[00:00] lets flip that around a bit.. what documentation are you reading that suggests that it should work? I can't find it on e.g. https://github.com/fujita/tgt/blob/master/doc/tgtd.8.xml [00:03] https://community.mellanox.com/docs/DOC-1457 [00:03] which says to clone the github project you just mentioned [00:03] and gives the usage as -t [00:05] Hello, does ubuntu offer an integrated way to create MOK in an EFI secure boot environment and sign bootloaders and other efi binaries with it? [00:06] Or is it necessary to perform all the steps involved manually? (create all the keys, sign all the binaries, resign on updates) [00:07] DirtyCajun: wow... seven years ago. curious. https://github.com/fujita/tgt/commit/18868050236b065b0c66f7832df3763aea2968c8 [00:08] so it just... does the num of your vkernels [00:09] thats stupid [00:11] -h should print "-t, --nr_iothreads NNNN specify the number of I/O threads" if supported, apparently man pages didnt get updated. [00:15] afaict, still not in manpage upstream https://github.com/fujita/tgt/blob/master/doc/tgtd.8.xml [00:15] last changed june 5, 2017 :) [00:17] here is the real question. currently, I/O threads are per lun. so 2 luns = 32 threads for 16 core server [00:17] if i specify 8 is it total or per lun! [00:20] wait, you do HT, isnt that considered unsafe now? [00:21] .. unsafe security or stability wise? [00:21] * tomreyn trolling at https://www.mail-archive.com/source-changes@openbsd.org/msg99141.html - sorry for deflecting [00:49] tomreyn: and it's not even April Fools! :) [00:50] Next in OpenBSD land: Moving data to memory considered harmful. === giraffe is now known as Guest33711 === DropItLikeItsHot is now known as PityDaFool [05:15] good morning === qwebirc4945 is now known as Benl90 [05:20] Hello, I've question about iptables, I only allow port 22, 80, 443, 10000 and allow from localhost and from my public ip and drop another packet. I've problem with resolving the domain to ip after that, that lead to my ubuntu server can't update and solve ubuntu repo domain name. any idea how to allow nslookup too? Thanks [05:26] Benl90: should be port 53, but if you really allow everything from local and your IP that should not be blocked [05:26] I'd assume something blocks it [05:26] but there are plenty of examples if you search for the buzzwords, like https://gist.github.com/thomasfr/9712418 [05:26] cpaelzer: Yeah I drop the other, Ive try allow port 53 both tcp udp but still no luck [05:35] cpaelzer: In the end I fix the setting, name server allow -_-. bad ways, -_- [05:39] well, you got it working which is all that matters :-) [06:13] Good morning === gogbog is now known as Guest27691 [10:51] !info dovecot [10:51] Package dovecot does not exist in bionic [10:52] !info dovecot-core [10:52] dovecot-core (source: dovecot): secure POP3/IMAP server - core files. In component main, is optional. Version 1:2.2.33.2-1ubuntu4 (bionic), package size 2877 kB, installed size 8809 kB [10:53] Is there an SRU or something planned to upgrade Dovecot to 2.3.x? 2.2.x is EOL, not sure it'll receive even security fixes [13:11] I am running Ubuntu Server 16.04 and having issues setting up Postfix with Dovecot SASL [13:29] then you should probbaly try to get support with it! [13:33] I have followed numerous guides and cannot get it to work for the life of me on Ubuntu 16.04 had it working flawlessly on Ubuntu Server 14.04 [13:36] cryptodan: try http://www.sabi.co.uk/Notes/linuxHelpAsk.html to start. That'll help you ask the right questions here. [13:37] no need it has been fixed [14:26] possible to see what packages come with server 18.04? [14:27] madLyfe: https://packages.ubuntu.com/ ? [14:30] Is this channel dead or can I ask a really quick stupid question? [14:30] madLyfe: install and run dpkg -l? also a hint are packages that are dependencies (and their dependencies, and ...) of the ubuntu-server metapackage [14:30] v0lksman: that looks like for desktop? [14:30] It's about netplan. I've switched to Ubuntu Server from Windows, and am now trying to set things up as they were before [14:30] Luxray5474: you alraedy did, so.... :) [14:31] I've installer 18.04 and after a bit of research I've foudn that netplan is the only way to set a static IP [14:31] basically I wanted to see if SSH and tmux were installed with iso [14:32] Luxray5474: it isn't though. netplan is the novel network configuration abstraction tool. you can