jhutchins 2017-03-05 11:45:25
Klaus_Dieter: The traditional Linux partitioning scheme really doesn't make sense with modern hardware. Back when hard drives were small and expensive and systems were all multi-user it made sense. These days it just guarantees that you will run out of spave on one partition when you have plenty on another.
jhutchins 2017-03-05 11:47:01
I usuall receommend (and practice) one partition + swap unless you know of a specific reason you need something else.
userro 2017-03-05 11:51:45
lulwut
Klaus_Dieter 2017-03-05 11:51:47
jhutchins: this will also give me a heads-up if there is a new kernel in dist-upgrade as that will then fail...
jhutchins 2017-03-05 11:51:50
Oracle servers tend to have their own special partitioning scheme (usually on NAS).
Klaus_Dieter 2017-03-05 11:52:42
unfortunately oracly is a different can of worms entirely... with everything. You cannot trust a software having version 12 or whatever if it never had version 1.
Klaus_Dieter 2017-03-05 11:53:20
by version 2 at the latest you already know that the marketing department had taken over. ;)
Klaus_Dieter 2017-03-05 11:54:01
jhutchins: actually for me it does not matter if the device is attached physically or over san.
jhutchins 2017-03-05 11:54:16
Given databases that can take well over 24 hours to move over high speed connection, Oracle is a fact of life.
Klaus_Dieter 2017-03-05 11:54:56
unfortunately yes. still applications using it have slightly more incidents than the ones running db2.
jhutchins 2017-03-05 11:55:31
I don't see things like the Federal payroll or the DMV registry for Texas on db2 these days.
Klaus_Dieter 2017-03-05 11:56:11
I am seeing a lot of banking backend software...
jhutchins 2017-03-05 11:56:11
100 years of Aviation Week and Space Technology is actually on MySQL.
Klaus_Dieter 2017-03-05 11:56:22
interesting :)
Klaus_Dieter 2017-03-05 11:56:54
frankly that probably is a good choice given its foss nature....
tw 2017-03-05 11:57:07
I disagree that partitioning doesn't have a place; /home, /srv, and /var/lib should run out of space before /var/log and /run (if /run is not tmpfs)
theartificial 2017-03-05 11:58:06
Quick unrelated question. Two computers and one email/keybase. Should I have a 3 ssh keypairs or one for all?
tw 2017-03-05 11:58:07
As soon as you get into a position where a user can induce init to start vomiting, something is designed wrong.
rx777 2017-03-05 11:58:25
Hello, I am experiencing an issue described here https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=844487 though I am not using disk encryption. Does anybody know how to fix it? Thanks
judd 2017-03-05 11:58:26
Bug http://bugs.debian.org/844487 in src:linux (open): «linux-image-4.8.0-1-amd64: Debian fails to prompt for encrypted disk password on boot»; severity: important; opened: 2016-11-16; last modified: 2016-11-16.
Klaus_Dieter 2017-03-05 11:58:41
theartificial: it probably depends on your use case. If in doubt chose the 3 pairs solution.
Klaus_Dieter 2017-03-05 11:58:59
tw: ack.
Klaus_Dieter 2017-03-05 11:59:34
rx777: how can you suffer from debians failure of asking for disk encryption if you do not even use it?
Klaus_Dieter 2017-03-05 11:59:47
asking for disk ecnrpytion passwords
rx777 2017-03-05 12:00:34
Klaus_Dieter: I am getting the same message the user gets, and then Debian boots.
rx777 2017-03-05 12:00:48
tomreyn: you are right, I'm sorry.
rx777 2017-03-05 12:01:09
I get the same messages and then Debian boots fine.
rx777 2017-03-05 12:01:41
I'd like to know what's causing these messages to appear.
theartificial 2017-03-05 12:02:24
Klaus_Dieter: SSH and signing things. Wondering if the security outweighs the convenience of using one keypair.
tomreyn 2017-03-05 12:03:50
rx777: lvmetad does not run this early in the boot process, so lvm initialization using it fails, and the system falls back to device scanning. that's all this *warning* (not error) message is about.
rx777 2017-03-05 12:05:31
tomreyn: I understand. Is there some way to disable it? (just because I don't enjoy seeing warnings, but I could live with it)
Klaus_Dieter 2017-03-05 12:05:49
theartificial: at this point it is more about flexibility. how about you want to revoke one of those keys but in the beginning you picked to use just one?
tomreyn 2017-03-05 12:06:26
rx777: there is definitely some way, but i don't know it. you'd start by looking at the scripts contained in your initrd.
tw 2017-03-05 12:07:06
theartificial: If you're propagating your signing key to both hosts, I would just use the one. If you're generating different signing subkeys for each host, I would just use the host-specific signing subkey for ssh auth.
rx777 2017-03-05 12:08:43
tomreyn: thank you. Will look at it and let you know if I found a way :)
Klaus_Dieter 2017-03-05 12:08:49
tw: good catch I was assuming the latte rscenario.
