advantage status | 17:09 |
| rbasak | Or else, measure it and convince me :) | 17:09 |
| rbasak | Oh | 17:10 |
| rbasak | Well then that matters too | 17:10 |
| ahasenack | it's either that, or reimplement in the motd the logic that determines what's the status | 17:10 |
| rbasak | Yeah | 17:11 |
| ahasenack | you would gain a bit by reimplementing in the sense that "ua status" determines the status of all 3 services | 17:11 |
| ahasenack | livepatch, esm, and fips | 17:11 |
| ahasenack | livepatch is a snap call | 17:11 |
| ahasenack | esm and fips is an apt-cache policy call | 17:11 |
| rbasak | Another way might be to shortcut it. If a user has never had UA, then detect that and do nothing quickly. | 17:11 |
| ahasenack | you mean had an ua contract, or the ua package? The ua package will be there always | 17:12 |
| ahasenack | from the start, I mean, it can be removed | 17:12 |
| rbasak | Then the buck is passed to Canonical to make sure that adding UA doesn't slow things down. I'm fine with that :) | 17:12 |
| dpb1 | wait | 17:12 |
| ahasenack | if livepatch is available, then the motd becomes an advertisement | 17:13 |
| dpb1 | update-motd.d runs on a schedule | 17:13 |
| rbasak | ua contract, or the ua package> I mean had UA contract related functionality enabled. | 17:13 |
| rbasak | Or "has" would be better I suppose. | 17:13 |
| ahasenack | dpb1: no | 17:13 |
| dpb1 | it's not in critical path for login time? or am I very wrong | 17:13 |
| ahasenack | unless there is some smarts that only calls the scripts in update-motd.d in a schedule | 17:13 |
| dpb1 | ah | 17:13 |
| dpb1 | script-by-script basis? | 17:13 |
| rbasak | dpb1: AIUI, update-motd.d runs on every MOTD display, but some scripts do their own caching. | 17:13 |
| rbasak | Right | 17:13 |
| ahasenack | yes | 17:13 |
| dpb1 | got it | 17:14 |
| guzzlefry | howdy | 17:18 |
| dpb1 | ahasenack: what is the path for the default install case, does ubuntu-advantage get called there? | 17:18 |
| ahasenack | dpb1: default install of that, ubuntu? | 17:19 |
| dpb1 | install server, then login. | 17:19 |
| ahasenack | s/that/what/ | 17:19 |
| ahasenack | the motd script runs at login | 17:19 |
| ahasenack | ssh, terminal | 17:19 |
| ahasenack | nothing else calls "ua <stuff>" on its own | 17:20 |
| dpb1 | yes, I mean, in that case | 17:20 |
| dpb1 | does ua <stuff> get called there | 17:20 |
| ahasenack | yes, we do one "ua status" call to determine the livepatch status | 17:20 |
| ahasenack | save the output, and then look for the bits we want in that output | 17:21 |
| ahasenack | a grep and two seds, something like that | 17:21 |
| dpb1 | I'm looking for what is done for motd.ubuntu.com | 17:23 |
| ahasenack | that must be different, since it's a network call | 17:24 |
| * dpb1 can't find it :/ | 17:25 | |
| rbasak | 50-motd-news | 17:25 |
| dpb1 | is it on xenail? | 17:25 |
| dpb1 | xenial | 17:25 |
| rbasak | I don't see it on Xenial | 17:26 |
| * dpb1 checks bionic | 17:26 | |
| rbasak | Looks like 50-motd-news only prints from a cache to me. | 17:27 |
| rbasak | And the cache is updated out of band. | 17:27 |
| rbasak | Unless --force is used, which is commented to happen from the systemd timer | 17:27 |
| dpb1 | hm | 17:28 |
| rbasak | Seems to me that there is a common need for caching. | 17:28 |
| dpb1 | yes | 17:28 |
| dpb1 | and there is a systemd | 17:28 |
| dpb1 | that updates the same file | 17:28 |
| rbasak | Who wants to write a general caching infrastructure for update-motd.d? :) | 17:28 |
| dpb1 | lol | 17:28 |
| dpb1 | well | 17:29 |
| ahasenack | rbasak: if we cache the "ubuntu-advantage status" output, would you be ok with the motd script itself be in bash? | 17:38 |
| ahasenack | or do you want that in dash as well | 17:38 |
| ahasenack | or are you still looking and have other comments? | 17:39 |
| ahasenack | rbasak: of course, we have to find a common denominator involving trusty all the way up to bionic regarding that caching | 17:40 |
| ahasenack | systemd might be out of the picture | 17:40 |
| rbasak | ahasenack: you could use cron | 17:43 |
| ahasenack | probably | 17:43 |
| rbasak | ahasenack: I want everything in dash, unless it's demonstrated that if everything in update-motd.d moves to bash, it wouldn't cause any significant difference. | 17:43 |
| rbasak | ahasenack: since I think this package should be consistent with everything else; there's no reason to grant it an exception on this. | 17:44 |
| ahasenack | ok | 17:44 |
| ahasenack | can you make these comments in the mp please? | 17:44 |
| ahasenack | (if you haven't already) | 17:44 |
| rbasak | ahasenack: done just now :) | 17:47 |
| ahasenack | thx | 17:48 |