configure systemd-networkd directly, instead through netplan. [14:32] Generally there are a multitude of ways to do anything. But relatively few that are recommended. [14:32] which is easier though? [14:32] s/novel/new/ to avoid confusion with Novell [14:32] netplan is surely the easiest way to set a static IP. [14:32] madLyfe: https://packages.ubuntu.com/bionic/ubuntu-server [14:32] One config file and that's it. [14:32] okay [14:33] madLyfe: then dpkg -l |grep ssh will find that out for youy [14:33] but i need some help with setting it up [14:33] Luxray5474: netplan is meant to be One Config to Rule Them All, via yaml, and different backends -- currently networkd for servers and NetworkManager for desktops. [14:33] so there's this tag called 'addresses:' under ensxx and ethernets [14:33] Luxray5474: https://netplan.io/examples#dhcp-and-static-addressing [14:33] rbasak: technically, so is one tiny file in /etc/systemd/network ;) [14:34] now what I want to know is what is the number that comes after the slash after the IP? [14:34] It's the size of your netmask. [14:34] Do you know what your netmask is? [14:34] Luxray5474: that's the "CIDR" notation, look it up on googs [14:36] okay so, if it's like "192.168.1.75/24" that's the IP "192.168.1.75" and the subnet mask "255.255.255.0" right? [14:36] Yes [14:37] I get it now, thanks! :D [14:37] Luxray5474: try the "ipcalc" tool. "apt install ipcalc" and then "ipcalc 192.168.1.75/24". [14:38] okay [14:38] "sudo apt install ipcalc" I suppose. [14:38] It'll do all the calculations for you so you can experiment and verify things that way. [14:39] what does the 'ens' mean in say, 'ens33'? can i put just any number? [14:39] It's the name of your network interface. By default it's fixed, based on where it's physically connected inside your computer. [14:39] okay [14:40] and is 'nameservers' equivalent to the DNS? [14:40] Yes [14:40] Thanks :) [14:41] You're welcome! [14:47] okay so i think i did something wrong here [14:48] so when i do "ip a" it shows that I have 3 adapters i believe. one is wifi, one is Ethernet, and one is bluetooth [14:49] I'm not sure about the third one but that computer has a bluetooth adapter that supports internet, so I am positiv ethat the one [14:49] so i did sudo netplan apply and it did it without spitting anything out [14:49] but i went to my router page and my server is still 192.168.1.7, when i set it to 192.168.1.75 [14:50] actually one of them is loopback/localhost [14:51] so when i saved the new config and ran 'sudo netplan apply' it said "address already in use" [14:51] how do i reset configs so i can start from scratch? [14:52] do i simply delete the file? [15:00] What do you mean by "but i went to my router page"? [15:00] You mean your router's DHCP leases that it's handed out? If setting a static IP, the router won't know if you're stopped using your DHCP-assigned address. [15:23] i have an issue with my server, sometimes it crashes and have to reboot, i just manages to get a picture of the tty when it crashed, the screen is running with lots of messages i never saw before, can anyone please take a look and see if they know what it might be https://paste.pics/3AC18 thanks? [15:27] sruli: get a current bios, and all other firmware, too. your system bios dates back to 2014, and there are releases from this year. [15:28] you should probably download the SPP dvd and have it run through that. [15:28] also use ipmi to get a better idea of what may be going wrong. [15:28] i mean ilo, not ipmi [15:30] i dont have an account with HP so cannot download the SPP [15:30] if neither of those are options (why? this should not be.), review dmesg and syslog for issues reported during boot, and try to fix or work around them. [15:31] Hello, I am trying to install ubuntu server 18.04 on an intel board, I can install from usb just fine, but it won't boot afterwords, It looks like grub never gets installed, even though I see that as a step during install. [15:32] sruli: oh they require a login now, thats bad. well you can still download all those pices one by one. [15:33] tomreyn: this is final shot before it completely froze https://paste.pics/3AC3B [15:33] tomreyn: i updated as much as i could download from their site, did not notice the bios update, will check now [15:34] sruli: https://support.hpe.com/hpesc/public/home/driverHome?sp4ts.oid=5249572&lang=en&cc=us [15:35] tomreyn: all BIOS