Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-07-01

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:17 saibarspeis joined #fuel
00:30 eshumakher joined #fuel
00:48 saibarspeis joined #fuel
01:05 jaypipes joined #fuel
01:26 mattgriffin joined #fuel
01:49 rongze joined #fuel
02:02 xarses joined #fuel
02:14 mattgriffin joined #fuel
02:37 IlyaE joined #fuel
03:05 rongze joined #fuel
03:21 IlyaE joined #fuel
03:56 book` joined #fuel
04:15 nurla joined #fuel
04:22 dilyin joined #fuel
04:22 kat_pimenova joined #fuel
04:22 jkirnosova joined #fuel
04:22 sb joined #fuel
04:22 Bomfunk joined #fuel
04:22 ykotko joined #fuel
04:23 nurla joined #fuel
04:26 justif joined #fuel
04:27 evgeniyl joined #fuel
04:56 rongze joined #fuel
05:02 saibarspeis joined #fuel
05:28 e0ne joined #fuel
05:29 IlyaE joined #fuel
05:44 e0ne joined #fuel
05:50 e0ne joined #fuel
05:57 e0ne joined #fuel
06:04 taj joined #fuel
06:10 e0ne joined #fuel
06:12 ddmitriev joined #fuel
06:22 e0ne joined #fuel
06:24 IlyaE joined #fuel
06:33 e0ne joined #fuel
06:41 e0ne joined #fuel
06:42 strictlyb joined #fuel
06:47 e0ne joined #fuel
06:55 omelchek joined #fuel
07:02 artem_panchenko joined #fuel
07:28 al_ex6 joined #fuel
07:39 e0ne joined #fuel
07:56 IlyaE joined #fuel
08:04 IlyaE joined #fuel
08:05 e0ne joined #fuel
08:28 brain4611 joined #fuel
08:36 guillaume__ joined #fuel
08:39 brain4611 joined #fuel
09:14 evg_ CTWill: it seems 20MB
09:52 monester joined #fuel
10:20 brain4611 joined #fuel
11:02 agordeev joined #fuel
11:08 e0ne joined #fuel
11:16 akupko joined #fuel
11:19 eshumakher joined #fuel
11:31 rongze joined #fuel
11:50 igajsin joined #fuel
11:56 akupko joined #fuel
12:10 eshumakher joined #fuel
12:11 al_ex6 joined #fuel
12:15 al_ex7 joined #fuel
12:25 brain4611 joined #fuel
12:26 TVR_ OpenStack installs working as expected...
12:27 TVR_ figured with everyone posting about broken stuff, it would be nice to hear about a working set
12:34 al_ex7 joined #fuel
13:23 obcecado_ i had problems when i mixed roles
13:23 obcecado_ namely storage and controller
13:58 TVR_ I always run controller+OSD and then Compute+OSD in my environments
14:05 obcecado_ with neutron? gre or 802.1q?
14:11 jobewan joined #fuel
14:27 IlyaE joined #fuel
14:38 wrale_ How can I figure out which controller has the VIP for Horizon?  The arp table for that IP shows a MAC not present on any controller.
14:39 wrale_ per 'ip link show' | grepping for the mac
14:45 MiroslavAnashki- wrale_: `crm_mon -1Ar` on any controller in HA mode. Or `crm status`
14:46 MiroslavAnashki- And check vip__public_old location
14:47 wrale_ MiroslavAnashki: thank you
14:47 wrale_ that worked
14:48 wrale_ Can you think of any way to check the vip__public_old IP's netmask and gateway?
14:49 wrale_ I cannot reach this IP from a nearby subnet
14:52 wrale_ (the VIP's netmask should be 255.255.252.0 for my subnet)
14:52 wrale_ i'm wondering if it isn't hard coded as 24 bit
14:56 MiroslavAnashkin This VIP is inside the haproxy namespace. Please go to that node where this VIP is currently running and check its settings with
14:56 angdraug joined #fuel
15:00 MiroslavAnashkin `ip netns exec haproxy "ip a"`
15:01 wrale_ i was right.. well.. close to right.. it says my VIP is 16-bit
15:01 MiroslavAnashkin or  the same without quotes `ip netns exec haproxy ip a`
15:02 wrale_ thanks for the help.. i'll add this output to my open bug on the topic
15:20 geekinutah joined #fuel
15:24 wrale_ If I'm using ceph for all storage backing possible, do I still need "virtual storage" partitions ?
