Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-02-11

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

All times shown according to UTC.

Time Nick Message
00:40 richardkiene joined #fuel
00:57 rongze joined #fuel
01:33 IlyaE joined #fuel
01:50 rongze joined #fuel
02:13 phamby joined #fuel
02:17 xarses joined #fuel
02:27 e0ne joined #fuel
02:47 phamby joined #fuel
03:06 ilbot3 joined #fuel
03:06 Topic for #fuel is now Fuel for Openstack: http://fuel.mirantis.com/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
03:54 richardkiene_ joined #fuel
04:32 IlyaE joined #fuel
05:12 kpimenova__ joined #fuel
05:18 IlyaE joined #fuel
05:27 IlyaE joined #fuel
05:30 rongze joined #fuel
06:05 IlyaE joined #fuel
06:15 rongze joined #fuel
06:38 saju_m joined #fuel
06:50 saju_m joined #fuel
07:27 mutex I'm having a condition where my floating IP instances will routinely not return packets for about 14 seconds
07:27 mutex and I am assuming that is not normal
07:28 mutex and I do see iptables-save and iptables-restore running on the current l3 node
07:34 e0ne joined #fuel
07:37 saju_m joined #fuel
07:44 rvyalov joined #fuel
08:27 e0ne joined #fuel
08:35 mrasskazov joined #fuel
08:49 e0ne_ joined #fuel
08:51 saju_m joined #fuel
08:53 miguitas joined #fuel
09:25 AndreyDanin_ joined #fuel
09:38 anotchenko joined #fuel
09:53 anotchenko joined #fuel
09:54 mihgen joined #fuel
10:26 saju_m joined #fuel
10:31 anotchenko joined #fuel
10:39 tatyana joined #fuel
10:58 miguitas joined #fuel
11:38 vk joined #fuel
11:40 mihgen joined #fuel
11:48 anotchenko joined #fuel
11:50 e0ne joined #fuel
11:59 e0ne_ joined #fuel
12:03 vk joined #fuel
12:48 aglarendil joined #fuel
12:49 TVR__ joined #fuel
12:49 Dr_Drache joined #fuel
12:50 Arminder joined #fuel
12:56 MiroslavAnashkin mutex: iptables* commands does not matter on L3 agent node. L3 agent works in its namespace and provides routes via qrouter instances, running in different namespaces.
12:59 Arminder joined #fuel
13:18 Arminder joined #fuel
13:19 TVR__ hey MiroslavAnashkin, what is the rough timeframe for the fuel install node to have a redundancy option? I will not hold you to the deadline, but I am writing up the work I did on the POC for using fuel, and managers will need to see some timeline for that, so I say ~ June, then it give an idea, rather than simply saying it's on the roadmap.
13:28 MiroslavAnashkin TVR__: Hey, best estimate - not earlier than late April.
13:30 TVR__ thanks
13:58 Dr_Drache TVR__, MiroslavAnashkin, morning guys
13:58 TVR__ morning sir..
14:00 Dr_Drache and evening where approprate.
14:01 Dr_Drache blah, TVR__ what was it that allowed DHCP for external net on instances?
14:01 TVR__ set to shared
14:02 TVR__ by default, it is not
14:02 TVR__ that worked for my tests
14:03 Dr_Drache ..i just noticed something weird.
14:03 Dr_Drache had a windows server test image.
14:03 Dr_Drache used it for a day, setup stuff, etc.
14:03 Dr_Drache just went to the console..
14:03 Dr_Drache and it's "new"
14:06 Dr_Drache I need to make sure I get persistance working :P
14:11 Dr_Drache still no DHCP.
14:12 Dr_Drache TVR__, it's a stupid thing
14:12 Dr_Drache edit the subnet "enable DHCP"
14:13 TVR__ have you sniffed the subnet to see if the DHCP request is getting to your server?
14:13 TVR__ so it's working now?
14:14 Dr_Drache yea
14:14 TVR__ cool..
14:14 Dr_Drache dhcp wasn't "turned on"
14:14 TVR__ that will do it
14:14 Dr_Drache what a stupid mistake. of course, it being 4-5 windows deep doesn't help.
14:16 Dr_Drache cool, so... openstack from the words of the code monkeys "faster than the same machine on KVM/NFS"
14:17 TVR__ Yea.. I have to write up a report on what I have done / found... I need to do some benchmarks as well....
14:18 Dr_Drache i think it's all from ceph/novnc
14:18 TVR__ I just looked it up.. Canada is kicking some at the Olympics.....
14:18 Dr_Drache question oh great east coaster.
14:18 Dr_Drache :P
14:19 Dr_Drache this should be openstack 99, but... is there a way to make sure an instance is protected?
14:20 TVR__ protected?
14:20 TVR__ you can lock it down with security groups
14:20 TVR__ what are you looking to do?
14:24 Dr_Drache if I understand correctly, you terminate a instance, it's gone.
14:26 Dr_Drache I guess I just need to find something to read about being protected/volumes/images
14:26 Dr_Drache looks like I know less about running it, then I thought I did.
14:26 getup- joined #fuel