updates are "Entitlement Required" [15:35] sruli: hmm those also require "entitlement" [15:35] :-/ [15:36] i update everything that i can get, but hp needs to be slapped for requiring entitlement for bios updates [15:36] tomreyn: anything in the second screenshot that sheds light? [15:38] sruli: well the lower ones are edac (ECC memory) issues. but this could be a follow-up issue. you can run memory tests, and you can try disabling acpi if the boot logs suggest this might improve things. also, try using a non tainted kernel. [15:40] tomreyn: with regards to EEC yesterday "hpasmcli -s "show dimm" showed a few degraded, si i ran memetest all night no errors, all day today hpasmcli showed all ok. what do you mean by "tainted kernel"? [15:42] sruli: one that runs out of tree (often proprietary) kernel modules: https://unix.stackexchange.com/questions/118116/what-is-a-tainted-kernel-in-linux [15:42] check your screenshots which point out that the kernel is tainted [15:42] well just the first one does [15:44] tomreyn: i only install kernel from apt-update, no custom kernel [15:44] probably some modules though [15:45] the only modules which might not be from ubuntu repo are the HP tools [15:46] hpe "tools" are not drivers, not kernel modules for all i know. but maybe you have actually installed kernel modules from HPE. i woulod not know. [15:47] these are all the hp tools i installed "hp-health hponcfg amsd hp-ams hp-snmp-agents hpsmh hp-smh-templates ssacli ssaducli ssa" [15:48] btw. the md5sum for the SPP is 2d1047b00f8cc5645fdc18596ad1183e - you can always put that into a web search engine. [15:49] hp-health sounds like it could do kernel modules. running "lsmod" will show which modules are loaded. [15:50] i managed to download the bios file, updating it now [15:52] if this issue persists and you want to debug it properly you'll need to get the first of these errors / kernel oopses reported. [15:53] you'll need to attach a serial console (ilo got an internal one if you can use it) or use netconsole to have it written to a different computer on the network before the system freezes. [15:55] tomreyn: list of my loaded modules https://paste.ubuntu.com/p/vJcX9fMXBG/ [15:57] i'm not able to review this for you, sorry === miguel is now known as Guest66771 [15:58] rbasak: my router runs a little webpage that i can access using routerlogin.net. it has a tab showing all connected devices, including their ip and mac addresses. my server still had the IP 192.168.1.7 when I put 192.168.1.75 in the config file. [16:00] Luxray5474: I suspect that's really showing you DHCP leases. Use "ip addr" to show you what IPs a machine is really using. [16:01] yes i did that. It still shows ..1.7/24 [16:03] also when i tried to change it again it says that another daemon was running on the port ( i forgot which one) [16:03] sruli: actually the flags set by the kernel as seen on line 15 of your first screenshot https://i.paste.pics/3AC18.png are G D W L. and the G actually states that no out of tree modules are loaded. [16:03] sruli: ...as discussed at https://www.kernel.org/doc/html/latest/admin-guide/tainted-kernels.html [16:04] s/out of tree/non GPL/ [16:04] but the remaining flags indicate that waht this screenshot shows is just a follow-up issue. that's why i'm saying you need to get the full output. [16:07] tomreyn: just checked syslog, when i booted it today for a few seconds i have many 1,000's of messages "ureadahead[635]: ureadahead:dmi: Ignored relative path" is this normal? [16:09] on some releases it is. [16:09] 18.04 [16:10] you can uninstall ureadahead and they'll be gone. it's mostly relevant for desktops [16:10] but you can just keep it, too, no harm there [16:11] also see my private message [16:16] tomreyn: i see in syslog around the time of the crash a number of messages, regarding the MCE MEMORY, "mce: [Hardware Error]: Machine check events logged" ... "EDAC sbridge MC0: HANDLING MCE MEMORY ERROR" ... "EDAC sbridge MC0: CPU 2: Machine Check Event: 0 Bank 9: c8014c8700800091" and many 100's of "snmpd[1770]: Connection from UDP: [127.0.0.1]:44561->[127.0.0.1]:161" before and after those mem errors [16:21] sruli: so one of these messages states that memory bank 9 contains a faulty DIMM. this may or may not be so. you could try removing or replacing it and what gives, or to cross test [16:22] but before you do that, i'd recommend you try to get the full output first