Klaus_Dieter 2017-03-05 12:08:59
but actually the situation is not quite clear.
theartificial 2017-03-05 12:09:31
tw: Well I need to ssh into both computers from each other. Also I'd like to sign things with one keypair from both hosts.
zombo 2017-03-05 12:10:28
Klaus_Dieter: about the tasks; I don't use them because I'm a newbie (to computers anyway), I use them because in the last couple of years I've noticed that if I try to build my install up from the bare bones, some things magically won't work and I will have to scour forums and IRC, and nobody will be able to figure out why, and that is on a stable install (I personally blame a certain recent hotly debated change in debian). Somehow when I install stuff by vi
unborn 2017-03-05 12:26:52
heh
theartificial 2017-03-05 12:27:18
tw: I am dumb. You sign things with a pgp keypair not ssh.
tw 2017-03-05 12:27:58
theartificial: yes, but you can ssh with a pgp keypair.
theartificial 2017-03-05 12:28:25
tw: how?
tw 2017-03-05 12:29:11
gpg-agent.conf enable-ssh-support
tw 2017-03-05 12:29:56
It takes some setting up. And as far as I know, it's the easiest way to get ssh keys on a smartcard.
tw 2017-03-05 12:30:04
(which is how I know it exists)
NapoleonWils0n 2017-03-05 12:33:12
hi all
teraflops 2017-03-05 12:33:29
NapoleonWils0n: o/
NapoleonWils0n 2017-03-05 12:33:34
debian wiki kvm section seems outdated
NapoleonWils0n 2017-03-05 12:33:39
hi teraflops
NapoleonWils0n 2017-03-05 12:33:53
libvirt-bin package doesnt exist
teraflops 2017-03-05 12:33:57
not the only one sadly
NapoleonWils0n 2017-03-05 12:34:26
anyone know what the correct package for qemu-kvm and libvirt-bin are
teraflops 2017-03-05 12:34:51
,v libvirt-bin
judd 2017-03-05 12:34:52
Package: libvirt-bin on amd64 -- wheezy: 0.9.12.3-1+deb7u1; wheezy-security: 0.9.12.3-1+deb7u2; wheezy-backports: 1.2.9-9+deb8u2~bpo70+1; jessie: 1.2.9-9+deb8u3; jessie-security: 1.2.9-9+deb8u3
NapoleonWils0n 2017-03-05 12:35:14
im on unstable
NapoleonWils0n 2017-03-05 12:35:28
no hits
teraflops 2017-03-05 12:35:35
NapoleonWils0n: well, this channel is mainly for stable
NapoleonWils0n 2017-03-05 12:35:45
sorry my bad guys
bazhang 2017-03-05 12:35:54
NapoleonWils0n, check the topic here
tw 2017-03-05 12:35:54
on stretch, you want libvirt-daemon%
tw 2017-03-05 12:36:04
but yes, OT.
NapoleonWils0n 2017-03-05 12:36:05
tw cheers mate
NapoleonWils0n 2017-03-05 12:36:35
thought that might be the one, but wasnt sure
NapoleonWils0n 2017-03-05 12:37:10
switched main main from arch to debian unstable
NapoleonWils0n 2017-03-05 12:37:13
machine
teraflops 2017-03-05 12:37:16
NapoleonWils0n: no worries just saying
NapoleonWils0n 2017-03-05 12:37:47
does that mean all the people here are stable then :)
teraflops 2017-03-05 12:38:20
NapoleonWils0n: I dont think so xD
teraflops 2017-03-05 12:38:53
NapoleonWils0n: you can ask about sid in #debian-next at oftc network
NapoleonWils0n 2017-03-05 12:40:21
still getting used to debian names for packages
NapoleonWils0n 2017-03-05 12:41:00
on arch you have openbsd-netcat on on debian its netcat-openbsd
teraflops 2017-03-05 12:41:43
what about wpasupplicant?
teraflops 2017-03-05 12:41:45
xD
