=== j_f-f_ is now known as j_f-f [12:40] hggdh, I just wanted to say thanks for adding your bot to our channel and we don't need anything else [13:47] mattymo: you are welcome === j_f-f_ is now known as j_f-f [17:16] jsalisbury: Bug 1241871 it's reported that the upstream now has a fix in it, maybe a candidate for update? [17:16] Launchpad bug 1241871 in linux (Ubuntu) "8086:0166 [ASUS ZENBOOK UX32VD] 13.10 kernels (3.11.x) Fail to light" [Medium,Triaged] https://launchpad.net/bugs/1241871 [17:16] penguin42, thanks! I'll take a look [21:47] Hello, can I ask for an apport-collect command in bugs like this one ? https://bugs.launchpad.net/ubuntu/+bug/1254867 or because it has not a package assigned apport-collect will not have any effect ? [21:47] Launchpad bug 1254867 in Ubuntu "Waking from Sleep Makes Some Games Perform Sluggishly" [Undecided,New] [21:48] I think a bot automatically sets it to incomplete and asks the reporter to pick a package [21:49] personally I'd guess for Linux and want a set of logs gathered after a resume [21:51] Thanks penguin42 . As I'm not sure what package is the appropriate one, I will leave it as it is. [22:15] what does one do when you can't use `ubuntu-bug` to file a bug about something crashing, because apport itself crashes? [22:16] TheLordOfTime: first step: cry. 2nd: verify if there is a bug opened against apport on that [22:16] hggdh, first step cry? lol [22:16] usually, apport -- being THE eror handler -- should not crash [22:16] so, I would cry first ;-) [22:16] hggdh, so if apport is crashing, go curl up in the corner, cry, panic, and then just lay down there and sob until you die? [22:17] or, of course, my favorite: [22:17] nuke the system with a clean install [22:17] which will prolly not solve the issue... [22:17] well... [22:17] and I did not say 'curl up & cry', I just said cry. Some 5 seconds is enough [22:18] how did apport crash? Backtrace available? [22:18] no idea [22:18] the user who asked it on ask ubuntu's main chat isn/t there [22:18] and i have questions for em [22:18] k. I do remember some issues on apport some time ago, perhaps the user needs to dist-upgrade [22:19] hggdh, http://chat.stackexchange.com/transcript/message/12337200#12337200 and the next several messages after that are relevant [22:19] 13.10 was the offending system [22:20] my first quesiton was "Did they in-place upgrade 13.04 to 13.10?" [22:20] but of course i'm 8 hours late to the party [22:21] heh [22:22] yeah, without knowing how ubuntu-bug crashes, it is difficult to say where is the issue :-( [22:23] and... who is filemanager? [22:33] nfc [22:35] hum. I cannot even log in to the stackexchange chat... [22:35] hggdh, do you have an askubuntu nick? [22:35] if so you login to askubuntu first :P [22:35] then access the chat [22:39] I did... [22:39] bah. I will get thru it later. Try with another browser, etc [22:41] mhm [22:41] the person who was having the issues isn't around though [22:41] yeah. Until s/he comes back, hopefully, I will have squared out my issue with login [22:41] I wonder if 2-factor auth is playing a role [22:48] could be [22:48] hggdh, you using google to login or something? [22:48] because i just use launchpad.net as my openid auth [22:48] and it works xD [22:50] I was also using lp.net. It does not work... So I tried having one tab on lp.net (logged), and one tab on askubuntu (logged). Stil, chat.stackexchange fails [22:50] checked probably blockage, found I was blocking data from stackauth.com; allowed it, checked OK, tried again -- still fails [22:51] s/probably/probable/ [22:52] cache nuke [22:56] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1254834 , how is this possible ? [22:56] Launchpad bug 1254834 in ubiquity (Ubuntu) "Can't report bug with apport because /var has no free space" [Undecided,New] [22:59] The whole root partition is filled up, I guess. /var is under /root. If the user had a separate /var , I guess s/he would know what to do in order to free /var space. ;-) [23:01] Should I convert it to a question or not ? [23:03] NikTh: my first feeling is that it should be a question -- be it the same partition (usual) or a different one (rare, user *must* know what is being done), the fact is there is no space... [23:05] I will convert it to a question then. Thanks hggdh [23:05] even more given the error message asks for cleaning up space [23:09] Correct :-) [23:11] hmm, I have a problem with the RSS feed (Firefox bookmark) right now. It shows only 10 results. [23:14] NikTh: you mean the Launchpad bugs feed? If so, you can /join #ubuntu-bugs-announce [23:14] Yes bugs feed I mean. Thanks hggdh === Jikan is now known as Jikai [23:32] yeah the bugs announce channel is nicer :P [23:32] latest bugs are always boring [23:33] penguin42, i have the ones for nginx on highlight [23:33] i'm expecting people from Apache to try and change the nginx default docroot [23:33] ... which violates Debian policy... [23:33] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=730382 [23:33] Debian bug 730382 in nginx "nginx: Please change the default document root to /var/www/html" [Important,Open] [23:34] why's it open it should be won'tfix [23:35] penguin42, apparently, 11.5.4 debian policy states not to use /var/www/html... http://www.debian.org/doc/debian-policy/ch-customized-programs.html#s-web-appl [23:35] i think the apache people are crazy [23:37] they tried in Debian, it was rejected by the maintainers. if they try in Ubuntu i'll slap them quoting the debian bug and debian policy... [23:37] hggdh, Ubuntu tries to follow Debian policy as closely as they can, no? [23:37] or is there a different policy set in play for packages? [23:37] TheLordOfTime: So where does nginx currently have it's root? [23:37] (and