[07:45] yo [07:46] I'm thinking of setting up a new VPS, but I started to wonder if I should wait 5 months so that I can get it started straight with fresh 18.04 [07:46] I know upgrading works, did it from 14.04 to 16.04, but still... [08:49] the spiders crawling through my file server are endless - I thought that they would get satisfied at some point that they have the whole picture :P [08:50] is there some tool that lets me cut through all the noise in Apache's access log to filter in just actual user downloads? === blackbaba_ is now known as blakebaba === blakebaba is now known as blackbaba [13:16] guys [13:17] i always get annoyed by the weekly kernel updates on ubuntu [13:17] anyway to get kernel updates to be less frequent, debian/centos style, once every 2 months? [13:51] gunix: you don't *have* to update. [13:51] You could watch the USNs and make individual decisions on security updates. [13:51] rbasak: is there a way to check if the kernel update is security or not? [13:52] can't apt just say "dude, this is security related, do the update" [13:52] ? [13:52] https://usn.ubuntu.com/usn/ [13:52] You could add an apt pin to lower the priority of non-security updates I think. [13:52] apt does know the difference. [13:53] But I suspect you may find that virtually every kernel update includes security updates, eg. for drivers you aren't using. [13:54] There's also https://www.ubuntu.com/server/livepatch [15:12] ...or drivers you ARE using [15:13] gunix: which part annoys you about it? a prompt to install updates, the bandwidth being used during the download, the prompt to reboot after the update succeeded? [15:14] oh we'Re talking server here, so no such prompts, i guess. then what is the issue? [15:14] tomreyn: the fact that i always have to reboot the server to stay up to date [15:14] you have to reboot redhat/debian every 2 months, but ubuntu every week [15:14] think about that when running an openstack infrasrtucture [15:15] reboot every node once a week [15:15] gunix: well as said before, you don't have to reboot [15:15] yea but security update of kernel won't be applied until you reboot [15:15] but if the other distros only ship kernel security updates less frequently it means they are doing a worse job [15:15] livepatch was also suggested [15:15] livepatch is risky [15:15] can break the system [15:16] not rebooting for security updates is also risky [15:16] how can it "break the system"? [15:18] if you compare it to other distros which make kernel security patches available less frequently, you really need to see it this way: ubuntu provides you an opportunity (but does not at all require) to have a shorter time-to-patch, i.e. you can close security vulnerabilities faster and thus increase the systems' security overall. you can maker use of it, but you don't have to. [15:19] so comparing to the other distros, consider it a free add-on offer. [15:20] (or, if you care about security, consider ubuntu to do a better job than the others which did not provide those security patches in a timely manner, keeping your system at risk.) [15:21] if patching / maintenances interrupt your service, you should rethink your service architecture === WieWoWasWozu is now known as hggdh === hggdh is now known as WieWoWasWozu [16:26] tomreyn: live migration of instances is not always the best for customers [16:31] i've uninstalled lxd (from xenial-backports) from my test xenial server and it left a sizeable /var/lib/lxd behind. when i try deleting it, i get "/var/lib/lxd/storage-pools/default/images...... Read-only filesystem" How do i get rid of this directory? i've already dpkg purged everything lxc/lxd related. [17:38] and i don't see anything relevant in /proc/mounts [17:54] gunix: yes, sometimes HA is [18:12] nevermind, lxd created those directories as btrfs subvolumes. [18:14] Hello! [18:15] I have a problem - my system doesn't recognise my network adapter, TL-WN721N, after the 1st reboot. [18:16] USB tethering with my phone doesn't work either. [19:05] MartesZibellina, what does lsusb/lspci say about that wireless TL-WN721N? [19:26] It IS recognised, but the network manager says "Device not ready". [20:23] MartesZibellina: lsusb/lspci knowing its name doesn't mean there is a driver [20:26] seems like there are at least two different adapters with that name... [20:26] sorry, *five* [20:27] three with Atheros chips and two with Realtek chips [20:29] MartesZibellina: does it tell you what version it is? [20:29] or check what the USB ID is [20:32] the USB ID would look like "0cf3:9170" or "2357:0107" [20:34] the version would likely also be printed on the adapter itself (but might be hard to recognize as such) === tomaw_ is now known as tomaw