of all. [16:22] as discussed above, your options for this are to attach a serial console, and the linux netconsole feature [16:23] how can i connect through ilo? [16:24] you probably have an ilo 4, the user guide is at https://support.hpe.com/hpsc/doc/public/display?docId=c03334051 [16:25] if you want to access the console through iLO, you want to use either the "iLO Virtual Serial Port" or just the Text-based Remote Console textcons [16:26] the latter is less user friendly. [16:26] ... but doesn't require configuration changes on ubuntu [16:31] from hpe doc, Text-based Remote Console "This method increases iLO performance significantly. However, the digital video stream does not contain useful text data, and text-based client applications such as SSH cannot render this data" [16:33] the idea there would be to ssh to your ilo, then run textcons to get the output there. but indeed, as i said, an actual serial console is much better. [16:37] what your quote is basically saying is that you will need to use a special purpose client (textcons) to render the data stream. just using netcat or an ssh client directly will not work [16:37] textcons - iLO Advanced License required for this functionality. [16:40] is this a personal project, an NGO or a business you run there? [16:41] just purchased a ilo license on ebay for 19GBP [18:10] rbasak: okay after a while of trying to figure it out, it's still down. can i send you my config? [18:10] (so you can check it out and tell me if there's anything wrong w/ it) [18:12] Luxray5474: re "ens" and so on, https://github.com/systemd/systemd/blob/master/src/udev/udev-builtin-net_id.c#L20 [18:12] Luxray5474: https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/ [18:13] Luxray5474: .. and finally https://lists.ubuntu.com/archives/ubuntu-devel/2015-May/038761.html for some ubuntu-specific background [18:13] ohhh [18:13] okay so enp3s0 is eth [18:13] and wlp3s0 is wlan [18:13] for some reason wlp3s0 showed up as "(ethernet)" when i ran "ifconfig" [18:14] Does anyone have a good guide on settings up DKIM with Postfix on Ubuntu Server 16.04, its just one last thing I need to have setup a good mail server [18:16] cryptodan: this looks kinda old but might still be useful https://help.ubuntu.com/community/Postfix/DKIM [18:16] ive seen that one, and for some odd reason my emails are not being signed by dkim [18:17] rbasak: i understand that naming now, thanks. but one of the things that are displayed when i run netplan -d apply is "netplan/tcp not found in /etc/services, using ports 2983 and 5444" [18:18] cryptodan: there's a much smaller mention of dkim on https://help.ubuntu.com/lts/serverguide/mail-filtering.html.en-GB -- but that might go through amavis too to work :/ [18:18] oops i mean sarnold lol [18:18] Luxray5474: that's a funny error .. maybe pastebin your config? it feels like a word out of place or missing punctuation mark .. [18:19] okay I'll do that right now [18:21] sarnold: https://pastebin.com/cBBGXzzr [18:23] Hi, all. I would like to downgrade a package that was inadvertantly upgraded, but the old version seems to be gone from apt. [18:25] The package in question was upgraded, from the apt history log file: [18:25] Upgrade: ansible:amd64 (2.4.2.0-1ppa~xenial, 2.5.4-1ppa~xenial) [18:25] NightMonkey: neither of those are ubuntu packages [18:25] !info ansible xenial [18:25] ansible (source: ansible): Configuration management, deployment, and task execution system. In component universe, is optional. Version 2.0.0.2-2ubuntu1 (xenial), package size 992 kB, installed size 7181 kB [18:26] NightMonkey: PPAs do not keep infinite numbers of packages going backwards [18:26] NightMonkey: you could see if it's stil present in /var/cache/apt/archives/ [18:31] nacc: Ah, thanks for your help. Sadly, I looked at the PPA after you pointed that out, and indeed, the old version is just gone. :/ [18:31] * NightMonkey rages at an indifferent sky [18:31] Luxray5474: hrm. I can't spot it :/ [18:32] me neither [18:39] Luxray5474: whats the issue [18:42] cryptodan: I was trying to set my server's IP to a static IP; 192.168.1.75. I found that making a config in /etc/netplan called 01-netcfg.yaml would allow me to do just that. So I did. But iI believe i did something terribly wrong and now, my server can't connect to the internet. I suspect it's because /etc/services/ doesn't have netplan/tcp. [18:42] cryptodan: my config is in