=== ksx4system_ is now known as ksx4system | ||
=== ksx4system_ is now known as ksx4system | ||
=== ksx4system_ is now known as ksx4system | ||
Geom | is there a signifacant changes in performance running a lower version of server? | 04:04 |
---|---|---|
=== rizzy_ is now known as riz0n | ||
=== rizzy_ is now known as riz0n | ||
Vmong | guys.. | 10:11 |
Vmong | can anyone help me with copy / paste issue between ubuntu vmware and host computer | 10:11 |
Vmong | thanks in advance | 10:11 |
patdk-lap | vmong, read the vmware manual | 11:38 |
=== PaulW2U is now known as pcw | ||
sypher | I'm curious as to why Ubuntu handles bonded interface configurations differently than Debian. http://paste.ubuntu.com/22444516/ | 13:11 |
sypher | Well, more specifically, I know that in trusty, this was a consequence of upstart, though I never got under the hood to determine why. With xenial on systemd, I'm only more perplexed. | 13:13 |
qman__ | both configurations _should_ work on both | 13:20 |
qman__ | specifyinng bond-master or bond-slaves does operate slightly differently and I usually do the latter config | 13:21 |
qman__ | I've not verified myself that they do on the current versions, but if they don't, that's a bug | 13:23 |
sypher | qman__: I'm going to make sure I'm not a moron (definite possibility), but the first style generally doesn't work correctly in xenial. I'll try it again. | 13:24 |
qman__ | well, the catch with the first style is that if the slaves aren't up, the bond doesn't come up | 13:25 |
qman__ | the second style brings the bond up independently of the slaves, then the slaves bring themselves up as they can | 13:25 |
qman__ | that's why I prefer the second | 13:25 |
sypher | qman__: i.e. if the switchports are off? | 13:25 |
qman__ | yes, or unplugged, or a driver bug or race condition messes things up | 13:25 |
sypher | Should I be setting the child interfaces to hotplug instead of auto, then, to take better advantage of that? | 13:26 |
qman__ | I'm not actually sure on the difference between those two settings, let me check what mine are at | 13:26 |
qman__ | yeah, mine are allow-hotplug | 13:28 |
sypher | As I understand it, allow-hotplug allows the interface to come up or down based on connection events. | 13:30 |
sypher | Rather than "it must be there, or die." | 13:30 |
patsToms | morning. So if I a do nmap to ubuntu 14 server I it shows that there is "25/tcp filtered smtp" open. But when I run netstat in server using root - it don't show any activity in port 25 | 13:42 |
patsToms | is there any way to found what keeps that port open? | 13:42 |
sypher | filtered doesn't mean open. | 13:49 |
sypher | filtered means no response at all. | 13:49 |
qman__ | right | 13:49 |
qman__ | are you running this scan over the internet, or locally? | 13:49 |
sypher | A closed port will still generate a response in the form of a TCP RST. | 13:50 |
sypher | A filtered port means that your incoming TCP SYN just got dropped into the ether. Not even a RST response. | 13:50 |
qman__ | most residential internet providers filter port 25 to prevent spam, and even some business ones do unless you specifically ask for it to be open | 13:51 |
sypher | qman__: Thank you for the pointers on lag config, btw. | 13:56 |
Geom` | w00t | 14:05 |
Geom` | srill no reply re my concern? | 14:06 |
ikonia | what concern | 14:06 |
ikonia | you've just joined - said nothing then said "w00t" | 14:06 |
ikonia | what do you expect ? | 14:06 |
Geom` | is there any significant change in speed n n lower server version compared to the latest? | 14:08 |
ikonia | what ??? | 14:08 |
* sypher just stares blankly. | 14:08 | |
Geom` | say.... 12.0.4 vs 16 (server version) is there performance difference in speed wise | 14:10 |
ikonia | no | 14:10 |
ikonia | not really | 14:10 |
sypher | Geom`: You also need to be concerned with the fact that support for 12.04 ends in, what, eight months? | 14:10 |
Geom` | ahh ok. downloading 16 then | 14:13 |
Geom` | i only have c2d with 4 gb ram. speed is my concern | 14:14 |
ikonia | thats fine | 14:14 |
ikonia | it really depends what you are going to do and the load on it | 14:14 |
ikonia | thats a lot of resource | 14:14 |
Geom` | file sharing and video conversion | 14:15 |
ikonia | so thats fine | 14:15 |
ikonia | you could do that on a raspberry pi | 14:15 |
Geom` | Great! | 14:15 |
Geom` | thanks ikonia :) | 14:16 |
sypher | Geom`: btw, 16.04 has been GREAT for me so far. | 14:16 |
sypher | Geom`: You'll love it. | 14:16 |
Geom` | i hope sypher... installing now | 14:19 |
Geom` | will it even run fine on a P4? | 14:20 |
qman__ | yes, though you can run into bugs with kernel modesetting on some video chips from that era | 14:22 |
qman__ | it will run on systems as old as the Pentium II | 14:22 |
qman__ | unfortunately it does not run on my AMD K6, which lacks the i686 CPU extensions | 14:23 |
qman__ | the last version that had a kernel compatible with it was 10.04 | 14:23 |
Geom` | i also have the K6. il keep that in mind if ever il try it | 14:30 |
qman__ | well, it required manually installing the -386 kernel, the standard kernel won't run on it, but 10.04 was the last version where the -386 kernel existed | 14:34 |
Geom` | im not a hardware guy.. what i have is this http://specsen.com/motherboard-epox/epox-ep-8hda5-/ is this K6? | 14:37 |
Geom` | im not so sure what the cpu is | 14:37 |
Geom` | oh its K8 :) | 14:41 |
teward | Geom`: um, in the modern era that's the Athlon 64 chips. (They don't use hte K nomenclature anymore) | 15:35 |
teward | (it's an amd64 arch) | 15:36 |
teward | (I have at least twelve such series' chips lying around heh) | 15:36 |
teward | blah nevermind me :) | 15:39 |
=== codedmart_ is now known as codedmart | ||
=== pavlushka is now known as Guest31425 | ||
=== Guest31425 is now known as pavlushka | ||
=== dzragon^ is now known as dzragon | ||
=== dzragon^ is now known as dzragon |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!