15:37 mattgriffin joined #fuel
15:38 MiroslavAnashkin wrale_: Yes, you need, until this https://bugs.launchpad.net/fuel/+bug/1262313 and one more related Nova bug will be fixed.
15:40 wrale_ any idea how big they should be ?  odd question i guess.. nodes have 1x240GB ssd for OS and journal and 2x3TB disks originally for OSD.. should I use one of the 3TB disks for virutal storage? or can I get away with ~120GB for virtual storage?
16:17 MiroslavAnashkin It depends on maximum size of your virtual machines. Nova uses this partition to download images temporary. And expands the image to it's full filesystem size
16:33 crandquist joined #fuel
16:43 xarses joined #fuel
17:04 CTWill joined #fuel
17:30 e0ne joined #fuel
17:50 jonathan1 i'm really trying to understand the networking aspects of openstack
17:50 jonathan1 i'm an openstack noob but a sr. linux/unix guy
17:51 jonathan1 it seems like my internal IPs are reachable from the controller node, and when i shoehorn in and create n account i can ssh into the internal IPs, but hwen i map a floating IP i can sort of ping it once i restart the firewall but i get pubkey permission denied.
17:51 jonathan1 it seems as though there is some kind of arp proxy.  i look at iptables and can't find the nat configuration anywhere
17:51 jonathan1 i am probably overthinking it
17:53 jonathan1 of course my IPv6 works like a champ ;)
17:59 TVR_ there are many levels to openstack networking....first... what networking did you choose?
18:01 CTMotorsports joined #fuel
18:06 jonathan1 just nova
18:06 jonathan1 199.188.96.0/24 is the block i am advertising through BGP
18:07 jonathan1 and that's my cloud range
18:07 jonathan1 eventually i want to carve out my entire /22 as openstack instances
18:08 jonathan1 this is part of a project to migrate my customers from proxmox to openstack
18:12 wrale_ may i ask.. do you use H/A with Proxmox?
18:12 wrale_ or do your customers, rather?
18:15 jonathan1 negative :)
18:15 jonathan1 it's more of a side business, tbh
18:15 wrale_ right on.. thanks.. i use both openstack and proxmox.. for different reasons
18:15 jonathan1 i'm trying to ramp up my openstack knowledge to get a day job...
18:16 jonathan1 thoughtwave.com is my company.. i seriuously got my ip blocks in 1994 and then a /22 in 2012 and set up bgp a few years ago
18:16 jonathan1 was at NASA JPL for 9 1/2 years but now i am trying to do my own thing
18:16 jonathan1 :)
18:17 wrale_ move to mountain view .. lol.. jobs galore
18:19 mutex srs
18:28 wrale_ MiroslavAnashkin:  if I want to boot ubuntu cloud from a 2TB volume from ceph, do I need at least 2TB of virtual storage?
18:30 jonathan1 i lived in the bay area in 2003
18:30 jonathan1 until 2003*
18:30 jonathan1 mountain view, palo alto, san francisco.  moved to LA.. now i'm stuck here sharing custody of my kids :)
18:35 MiroslavAnashkin wrale_: Nope. Virtual storage is only used in scope of import/export operations to/from ceph.
18:36 wrale_ So maybe 120GB would be enough, if I plan to use only base ubuntu cloud images
18:37 MiroslavAnashkin jonathan1: Please check security groups, assigned to your tenant, where you run instances with floating IP
18:43 jonathan1 do i have to explicitly open it in the default security group?
18:43 wrale_ yes
18:43 jonathan1 oh
18:43 jonathan1 *foreheadslap*
18:43 jonathan1 maybe i should spend less time watching house of cards while configuring openstack
18:44 wrale_ it would help your job chances..
18:44 wrale_ :)
18:47 jonathan1 wait, so i add a rule opening up port 22
18:47 jonathan1 and i still get thsi
18:48 wrale_ yes
18:48 wrale_ ?
18:48 jonathan1 wolfman:jonathan {107} ssh tw-user@199.188.96.32
18:48 jonathan1 The authenticity of host '199.188.96.32 (199.188.96.32)' can't be established.
18:48 jonathan1 RSA key fingerprint is 29:36:74:06:59:98:26:57:a2:60:17:e4:53:d0:b5:8d.
18:48 jonathan1 Are you sure you want to continue connecting (yes/no)? yes
18:48 jonathan1 Warning: Permanently added '199.188.96.32' (RSA) to the list of known hosts.