this pastebin: netplan/tcp not found in /etc/services, using ports 2983 and 5444 [18:42] oops [18:43] https://pastebin.com/cBBGXzzr [18:43] Luxray5474: wait a sec [18:43] netplan != nplan [18:43] Luxray5474: why not just use /etc/network/interfaces [18:43] i'm on 18.04 and it doesn't exist whenever i try to cd to it [18:44] Luxray5474: you need to do some work to go back to that if you want [18:44] yes please [18:44] Luxray5474: iirc, reinstall ifupdown, then dpkg-reconfigure resolvconf [18:44] you should have an interfaces file [18:44] Luxray5474: where are you seeing that message? [18:44] cryptodan: no, you sholdn't not in 18.04 [18:45] well, not necessrily, at least [18:45] "This new tool [netplan] replaces the static interfaces (/etc/network/interfaces) file that had previously been used to configure Ubuntu network interfaces. Now you must use /etc/netplan/*.yaml to configure Ubuntu interfaces." -https://websiteforstudents.com/configure-static-ip-addresses-on-ubuntu-18-04-beta/ [18:45] Luxray5474: yes, we know that :) [18:45] Luxray5474: ok, 1) where do you see the netplan/tcp message ? [18:46] 1. I make the config. [18:46] 2. I run "sudo netplan apply" [18:46] 3. spits out the message as the last line, and starts using ports 2983 and 5444 [18:47] now, i try to reconfig, and it says another daemon is using the ports, even when i kill the netplan daemon [18:47] Luxray5474: `ls -ahl /usr/sbin/netplan` ? [18:47] nacc: btw where'd you spot 'nplan'? [18:48] sarnold: well, i did a quick google for the message pasted, there's a bug from 1998 mentioning it for the *netplan* package [18:48] "-rwxr-xr-x 1 root root 40K Apr 3 12:39 /usr/sbin/netplan" [18:48] which is not the nplan package (which is what is netplan.io) [18:48] yeah [18:48] nacc: oh god. [18:48] Luxray5474: did you try and install netplan at some point? [18:48] the package netplan that is [18:48] it comes installed [18:48] `apt-cache policy netplan` [18:48] no, *nplan* comes installed [18:48] is it too late to give this thing a name that wasn't taken twenty years ago? P) [18:49] they are very different packages [18:49] and I ofc can't install anything bc network is unreachable [18:49] sarnold: just a bit too late, yes [18:49] :) [18:49] Luxray5474: please pastebin the above (`apt-cache policy netplan`) [18:49] but wth, we've changed twice already [18:49] sarnold: i then did a grep in the srouces and didn't see the strings, [18:49] sarnold: so i'm thinking that's the bug here, but i'm waiting on confirmation [18:50] dpb1: insert joke about snap and click names colliding with existing projects and traditions.. [18:50] nacc: Luxray5474: the name of the binary is always "netplan", but the name of the package is "nplan" in 16.04 and 17.10, and "netplan.io" in 18.04 [18:51] sarnold: ha [18:51] oh [18:51] cyphermox: right, but if a user has "netplan" installed, it *also* provides a netplan binary [18:51] nacc: that should conflict [18:51] is that handled correctly by nplan/netplan.io [18:51] cyphermox: i agree, but i wonder what is up with the above user's system [18:51] because on my systems /usr/sbin/netplan is a symlink [18:51] (one an upgrade, one a fresh install) [18:51] to /usr/share/netplan/netplan.script [18:52] yes [18:52] it's too long for me to type, and my server is not in the best spot for me to be looking back and forth to, so: https://i.imgur.com/DRq3LOT.jpg [18:52] Luxray5474: yeah, so you have netplan installed which you don't want [18:52] Luxray5474: yes, that's the netplan you don't want ;) [18:52] oh [18:52] Luxray5474: can you take a pciture of `apt-cache policy nplan netplan.io` ? [18:53] I'm surprised, because 18.04 should have netplan.io installed by default, and when you try to install "netplan" you'll get semi-loud warnings about removing things [18:53] cyphermox: agreed [18:54] https://i.imgur.com/LlmQqQK.jpg [18:54] urgh [18:54] you don't have netplan installed at all :/ [18:54] (netplan as in the network configuration tool) [18:54] Luxray5474: is this a fresh install? [18:54] yes. fresh install this morning [18:54] That makes sense as it sounds like the two packages conflict? :) [18:54] cyphermox: --^ ... not good [18:54] unless they were using old install media, maybe? [18:54] rbasak: yeah :) [18:55] okay so I guess instal of doing netplan apply i do netplan.io apply or nplan apply? [18:55] Luxray5474: can you look in /var/log/apt/history.log to see when 'netplan' was