| rbasak | ahasenack: I'm EODing soon. I'll finish reviewing on Monday. | 17:48 |
| ahasenack | ok | 17:49 |
| HackeMate | hello folks | 19:40 |
| HackeMate | i have configured bind9 for dns forwarding as cache, it works fine, now i want to set up a dns proxy, or any system that filter all dns queries in order to block porn sites | 19:41 |
| HackeMate | is there a project working in this way? | 19:41 |
| sarnold | HackeMate: https://dnsdist.org/ is cool stuff | 19:42 |
| HackeMate | oh, thanks | 19:42 |
| HackeMate | i wonder how big should be your curriculum sarnold | 19:43 |
| dpb1 | HackeMate: if you bottom out on anything, you could update https://help.ubuntu.com/community/ParentalControls | 19:43 |
| sarnold | HackeMate: what do you mean? | 19:43 |
| HackeMate | dpb1: thanks | 19:43 |
| HackeMate | sarnold: you have a huge knoweldge | 19:43 |
| HackeMate | that's what i mean | 19:44 |
| sarnold | heh, it feels like I'm learning less than is being created every day .. it feels like endlessly slipping backwards :) | 19:44 |
| HackeMate | that feelings makes you powerfull | 19:45 |
| sdeziel | HackeMate: the implementation probably depends on the kind of block list you use. For example, you could leverage OpenDNS FamilyShield by setting their IPs as forwaders in your bind | 19:46 |
| BenXVII | Hello is there anybody on that could enlighten me with an issue I have on my HP server ? | 20:52 |
| teward | BenXVII: not unless you give details about the 'issue' and ask your *real* question | 20:53 |
| teward | failing that nobody can help you. | 20:53 |
| BenXVII | I didn't know wether this was the proper place to ask. I plugged a 4TB SAS drive. But for some reason it is not recognized. In hpssacli it says "Status: Unsupported". It's an HP ProLiant DL370 G6 server. | 20:54 |
| teward | BenXVII: that's an HP hardware question, so this isn't the right place | 20:55 |
| teward | so you'd need to contact HP probably for assistance with the system saying it's unsupported | 20:55 |
| BenXVII | Thank you teward :) | 20:55 |
| BenXVII | It doesn't even appear in gparted :) | 20:55 |
| teward | probably because it's not a supported drive or something else, the fact it doesn't show up *sounds* like the drive and your hardware aren't compatible | 20:56 |
| BenXVII | I will contact HP support. I hope it won't be too expensive :p | 20:57 |
| patdk-lap | are you even sure the port you plugged it into supports sas disks? | 21:04 |
| patdk-lap | you didn't plug it into a sata port did you? | 21:04 |
| BenXVII | patdk-lap, yes I am sure, the server only supports SAS | 21:04 |
| patdk-lap | I have never seen a server that only supports sas | 21:04 |
| BenXVII | Might be that the controller is outdated and doesn't support 4TB volumes | 21:05 |
| teward | patdk-lap: I have. some older poweredges here :P | 21:05 |
| patdk-lap | teward, heh? | 21:05 |
| teward | they're sitting in the scrap pile for destruction, but they only have 3 hotswappable disks, all SAS | 21:05 |
| patdk-lap | every poweredge and hp I have ever had supported sata, and ones with raid cards did both | 21:05 |
| patdk-lap | hotswappable sas != sas only | 21:05 |
| BenXVII | I tried to plug in a SATA drive once and it wasn't supported | 21:06 |
| BenXVII | all drives are in the front bays, there are 8 of them | 21:06 |
| teward | patdk-lap: well, this one *only* had SAS drives, I guess whomever set it up played safe. | 21:07 |
| teward | *shrugs* | 21:07 |
| teward | ultimately irrelevant, we have much newer servers :p | 21:08 |
| patdk-lap | https://h20195.www2.hpe.com/v2/getpdf.aspx/c04282679.pdf?ver=1 | 21:08 |
| patdk-lap | supports sas and sata | 21:08 |
| patdk-lap | the raid card that comes in it only supports 2tb drives though | 21:08 |
| teward | ah that explains that | 21:08 |
| teward | BenXVII: unsupported drive size. | 21:08 |
| BenXVII | Thought so. Do you guys think I can use 2 2TB SAS and make them in the OS as one logical volume of 4 ? The purpose of the drive is to backup an attached NAS. | 21:09 |
| patdk-lap | hmm, it can support 4tb, if your lucky and have upgraded firmware | 21:10 |
| patdk-lap | also, those raid cards do not support passthough that I know of | 21:10 |
| patdk-lap | so you have to setup a raid config | 21:10 |
| BenXVII | Oh well, it's for a local non-profit, I will tell them to the backup of the NAS on two 2 TB dirves :) | 21:11 |
| BenXVII | RAID is already setup, so I don't want to start all over again :p | 21:11 |
| crester48 | hello | 21:27 |
| BenXVII | patdk-lap and teward .. thanks a lot for your help | 21:47 |
| BenXVII | Have a good night everybody :) | 22:19 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!