[00:39] <Phanes> hello
[00:39] <Phanes> where is the ubuntu maintainer and developer channel
[00:53] <teward> Phanes: the general development channel is #ubuntu-devel, but if you're trying to reach the maintainer of a specific package, filing a bug is more useful
[00:53] <teward> anything under the Server Team's purview, we can poke here
[01:00] <Phanes> teward, im building a new distro so was looking for a doc that outlined the considerations made when deciding the components of the system and how they'd be implemented
[01:01] <Phanes> or a channel to spray questions into
[01:06] <teward> Phanes: not sure what kind of support you'll get from Ubuntu with that, just saying.
[01:07] <Phanes> what
[01:07] <Phanes> why
[01:07] <teward> there's really no specific channel I can direct you to for that...
[01:07] <teward> because there's more than just the "maintainer" and "Developer" aspect
[01:07] <Phanes> if it is worth a damn, they'll have documented it
[01:10] <Phanes> surely
[01:14] <tomreyn> i would bet that not all stretegic documents are open for the general public.
[01:14] <tomreyn> *strAtegic
[01:15] <tomreyn> there are those blueprints on launchpad, however, but that's mostly drafts.
[01:15] <Phanes> surely they're not just hoping good will somehow makes engineering happen
[01:16] <Phanes> please tell me canonical is smarter than that
[08:46] <DK2> what is "killall -0" doing?
[09:59] <Genk1> hello folks
[09:59] <Genk1> I have a basic question. I want to synchronize two web repertories /var/www on server 1 and /var/www on server 2
[10:00] <Genk1> I have choosed to use unison for two-direction synchro feature
[10:00] <Genk1> my problem is how to figure out the best strategy to deal with permissions in such situations ?
[10:01] <Genk1> do I have to set www-data as an ssh account in both servers ?
[10:01] <Genk1> Do I need to setup a new account who has r/w access to /var/www ? and use it for the synchronization ?
[10:02] <Genk1> what kind of stuff do people choose in such situations ?
[10:03] <Genk1> some documentation suggest to use sudo ?
[10:04] <Genk1> is this a healthy way to deal with such problems ?
[10:25] <ziyourenxiang> better to use a new account and keep www-data read only (assuming your web server is running as www-data)
[10:32] <maxb> unison sounds like a bad idea because it inherently requires human interaction to resolve in which direction changes need to be propagated
[11:35] <Genk1> ziyourenxiang, OK thanks for your advice
[11:36] <Genk1> maxb, you're some how right but can you suggest a better option ?
[11:37] <Genk1> maxb, I was thinking about distrubuted FS like Gluster and CEPH but it seems to me that it's a little bit complex to manage
[12:14] <ziyourenxiang> other possibilities: if all your content are public then just mirror one of your web server from the other web server using wget or similar
[12:14] <ziyourenxiang> or use rsync driven by one side
[12:15] <ziyourenxiang> or follow any number of tutorials you can find to use git to 'publish'
[12:34] <tomreyn> NFS would have been an option, but he left
[15:04] <MASM> someone here, i have a problem with mdadm resync, the speed y so slow, '  finish=95954.6min speed=168K/sec '
[15:16] <Poster> In my experience, slow resync with mdadm is generally caused by moderate/high IO to the device in question
[15:43] <MASM> when i start sync all system, the speed decreased, but i cancel the resync of mdadm and speed of hard disk get normal right now
[15:44] <MASM> i change the dev.raid.speed_limit_min
[15:45] <MASM> and it didn't change anything
[15:47] <Poster> There is a finite amount of bandwidth available to each device that makes up your mdN ; if you are writing to either the md or any other partitions on the physical disk, all contribute to the total amount of bandwidth to the device
[15:55] <MASM> Poster: you mean that will be the connector of the raid that is damaged?
[15:56] <MASM> when i start the sync all speed decrease in both disk, but when i stop, i check speed in both but it is ok
[15:58] <lordievader> MASM: What Poster means is that the available bandwidth of the drives is shared over the sync action and any other action it performs. Hence the speed dwindles if a sync action is going on.