installed? [18:55] okay [18:55] Luxray5474: no, you don't hve the correct package installed [18:55] Luxray5474: which is confusing, because you say it's a fresh install and the "netplan" (the package) is not installed by default and "netplan.io" (the package) should be installed by default [18:56] Luxray5474: is this stock Ubuntu etc.? [18:56] (sorry for all the questions, just want to understand how you got in this situation) [18:56] steps i took to install: [18:57] 1. Grabbed Ubuntu Server fomr the official site [18:57] 2. Used Rufus to burn it onto a USB thumb drive, ISO mode. [18:57] from* [18:57] 3. Installed as normal and booted as UEFI [18:57] Luxray5474: ok, any luck with the apt log? [18:57] tryna get it to focus properly so you can see the text [18:58] Luxray5474: sure [18:59] Luxray5474: as background, there are two pakcages that provide the "netplan" binary. One is old and one is new. The old one is what you have installed, but has nothing to do with netplan.io or the replacement of ifupdown (/etc/network/interfaces). So you've been asking something totally unrelated to "apply" netplan config files [19:00] now, i think it's rather funny (not haha) that the old netplan doesn't barf on an "apply" subcommand :) [19:00] just a question why remove things that worked and worked well for users to replace it with more complexity [19:01] cryptodan: because ifupdown has well known limitations [19:01] ifupdown definitely didn't work well for users [19:01] cryptodan: and it's not been removed, it's just not hte default [19:01] now its uploading [19:01] (when trying to do anything remotely complicated) [19:01] it has worked well for me on multiple linux distros including ubuntu [19:01] cryptodan: that's nice. [19:01] i was trying to get the right angle so it doesn't reflect the bright light from outside and it doesn't reflect my whole house lol https://i.imgur.com/HRhpYMy.jpg [19:02] Luxray5474: thanks! [19:02] and it is barely visible now that i look at it [19:02] Luxray5474: ok, so that says at 2:23 your time, you issued "apt install netplan" [19:02] that is the problem [19:02] don't do that :) [19:02] ohhh [19:02] that in turn removed netplan.io [19:02] and probably told you it would have to [19:02] okay so i guess i have to do a whole reinstall? [19:02] it also removed ubuntu-minimal [19:03] Luxray5474: no, i think you can do `sudo apt install netplan.io nplan ubuntu-minimal` [19:03] that should remove 'netplan' in response [19:03] but it isn't connected to any network [19:03] it might have it cached [19:03] i'll try [19:04] yep it can't :P [19:04] Luxray5474: and then problem remove ifupdown again (as you don't need it) [19:04] Luxray5474: ok, it's probably fastest to reinstall then honestly [19:04] okay lol [19:04] you could also probably configure your static network by hand (`ip`) [19:04] just to do the apt commands [19:04] dhclient $NIC [19:05] well thanks for all the help you guys :D [19:05] sdeziel: i think they were doing static networking [19:05] ah\ [19:05] i'm a noob with linux servers, and still learning in regards to servers in general [19:05] Luxray5474: but if you do have a dhcp server, sdeziel's suggestion would be ideal [19:10] holy [19:10] oh my god. [19:10] Oh [19:10] so [19:10] I think I know why he'll have done that. [19:10] i just found that... when you install it in the first place... you can set a static ip. [19:10] Type "netplan apply" and does command-not-found suggest typing "sudo apt install netplan"? [19:11] i.... i'm a complete idiot for not realizing that XD [19:12] Hmm. No. It does nothing (ie. it follows the symlink) [19:12] So I don't know. [19:22] what should be the search domain? [19:22] rbasak: right, i didn't think cnf knew about the arguments to the binary, cool if it does :) [19:22] I know the nameservers i want to use and I put in google's and cloudflare's DNS [19:23] Luxray5474: yes, re: static ip at install time :) [19:23] Luxray5474: you might not have any, it depends on if you are in a domain [19:24] also it says that 192.168.1.75 isn't included in the subnet 255.255.255.0 when i obviously have a similar ip and the same subnet on my windows machine (one i'm typing with rn) [19:25] I put in 255.255.255.0/24 in the subnet field btw [19:27] Luxray5474: you seem to be combinig two fields [19:27] you want either 192.168.1.0/24 or 255.255.255.0 iirc [19:27] ? [19:27] 'search' just means how many wrong names