[01:14] hi, is it possible to connect freshly installed server to wlan accesspoint (my phone)? It worked fine with desktop via bluetooth and wlan, but I'm clueless how to do it from server. [01:17] There is plenty of guides how to set up server as accesspoint but not the otherway around. Unless you can manually set ip, but I don't think that works on phone accesspoints. [01:29] Might be I need to get some drivers first, while installation seemed to find wifi, server now doesn't seem to find ath0. [01:37] mko: google it - http://askubuntu.com/questions/16777/connecting-to-a-wireless-network seems a good start [01:37] yeah just found it thanks [01:37] mko: try iwconfig [01:41] wlp3s0 shows IEEE 802.11bhn ESSID:off/any [01:44] then try to configure that one [01:45] network devices in modern distros don't have the same names as in the old ones [01:46] slight problem is, I can't configure wifi on my phone, its some preset stuff I don't know about [01:46] other than password ofcourse [01:46] I'm not sure how to scan it [01:46] ooh [01:48] either I broke it or it takes a loong time [01:50] should it work if I add "auto wlp3s0; iface wlp3s0 inet dhcp" to interface? [01:52] at least iwlist scan returns no scan results, so that might be a good sign, but there is wlans so it's still not working. [01:52] you'll need the ssid and the password obviously [02:09] okay I might be getting there, now iwconfig didnt return error [02:09] but i didnt give it password, I might be blind as I can't seem to find how to from man :o [02:12] I just get invalid argument "password" [02:41] I dont understand how i can misspell "iwconfig ssid key s:pw" [02:41] ssid -> was supposed to write iface [02:41] anyhow it returns "set failed on device iface ; invalid argument [02:45] most likely iphone uses wpa/wpa2 and its not working? [02:47] wpa_supplicant is used to connect with wpa [03:29] I might've gotten further, but wpa_supplicant returns error invalid group wheel, and failed to initialize control interface /var/run/wpa_supplicant [03:37] mko: did you use the right driver? You can see your list of supported drivers with wpa_supplicant -h [03:41] i tried nl80211, wext and wired but no luck [03:42] wext gave also ioctl[siocs]wencodeext]: Invalid argument [03:42] mine was nl80211 [03:42] I have panda wifi adaptor [03:44] i ordered it because it said it works on linux [03:46] integrated 802.11 b/g/n or 802.11 ac [03:46] whatever that means [03:47] thanks for helping, nightshift is over and time to go to sleep. :) [03:48] np [04:34] I'm trying to run a python script on startup. Where should I put the script on my filesystem? And how do I make it start when booting? === Thorn__ is now known as Thorn [05:18] Am I supposed to use /etc/init or /etc/init.d for my own daemons in the latest ubuntu? The internet doesn't say === tikund is now known as sikun [09:26] i am new to maas ,i think i have setup a maas (getting web gui ) but i have no idea now how do i add node and how do i boot client from host using pxe /remote boot [09:27] using ubuntu 16.04.1 === 1JTAA53XX is now known as m1dnight === JanC is now known as Guest44273 === JanC_ is now known as JanC === Ph4edrus is now known as Pha4drus === JanC is now known as Guest81597 === JanC_ is now known as JanC === spammy is now known as Guest39703 [14:31] sadly, i must run an ftp server. can i solicit recommendations for a reasonably unshitty choice of software? [14:32] ideally, it would support ldap directly. [14:32] [as opposed to indirectly, through nss/pam] [14:34] pureftpd? just apt it, it has ldap,mysql versions too [14:34] so far, the consensus seems to be one of pureftp, vsftpd, or proftpd [14:35] cncr04s: thanks for the vote. do you specifically prefer pureftpd over the others? [14:35] oh - it must support ftps too, of course. plaintext ftp won't be allowed === Guest39703 is now known as spammy === m1dnight1 is now known as m1dnight_ [16:02] Could someone please help me figure out why unattended-upgrade isn't sending emails, even though it's configured to do so, and it seems to be applying updates correctly? [16:03] can you machine send emails [16:03] thats the first test [16:03] Yes. mailx is installed and verified functioning. [16:04] Or at least the 'mail' command is. [16:04] so your host can send mail - how is the unattended-upgrade calling the mail command [16:05] No idea; I didn't see any configuration for that. [16:06] MAIL_BINARY = "/usr/bin/mail" [16:06] And SENDMAIL_BINARY = "/usr/sbin/sendmail" [16:06] ok - so it's calling /usr/bin/mail [16:06] thats good [16:06] check /usr/sbin/sendmail exists and is a s symlink to postfix [16:07] /usr/sbin/sendmail looks like an ELF [16:08] really ? you're running sendmail on your box ? [16:08] thats not default [16:11] I'm not sure where to go from that. [16:11] do you have the sendmail package installed ? [16:12] Lemme figure out how to do that on ubuntu; if these things had Portage I would be able to tell you now. :3 [16:12] dpkg-query says sendmail is not installed. [16:13] And if I use apt-get to try