[15:58] <Vamp898> Hi there. I used this guide to setup SSSD on Ubuntu 14.04 https://help.ubuntu.com/lts/serverguide/sssd-ad.html and everything works perfect. But when i use the exact same guide for Ubuntu 16.04, i can join the AD, i can use getent to get user ids and groups and so on, everything works almost fine. Everythig except logging in (no matter if i use su,
[15:58] <Vamp898>  ssh or an DM) does work. Login as an user does not work (i type the password an then get "System error". When i switch user with root it works, but i also get there "su: system error"
[15:59] <Vamp898> I tried this on a completely new/fresh Ubuntu 16.04 installation
[15:59] <MASM> lordievader: the only solution is?, restart in security mode, and do it?
[16:00] <lordievader> MASM: What? There is no solution, this is a hardware limitation. Each drive has a maximum bandwidth...
[16:01] <MASM> lorddievader: but "speed=168K/sec" is so slow, i saw in internet that the speed is more than this :S
[16:02] <lordievader> MASM: Oh, it could very well be that mdadm limits the sync speed in order to keep the raid usable.
[16:03] <Poster> I don't think it's that low by default
[16:03] <Poster> if you really want it to speed up, consider booting into single user mode to allow the resync to be (more or less) the only thing accessing the disks
[16:05] <MASM> the problem is hosting :S and don't see directly
[16:06] <MASM> i have a server dedicate
[16:06] <MASM> it is possible?
[16:07] <Poster> I don't understand the question or solution you are proposing
[16:09] <MASM> sorry, English isn't my native lenguaje, to consider booting into single user mode, is posible if i have a server dedicate, if the server isn't here with me?
[16:10] <MASM> the ssh will work?
[16:10] <Poster> single user mode has no network connectivity, you would need console access
[16:10] <Poster> well, you could start it I think
[16:10] <Poster> but generally console access is what you would need
[16:10] <Poster> either via KVM switch or if your system is virtualized, via whatever virtualization manager
[16:12] <MASM> Poster: i understand.
[17:07] <jge> anyone used chrony before? I'm trying to figure out how I would query another client for stats
[17:08] <jge> looking for an equivelant of ntpq -p
[17:25] <soop> anyone familiar with the official ubuntu mail-filter howto (spamassassin/amavis/postfix)?
[17:45] <tomreyn> where is that?
[17:51] <soop> https://help.ubuntu.com/lts/serverguide/mail-filtering.html
[17:51] <soop> there
[17:54] <tomreyn> i guess my question already indicated that i am not familiar with it.
[17:56] <jge> nvm, looks like chronyc -h someip tracking does what I want
[21:43] <ws2k3> my ubuntu machine has a gigabit nic and is connected to a gigabit switch but it still shows 100Mb/s what can this be?
[21:45] <mybalzitch> failed autonegotiation, bad cable would be my guesses
[21:56] <soop> are you sure your port is gigabit?
[22:01] <ws2k3> soop yes
[22:01] <ws2k3> soop i probarly need to find the source of the issue on the cisco switch
[22:09] <JanC> can be a driver issue too
[22:12] <patdk-lap> can be most anything
[22:12] <patdk-lap> every wire must work, and configs set to autodetect
[22:13] <patdk-lap> or you dont' get gigabit
[22:13] <JanC> I've seen a NIC fall back to 100 or 10 Mbit/s depending on the kernel (& thus driver) version used
[22:14] <JanC> or Gbit/s
[22:29] <compdoc> it does need all 8 wires in the connectors. if you made any of the cables, then you are likely the cause
[22:33] <ws2k3> compdoc they are not self made connectors just patch cable from factory so unlikeley the issue is in the cable
[22:59] <patdk-lap> unlikely? I receive lots of bad patch cables
[22:59] <patdk-lap> like 100 bad black patch cables, I don't use black ones anymore
[23:00] <patdk-lap> funny only black, the grey, blue, green, red, where all fine