do you want to look up before getting right names :) if you owned e.g. luxray.example.com. as your own domain, you could have machines www and ns1 and irc and so on, www.luxray.example.com, ns1.luxray.example.com, .. [19:28] so if you stuffed 'luxray.example.com' into the search field, your resolver would look for www.google.com.luxray.example.com and then www.google.com ... [19:28] nacc: if i only enter the latter, it says it has to be in CIDR notation [19:28] sarnold: that's pretty neat [19:28] i just had an interesting question asked of me - namely why is the MySQL server version in the repositories so old even for Bionic. Anyone want to give me a proper answer to give them? [19:29] (MySQL 8 exists upstream now so_ [19:30] teward: we're working on MySQL 8 packaging. [19:31] teward: 5.7 isn't particularly old. [19:31] did debian decide to stop packaging mysql and focus on the altneratives instead/ [19:31] 8.0 was released GA only on April 19 this year. [19:32] The release team removed MySQL from testing in Debian. We're still maintaining it together with MariaDB in unstable. [19:32] Luxray5474: which field are you typing this in? (picture)? [19:33] rbasak: i know it's not particularly old. Just proxying the request ;P [19:33] sarnold: isn't the search domain only used for non-FQDN names? Unless you have unusual/non-default options in /etc/resolv.conf ? [19:33] I *think* it's used to try and resolve any name that fails to resolve [19:33] (on its own) [19:34] as a FQDN isn't known to be a FQDN until it resolves (in my head) [19:34] sdeziel: no one ever types the . though, heh :) [19:35] by default, having at least 1 dot means the initial query will be tried first then search will be attempted [19:35] hrm, really? that's not my recollection, lst time i looked at it [19:35] sarnold: I didn't mean '.' (root) but in your example www.google.com this has dots [19:35] I thuoght it required the finial . to skip the search [19:35] quick, somebody strace! [19:35] man resolv.conf see ndots [19:36] interesting [19:36] bummer, no notes on when it was introduced [19:40] could be I just noticed 401202 instead and assumed I saw something else, last time I used search [19:46] sarnold: looks like it was there in 2008 at least, looking at the git-ubuntu repository for the manpages package. [19:46] pkg archaeology ftw [19:52] rbasak: oh handy :) [21:02] anyone know the right way to call depmod in a chroot when updating the netboot image with newer ethernet drivers [22:13] /nick Epx998 === Epx998- is now known as Epx998 [22:14] Install hung at finishing installation -> preseed, never seen that before [22:14] Epx998: you didn't miss anything while you were away [22:14] yeah [22:17] Aww yeah [22:18] I got almost everything to work now [22:18] lies [22:18] work is STILL making me build systems on ub12 buh [22:18] Luxray5474: nice, so the reinstall worked ok? [22:18] Yep [22:18] Now have Plex and vsftpd running [22:18] (after much frustration of course) [22:19] I'm gonna install UnrealIRCd, and node (for my IRC bot) [22:19] that reminds me, is there any remote server management applications like Pulseway, but for linux [22:19] ?* [22:20] dozens? [22:21] typically the answer to "is there any ... " queries is "yes". [22:21] and or ltmgfy [22:21] lmgtfy rather! :) [22:22] is there any for android tho [22:22] (like pulseway) [22:22] cf my answer above. [22:22] as in, go google it? [22:22] I installed a nagios frontend on my phone once [22:23] it was meant to motivate me to set up nagios [22:23] pulseway has an android app [22:23] sarnold: lol [22:23] but nagios looks kinda crunky [22:23] and the replacements are crunky [22:23] and prometheus looks neat but also looks like deploying it is a bit of a procedure [22:24] * nacc pictures sarnold dancing to crunk [22:24] crunky ain't dead man [22:24] heh [22:24] how do you do those texts? https://s15.postimg.cc/64hfvpwjv/screenshot_61.png [22:24] of course reading the pcp sources and docs for the pcp MIR made me want to use it on my own systems.. and netflix's vector thing looks neat.. [22:25] Luxray5474: /me foo [22:25] * sarnold foo [22:25] oh [22:25] thanks lol [22:29] really off the subject at hand, who else likes i n t e n s e t e r m i n a l f l o o d ? [22:31] it happens a lot when i tail all Plex's logs [22:43] less helps with that [23:32] or maybe go the opposite way and use ccze or something to make it much noisier still