18:48 jonathan1 Permission denied (publickey).
18:48 wrale_ it's probably your username
18:48 jonathan1 can't have dashes?
18:48 wrale_ what system image are you booting?
18:48 jonathan1 tw-user is an account i created
18:48 jonathan1 i took the centos 6.5 image and hax0red it to put in my own account called tw-user where i know the password
18:49 jonathan1 also i enabled httpd but
18:49 jonathan1 hmmm
18:49 wrale_ :) clever... dunno why you'd do that
18:49 wrale_ that's anti-cloud
18:49 jonathan1 well i can not get the key pairs to work to save my life
18:49 jonathan1 maybe my keypair is fubar
18:49 wrale_ it's usually the user name
18:50 TVR_ did you check permisions on autorized_keys and such?
18:50 jonathan1 yeah
18:50 jonathan1 permissions all good
18:50 TVR_ 600?
18:50 jonathan1 yes
18:51 TVR_ I get burned with that often enough to commit to memory
18:51 wrale_ are you doing a -i with ssh? do you know when you need to?
18:51 jonathan1 yes
18:51 wrale_ (and need not) :)
18:51 jonathan1 i've been using ssh since 1995 :)
18:51 jonathan1 i just think it may be a firewall issue
18:51 wrale_ just sayin :)
18:51 jonathan1 it seems like it works to internal IP 10.0.0.5 but not my floating one
18:52 jonathan1 but the error to the floating one is an application one about pubkey not one about connection refused which would be indicative of a layer 4 issue
19:12 TVR_ so.. are you using nova flat networking, neutron? (and if neutron, neutron with gre or vlans?)
19:13 TVR_ also.. in HA mode on all in one / single controller?
19:24 jonathan1 flat networking.. nova...
19:25 jonathan1 got a sick kid.. i'll pick this up later :)
19:26 dhblaz joined #fuel
19:26 dhblaz When fuel 5.1 comes out what is the process to upgrade the cluster?
19:27 gleam i believe fuel 5 brings the ability to upgrade the fuel node but not hte cluster underneath
19:27 gleam which.. yeah
19:29 dhblaz Thanks
19:48 angdraug joined #fuel
19:48 mutex gleam: really ?
19:49 mutex I was under the impression it was cluster upgrade
19:56 taj joined #fuel
19:58 gleam https://launchpad.net/fuel/+milestone/6.0
19:58 gleam see #3 on that list
19:58 gleam https://blueprints.launchpad.net/fuel/+spec/upgrade-major-openstack-environment
19:59 mutex Interesting
20:02 IlyaE joined #fuel
20:06 eshumakher joined #fuel
20:10 gleam right.
20:11 gleam upgrading the fuel node lets you then deploy a new openstack environment and migrate your VMs and such over there, but you need new controllers/etc for that
20:11 gleam in-place upgrades aren't a thing yet for it
20:15 monester joined #fuel
20:25 e0ne joined #fuel
20:32 IlyaE joined #fuel
20:32 ogelbukh joined #fuel
20:52 IlyaE joined #fuel
21:14 IlyaE joined #fuel
21:45 IlyaE joined #fuel
21:47 angdraug joined #fuel
22:02 dhblaz gleam: I would be inerested in understanding how you would “migrate your VMs and such” after an upgrade.
22:02 dhblaz Unfortunately I have to run; perhaps if you have time tomorrow you can share more.
22:02 gleam that's a good question and it depends entirely on your infrastructure
22:10 Kupo24z xarses: MiroslavAnashkin do you know where this comes from when changing quotas? 'DEBUG (shell:777) Quota value 10 for cores are less than already used and reserved 13 (HTTP 400)'
22:16 IlyaE joined #fuel
22:22 xarses nope
22:39 IlyaE joined #fuel
22:42 jonathan1 hey it worked, i just had to generate a keypair and not try to use my crappy id_rsa that was borked
22:43 jonathan1 if i have a freebsd zfs backend that served out iscsi targets is it easy to use that for storage instead of ceph/lvm ?
23:18 IlyaE joined #fuel
23:41 christop1eraedo jonathan1: nobody has done that yet AFAIK, nexenta was closest with their plugin. Could be done using the nexenta plugin plus creating an API service on the freebsd box.
23:41 christop1eraedo In theory it wouldn't be that difficult, it doesn't have to do all that much - but I haven't heard of anyone working on that.

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary