[01:59] <mtl> upgrading ubuntu 16.04 server to 18.04 was quite "painless"
[03:35] <teward> sopparus: dav_ext_methods isn't a directive - http://nginx.org/en/docs/http/ngx_http_dav_module.html
[03:36] <teward> sopparus: i think you meant dav_methods ?
[03:36] <teward> can you provide an example configuration you're testing with? (I don't have any webdav examples lying around at the moment)
[03:37] <teward> sopparus: the dav_ext_methods directive is part of third-party code
[03:37] <teward> so i'd need an example config to test with
[03:41] <teward> sopparus: nginx-full works fine in 18.04, but only if you use the version in the repositories.  Verify you have the proper version with `apt-cache policy nginx-full` and `nginx -v` which in Bionic will say "nginx/1.14.0 (Ubuntu)"
[03:41] <teward> i just pulled a webdav example from the web and it works and runs without that error.  *Only* if you use nginx-full or nginx-extras, though, it's not in nginx-core
[03:41] <teward> (and apt-get install nginx will install nginx-core by default)
[03:47] <teward> (E: NOREPRO)
[05:40] <sopparus> teward, its from an external package
[05:40] <sopparus> teward, in ubuntu 16 it was this https://github.com/arut/nginx-dav-ext-module
[05:41] <sopparus> ill post config
[05:47] <sopparus> teward, https://pastebin.ca/4019228
[05:50] <sopparus> teward, the same config works on freebsd with nginx 1.14 and that webdav module
[05:50] <sopparus> and ofcourse it worked on 16.04 :)
[05:58] <sopparus> teward, solved it. include /etc/nginx/modules-enabled/*.conf; was added
[05:58] <sopparus> i mean, had to be added
[05:58] <sopparus> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=851784
[06:23] <nostrora> Hi everybody, i'm looking for advice about my new Ubuntu LTS server. I'm setting up a home server to host NextCloud. i have intel C3758 + 16GB ddr4 + 2*4TB ZFS . Do i need swap ?
[06:42] <OpenTokix> nostrora: yes
[06:42] <OpenTokix> nostrora: but it does not have to be very big
[15:08] <Meerkat> Is the server install UI for 18.04 suppose to be different than earlier releases?
[15:09] <rbasak> Yes
[15:09] <rbasak> https://wiki.ubuntu.com/BionicBeaver/ReleaseNotes#Server_installer
[15:16] <Meerkat> I see. Good thing the server installer with RAID, VLAN, and reusing existing partitions still exist. Thanks!
[16:18] <RoyK> Meerkat: the new installer sucks - shouldn't have been released in this state
[16:23] <Meerkat> I went with the iso that uses debian-installer. I do love that installer.
[16:23] <Meerkat> https://help.ubuntu.com/18.04/serverguide/apparmor.html.en#apparmor-usage This page says there is a bug. But the link points to a bug that has been marked as duplicate. And the bug it is a duplicate of is fixed.
[16:25] <albech> Worst part of a new LTS is that my stupid Xen Server probably wont support it :(
[16:27] <ChmEarl> albech, worst part of bionic as a xen VM is the initrd.img defaults to 40MB
[16:27] <ChmEarl> ^^ causes a boot failure in some situations
[16:30] <albech> We've been debating for a long time now to switch away from Citrix XenServer due to their changes in licenses since 7.3. It's just a huge undertaking to rebuild everything. About 250 VMs :(
[16:37] <teward> sopparus: then i need to give you two tidbits of advice:
[16:37] <teward> (1) We don't provide support for third part packages
[16:38] <teward> (2) if the packge is 'third party', then I can't vouch for anything.  Base 18.04 with nginx-full from the repos (or my PPA) "works" out of the box
[16:53] <SuperLag> Mad props to the Ubuntu Server devs. 16.04 -> 18.04 has been the smoothest upgrade I've done to date.
[16:55] <sopparus> teward its part of nginx full
[16:55] <sopparus> but ok
[17:10] <teward> sopparus: what's unclear is if you're using `nginx-full` from the repos, and if this was an 'upgrade'
[17:11] <teward> because like all other bits of apt, sometimes things don't get installed if existing directories exist
[17:11] <teward> you say "from an external package" which means it's not from the repos
[17:12] <teward> and at that point you run into the problem of "third party packages" which are beyond the ability for us to determine the 'behavior' of
[17:12] <teward> we know how the nginx packages in the repository will work.  throw in 'external packages' and we can't predict behavior :P
[17:12] <teward> i know it's part of nginx-full, I'm the Ubuntu package maintainer for nginx alongside the server team
[17:12] <teward> but if you use *external* packages that aren't in the repos, we can't predict the behavior.
[17:13] <teward> 'tis what i meant in my tidbits earlier
[18:44] <rbasak> RoyK: could you be more specific please, so that the developers involved can make it better?
[18:56] <webchat146> Just installed Ubuntu 18.04 on a standalone server. I noticed that is using netplan and systemd-networkd for network configuraiton now, which I really like. However, I am a bit confused that /etc/netplan/50-cloud-init.yaml also says that cloud init will overwrite the file.
[18:56] <webchat146> Did I do something wrong during installation or is cloud init getting installed by default? Does it make sense in a single server environment?
[19:13] <teward> webchat146: cloud-init is how the server was 'started' if you used the live server disk and not the alternate images.
[19:14] <teward> i have a server running that for just under 2 days now with multiple reboots because I kept fiddling with the configs at the virtual level; cloud-init hasn't overwritten the configs.
[19:20] <webchat146> teward: hm I see. Yeah I also noticed that my network config does not get overwritten. But I rather like to either make use of cloud-init and let it handle network, or get rid of it (or disable the networking part) and use netplan
[19:21] <webchat146> teward: But I don't know cloud-init really, before just blowing it away I rather would like to understand if it is useful to embrace it in a single server environment....
[19:38] <madLyfe> possible to take Ubuntu Server 16.04.4 LTS (GNU/Linux 4.13.0-38-generic x86_64) instance and upgrade it to Ubuntu Server 18.04 LTS
[19:38] <madLyfe> ?
[19:41] <Meerkat> yea. sudo apt dist-upgrade
[19:41] <Meerkat> actually, do-release-upgrade is the command I think
[19:42] <madLyfe> "Upgrades from 17.10 will not be enabled until a few days after 18.04's release. Upgrades from 16.04 LTS will not be enabled until a few days after the 18.04.1 release expected in late July."
[19:43] <tomreyn> madLyfe: well quoted.
[19:43] <madLyfe> so i should wait till then?
[19:43] <tomreyn> the depends on you. you could also do an unsupported upgrade now
[19:43] <tomreyn> but for any production server you most definitely want to wait until the upgrade path is tested
[19:45] <madLyfe> 'sudo do-release-upgrade' is the command though?
[19:46] <tomreyn> yes, but you'd need to add -d since it's not supported, yet
[19:46] <madLyfe> -d = do?
[19:46] <tomreyn> i wont spend the time answering this if you can just do a quick --help
[19:47] <madLyfe> --help: command not found
[20:08] <irwiss> -d = hold my beer
[20:09] <madLyfe> i like to live dangerously.
[20:11] <Meerkat> looks like 18.04 server doesn't blank the screen. Good stuff!
[20:28] <Meerkat> I was restarting apache and this came up: "Authenticating as: test,,, (test)" Why the commas? Username has no commas in it.
[20:30] <tomreyn> we don't know your apache httpd (? or some other apache project, one of the dozens they have?) configuration nor the users who can authenticate against your system, so that's difficult to answer.
[20:33] <Meerkat> Fresh install of 18.04. I only just installed apache and I enabled mod_speling. Then I was restarting apache.
[20:33] <Meerkat> mod_speling is a module for apache enabled with "a2enmod mod_speling"
[20:36] <Meerkat> The command "systemctl restart apache" returned that string. Do I check its source code or does systemctl call code from other packages? Like apache in this case?
[20:41] <dpb1> mad "-d" Lyfe.  -d for dangerous
[20:41] <dpb1> :)
[20:49] <madLyfe> 'System upgrade is complete.'
[20:50] <madLyfe> 'Welcome to Ubuntu 18.04 LTS (GNU/Linux 4.15.0-20-generic x86_64)'
[20:56] <madLyfe> got the error when running my application:
[20:56] <madLyfe> ./xmr-stak: error while loading shared libraries: libmicrohttpd.so.10: cannot open shared object file: No such file or directory
[20:56] <madLyfe> tried to do 'sudo apt install libmicrohttpd-dev'
[20:56] <madLyfe> 'libmicrohttpd-dev is already the newest version (0.9.59-1).
[20:56] <madLyfe> 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.'
[21:09] <tomreyn> why -dev? it's looking for an .so
[21:10] <tomreyn> unless you meant to build it?
[21:12] <madLyfe> not sure those are the build instructions for the application. so i figured id just go back and try to install it again since it couldnt find it.
[21:13] <madLyfe> do i need to recompile the application or can i just add the lib back in?
[21:15] <TJ-> madLyfe: 18.04 has libmicrohttpd12: /usr/lib/x86_64-linux-gnu/libmicrohttpd.so.12 not .10
[21:16] <webchat146> FYI I wrote a short post about the Ubuntu Server installer topic: https://blog.printk.io/2018/04/ubuntu-18-04-lts-bionic-beaver-server-installer-differences/
[21:19] <RoyK> rbasak: it's documented, they have acknoledged the probleem, they haven't fixed it
[21:23] <madLyfe> TJ-: how do i get .10 so my application can function?
[21:24] <TJ-> madLyfe: Other way around generally; rebuild the application so it links to the installed library, or else do some customisation using LD_LIBRARY_PATH and a private copy of the .10 library
[21:25] <madLyfe> are the make options set somewhere in the file to just run the rebuild again?
[21:25] <madLyfe> like, whatever make build options i designated from the prior build?
[21:31] <madLyfe> looks like that was the case
[21:32] <madLyfe> the make build options must have been stored from the previous build