to install it, it wants to remove postfix and install a bunch of sendmail packages. [16:14] Default or not, sendmail is functional. [16:15] /usr/share/boot-sav/gui-g2slaunch.sh: line 29: 11520 Segmentation fault (core dumped) $G2S $1 -g ./$PACK_NAME.glade -s ... full details here: pastebin.com/kSjuHkPP [16:15] natmal: ls -la /usr/sbin/sendmail [16:16] ikonia: -rwxr-xr-x 1 root root 26648 Apr 13 12:58 /usr/sbin/sendmail [16:16] It's not a symlink. [16:16] natmal: Are you sure they're not just stuck in the queue? Type mailq [16:16] thats very odd, it's normally a symlink, I guess postfix must now be shipping it's vinary as "sendmail" [16:16] Mail queue is empty. [16:17] does the mail log show it trying to be sent [16:18] and does unattended-upgrade have a log? check that for errors. [16:21] No errors in the unattended-upgrades logs. I'm not sure about the mail log; my assumption is that they're all test emails when I was trying to turn this functionality on last week. [16:21] Is running unattended-upgrade manually supposed to trigger emails? [16:22] check the mail log [16:22] I have MailOnlyOnError set to "false" [16:22] it will show anything that is proccesed by the mail system [16:24] mail.log only has startup info for postfix (Doesn't have anything about the test email I sent earlier today). mail.log.1 only has entries from the 3rd. [16:24] natmal: there should be a postfix log of it's own [16:24] the syslog may also show mail attempts (I can't remember the channel split without looking) [16:27] No mail/postfix attempts that I can find in any of the syslog files. [16:28] doesn't look like it's trying to mail out [16:28] I'd check the docs to see what the behvaiour is expected [16:30] Docs say that /usr/bin/mail is required, but the code tries to use /usr/sbin/sendmail? :P [16:31] look at what the expected behaviour is [16:34] There doesn't seem to be any documentation beyond "notifications will be sent to this address assuming you have a working mail setup". [16:35] I've looked at the Automatic Updates page in the Official Ubuntu Documentation, the AutomaticSecurityUpdates page in the Community Help wiki, and the UnattendedUpgrades page on the Debian wiki. [16:35] look in /var/log/mail.log [16:36] and search for pickup [16:36] patdk-lap: 2 matches from the 3rd; test emails. This system was automatically updated requiring a reboot between then and now. [16:37] well, the sendmail binary puts the email in the maildrop folder [16:37] and pickup should see it, and then act on it [16:37] is postfix running? [16:37] does /var/spool/postfix/maildrop have files in it? [16:38] No files in maildrop; postfix is not running. [16:38] Does postfix need to be running if configured solely as a relay? [16:38] I take that back; postfix is indeed running. [16:39] postfix is not running....you will not get mail [16:39] does a webserver need to be running to solely serve webpages? [16:39] Wrong flags on ps. :3 [16:39] if your doing relay only, you should just use a nullmailer instead [16:39] msmtp, or something [16:42] Will switching to that make unattended-upgrades start wanting to send mail? [16:49] I might stick with postfix; I'm not finding much information about SASL support in nullmailer, and msmtp doesn't appear to have a global config? [16:50] Oh, I think I found it. === phunyguy is now known as _phunyguy [16:52] it's doubtful either of them will solve the issue [16:53] but it will remove a lot of configuration problems from being the issue [16:53] using postfix as a nullmailer is generally extreem overkill and extra complications for most === _phunyguy is now known as phunyguy [16:57] Okay, msmtp is configured. [17:04] Yeah, I think I'm going to stick with postfix. I don't have to jump through a bunch of hoops trying to securely store upstream passwords with msmtp. [17:16] Huh. I think email notifications are working; I may have been chasing my tail. [17:18] Thanks for all the help anyway [17:23] securely store upstream passwords? [17:24] how do you "securely store upstream passwords" with postfix? [17:27] when is this a use case? [17:29] what? [17:33] which passwords do you want stored securely? [17:33] "upstream passwords", he says [17:34] right, that's not very clear, and so i'm asking [17:35] "upstream passwords" is pretty clear. "securely storing" them with postfix isn't though [17:49] hi [17:51] installed server 16.04.1 on: 3 disks in raid5 on sda1 sdb1 sdc1. then on md0 created an lvm PV, there a VG with 3 LV: root, swap, home. so /boot is part of /. it then proceeded to install, then install grub. [17:52] won't boot. says "cannot find filesystem" -> grub rescue. [18:11] DexterF: you probably need a separate boot LVM. it's even possible that you need to have it as a separate partition without extra layers. [18:13] /boot on lvm is fine [18:13] also make sure the UUIDs in your /etc/fstab are correct (compare to 'blkid' output) and that you have run update-grub, and that you are using an initrd with all modules, not the limited set. [18:14] i'd suggest mounting with labels