how they behave) [23:37] penguin42, /usr/share/nginx (WHICH IS ACKNOWLEDGED AS WRONG!) [23:38] gonna be /usr/share/doc/nginx next upload [23:38] (per 11.5.4 http://www.debian.org/doc/debian-policy/ch-customized-programs.html#s-web-appl ) [23:38] at least that's what was implied [23:38] TheLordOfTime: What's a bit weird is that's where I'd expect documentation about the package, not the base place it serves it's documents from [23:38] penguin42, it's debian... [23:38] blame them [23:39] 11.5.4: [23:39] Web Document Root [23:39] Web Document Root [23:39] Web Applications should try to avoid storing files in the Web Document Root. Instead they should use the /usr/share/doc/package directory for documents and register the Web Application via the doc-base package. [23:39] whoooopsies [23:39] penguin42, i blame debian policy, because i'd expect the same you expect. [23:39] apache is the only webserver that has a requirement use of /var/www/html apparently though [23:39] and apparently Apache is saying "We're the debian standard, so follow us!' [23:39] when they're violating debian policy. [23:40] TheLordOfTime: Are we reading this the right way - does 'Web application' mean a server? or does it mean a set of things that get served? [23:40] 11.5 header penguin42 [23:40] 11.5 Web servers and applications [23:40] This section describes the locations and URLs that should be used by all web servers and web applications in the Debian system. [23:40] lrn2read? [23:40] sigh, ok [23:40] (sorry if i come off as an ass, i'm slightly under the weather, it's not intended) === Jikai is now known as Jikan [23:40] penguin42, the one time i get to tell someone from Apache off, though, was quoting the exact policy on the debian bug xD [23:41] TheLordOfTime: The thing is that's also insecure, it means that the default setup is to serve the things including things like changelog etc so you can see package version [23:41] (note: Debian BTS and iMail's mail formatting don't play nice) [23:41] penguin42, true, which was a concern [23:41] penguin42, however, this could also be fixed by /var/www symlinking to the file source [23:41] TheLordOfTime: And you don't want people screwing around with files in /usr/share/doc which will break next time the package gets updated [23:41] but until Debian policy is changed i don't think nginx is going to be changing things [23:41] penguin42, that's for the default config... [23:41] all default configs are designed as examples [23:42] not "Serve from here" things [23:42] most SANE sysadmins write their own directives and different docroot locations [23:42] * TheLordOfTime shrugs [23:42] penguin42, still, i find it interesting Debian Policy and APache disagree [23:42] but i do see your security observation there [23:43] (if nginx has an MIR and the security team notices that i may bring it up) [23:43] TheLordOfTime: Yeh I suppose everyone does do their own config - I suppose I was thinking about the simplistic cases [23:43] penguin42, you mean the people that want everything to "just work"... right? [23:43] the people who are ignorant to the intricacies of sysadminning :P [23:44] TheLordOfTime: Yeh, like if you just want a simple little internal webserver - not a full serving critical stuff [23:44] mhm [23:45] penguin42, but there's a difference between "make it work" and sane, IT-security-common-sense configurations [23:45] yeh [23:45] a good sysadmin knows how to configure things. [23:45] ... hmmm, do we even have /var/www in a default setup? [23:45] shrug, don't know [23:46] * TheLordOfTime pokes -server because the server team would know [23:46] TheLordOfTime: I'd point it to a subdir of /usr/share/doc/nginx - e.g. /usr/share/doc/nginx/default-doc-root or something like that [23:46] penguin42, however nginx does it they'll do it [23:46] i think there's suggestion to have an /html folder inside /usr/share/doc/nginx [23:46] nod [23:46] at least, drifting around [23:46] that would make sense [23:47] root /usr/share/nginx/html; <-- default config [23:47] well... [23:47] default config as of 1.4.3 [23:47] i haven't packaged 1.4.4 from Debian yet [23:47] although that would be the same, the change hasn't popped up until after 1.4.4 [23:47] so it'll be in 1.4.4-somenumber or 1.4.5 [23:47] * TheLordOfTime shrugs [23:48] TheLordOfTime: Do you do things like add selinux rules for it? [23:48] I personally don't. [23:48] * TheLordOfTime isn't maintainer [23:48] that'd have been done by the maintainers before me [23:48] if there's SEL rules for it, they're drifting around somewhere, or just not in the debian packages [23:48] probably be in the centos repos though [23:49] sorry, I actually meant apparmour [23:49] i don't think there's any rules for that [23:49] but i only run basic-needed features, lock down everything else in the nginx config [23:49] * TheLordOfTime yawns [23:49] bleh coffee time again... [23:56] penguin42, it shouldn't be too hard to generate an apparmor profile though [23:56] * TheLordOfTime shrugs [23:56] never tried it, maybe i'll mess with it on my testing system [23:57] TheLordOfTime: Probably best not to enable by default (because it complicates setup) but might be useful to ship disabled [23:57] mhm [23:57] penguin42, i don't know if there's any default profiles made available, i could find out but i don't see one anywhere [23:58] penguin42, remind me where the apparmor profile(s) would exist in a pacage? [23:58] package* [23:58] woahhhh [23:58] one of my EC2s has 150+ updates o.O [23:58] i guess unattended-updates isn't doing its job [23:59] somewhere under /etc/apparmor.d I think [23:59] WORK YOU STUPID PIECE OF... [23:59] * TheLordOfTime grumbles at his ec2 [23:59] it's hung up on downloading a bzip2 index with apt >.>