rather than uuids in fstab [18:14] lunaphyte: and boot on lvm on md raid5 is fine, too? [18:14] it works for me [18:14] i'd suggest mounting with uuids rather than labels in fstab ;) [18:15] because they are universal [18:16] no more so than the label specified by the admin ;) [18:18] if the bios has a bad day and passes devices in a different order (or you, that admin, made some silly changes which affe3ct the boot ordering), it will still work with uuids, but not labels. [18:18] incorrect [18:18] enlighten me! [18:19] labels of filesystems do not change just because the devices the filesystems reside on were detected in a different order [18:20] if a filesystem is labelled 'root', than that is its label. it doesn't matter when it shows up [18:29] for reference, here is a system with 4 disks, each with a gpt table containing a raid partition. assembled into a raid5 device, being used as an lvm pv, with an lv for "/". no separate /boot. http://dpaste.com/0A3DX29.txt [18:31] wag: you didn't install grub on the right disk [hint: install it on all of them], and/or you should include a bios_grub partition on each of the disks === seg_ is now known as seg [18:52] tomreyn: well, I'll have to rig a live system to assemble the raid first, at present I can't even access the array. update-grub and anything about the initd wsa not really queried, I ran with what the installer wanted to do. I did the entire setup with it, too [18:58] to do a setup like that, i would suggest running the installer in cli expert mode [18:59] lunaphyte: turns outr we had a misunderstand in regards to the term 'label' - i was assuming you were referring to device paths. [18:59] ...such as /dev/sda1 [18:59] oh [18:59] i've never heard that called a label [19:00] lunaphyte: does the installer offer that at the start...? don't remember such an option [19:00] lunaphyte: you're right, it's not, my bad [19:00] no worries, it's all good [19:00] DexterF: yes [19:00] lunaphyte: ok, i'll retry then. [19:00] you need to press an f key or something like that, iirc [19:01] is this for a server? [19:01] i guess regardless, i'd also suggest the minimal installer iso [19:01] one other thing: it seems I had to go LVM as I did not see any options in the installer to create md partitions, so md0p1 etc. not available? stick to lvm anyway? [19:01] that said, you could boot into the live cd, and assembly the bits yourself from a shell, i suppose [19:02] i would always use lvm, period. [19:02] personally i'd even go so far as to do the parititioning first, from a live linux, then use the installer to install. [19:02] lunaphyte: actually it's going to be a media center on kodi, so I will install a light desktop afterwards, I went for the server variant to directly install as raid5 on the existing 3 disks, I did not want to add another disk for keeping the power consumption at bay [19:03] if it were me, i'd do the minimal iso installer [19:03] I got the 600MB cd on usb here, any gain in switiching to minimal? that will start on expert cli right away? [19:04] no, but you'll not end up with a bunch of garbage that you have no use for [19:04] well, you still will - but less of it [19:04] ubuntu *sucks* at understanding the concept of minimal :) [19:05] sounds good, after all, I'll need X, kodi, tvheadend and compile the latter ones from src anyway [19:05] you can always sintall whatever you want, at any time. [19:05] *install [19:05] that's the plan [19:06] by the way, if I can boot the iso, can I do the installation remotely via ssh somehow? [19:06] probably [19:07] ok, no official feature then. [19:07] there might be. it's just not something i happene to be familiar with [19:07] *happen [19:10] there is talk of the "network installer" - that said "minimal" iso? [19:10] http://archive.ubuntu.com/ubuntu/dists/xenial/main/installer-amd64/current/images/netboot/mini.iso [19:11] https://help.ubuntu.com/community/Installation/MinimalCD [19:29] DexterF: ssh installation won't work unless it's manually selected from the advanced menu of the mini.iso installer, or preseeded. [19:32] i.e. unless you preseed the installer for ssh, you will need to be physically present, or have a networked KVM or serial console (but then you probably wouldn't ask for the ssh method), to boot the installer and select the ssh installation option. once that's done, you can finish the installation remotely. [19:33] that's neat [20:42] tomreyn: thanks [20:54] DexterF: welcome [21:02] Hello. [21:03] I recently upgraded a server from 14.04 to 16.04. Now when I use byobu (with tmux) when I press left/right arrow I get an orange circle int he middle of my screen: http://i.imgur.com/96BEyWt.png [21:46] I installed fresh server, then xen, but now i got stuck in the bootloop. last thing i see on screen is "loading initial ramdisk ..." [21:46] not sure if dmesg should have something in it but its empty [21:47] only advice i found was about updating bios but its up to date [23:21] on an EFI install, is grub installed on the ESP partition or the standard / partition