14:27 Dr_Drache i'm well versed in KVM vm management. but not so on how that relates to OS workflows.
14:28 TVR__ if an instance goes away... you can always start another from the volume that you created
14:28 Dr_Drache what about data persistance?
14:29 saju_m joined #fuel
14:31 anotchenko joined #fuel
14:32 TVR__ the volume would be at the state that the machine terminated from..
14:32 TVR__ I still have yet to test live migration
14:33 Dr_Drache I see what I did wrong.
14:34 Dr_Drache I booted from my "image'
14:34 Dr_Drache but, i didn't "Create new volume"
14:34 Dr_Drache from booting from the image.
14:34 TVR__ ah.. ok
14:34 Dr_Drache ok
14:35 Dr_Drache one more stupid question.
14:35 TVR__ one more stupid guess at the answer then
14:35 Dr_Drache when I launch an instance, choose a flavor.
14:36 Dr_Drache it has a drive size (40gb), but then, instance boot source, is "boot from image (creates a new volume)"
14:36 rongze joined #fuel
14:36 Dr_Drache and asked for a "Device size (GB)"
14:37 Dr_Drache does that override the flavor size?
14:37 TVR__ it seems to for me.. as I may choose 80G and I get an 80G volume
14:39 Dr_Drache thanks
14:39 Dr_Drache I think I'm done with being stupid today.
14:44 Dr_Drache well, that doesn't work
14:56 phamby joined #fuel
14:59 phamby joined #fuel
15:17 ruhe_ joined #fuel
15:21 Dr_Drache MiroslavAnashkin,
15:21 Dr_Drache I have 2 things I'd like to report today.
15:21 Dr_Drache let me know if a bug report is needed.
15:22 MiroslavAnashkin Dr_Drache: ?
15:22 Dr_Drache 1. when creating an instance as I outlined with TVR__, the timeout is shorted than the time it takes to download the image, so the instance errors out on me.
15:22 Dr_Drache 2. volumes created in this way, are nameless.
15:23 MiroslavAnashkin 1. Yes, it is already existing OpenStack bug.
15:24 MiroslavAnashkin 2. What does it mean nameless? Created volumes has their volime ID
15:24 MiroslavAnashkin volume ID
15:25 TVR__ I can understand the reasoning for them being named the way they are... every users project will want their volume to be called "80G" or some variant, so the naming is done the way it is to prevent naming conflicts... I am not sure what a good improvement would be except maybe to pre-make them?
15:25 TVR__ the naming based on Volume ID makes sense for this reason
15:26 Dr_Drache yes, but having a ID such as that, how do you keep track of them? there is a space in the meta for a common name.
15:26 Dr_Drache http://imgur.com/flHAlRX
15:26 Dr_Drache and I guess these are OS bugs, not fuel
15:28 mrasskazov joined #fuel
15:29 phamby joined #fuel
15:29 Dr_Drache but having a array of things called 80GB or an array of things with random, non-crossable IDs, in the end, you'll still lose track of what is what.
15:30 Dr_Drache just when you create a volume manually, you can assign a common name.
15:30 Dr_Drache I guess that's my only real point.
15:31 TVR__ like when you create multiple instances at once.. can't you rename it after?
15:31 Dr_Drache I'm looking
15:31 Dr_Drache but, so far, not from the gui.
15:32 Dr_Drache you can see where they are attached, one at a time.
15:32 MiroslavAnashkin Dr_Drache: It is possible to set volume name via CLI
15:33 Dr_Drache CLI is the devil!
15:33 Dr_Drache lol
15:35 MiroslavAnashkin Simply run command like following: cinder create --display_name <NAME> <SIZE>
15:36 Dr_Drache but why?
15:36 Dr_Drache I can create and name in the GUI
15:37 Dr_Drache but, if the volume is created VIA an instance creation, you cannot
15:37 Dr_Drache I'll go file a bug on that
15:39 phamby joined #fuel
15:46 saju_m joined #fuel
15:53 phamby joined #fuel
15:54 vk joined #fuel
15:58 phamby joined #fuel
16:08 mihgen joined #fuel
16:12 phamby joined #fuel
16:14 phamby joined #fuel
16:17 phamby joined #fuel
16:21 mihgen joined #fuel
16:44 vk joined #fuel
16:46 xarses joined #fuel
16:49 phamby joined #fuel
16:50 richardkiene joined #fuel
16:53 anotchenko joined #fuel
17:01 IlyaE joined #fuel
17:11 angdraug joined #fuel
17:33 rmoe joined #fuel
17:41 anotchenko joined #fuel
18:04 rmoe joined #fuel
18:23 vk joined #fuel
18:49 e0ne joined #fuel
19:10 rmoe joined #fuel
19:16 richardkiene joined #fuel
19:41 ruhe_ joined #fuel
19:43 rmoe_ joined #fuel
19:51 saju_m joined #fuel
19:52 IlyaE joined #fuel
19:55 ruhe- joined #fuel
19:58 ruhe joined #fuel
20:04 TVR___ joined #fuel
20:05 saju_m joined #fuel
20:19 MiroslavAnashkin joined #fuel
20:50 richardkiene joined #fuel
21:26 mutex MiroslavAnashkin: ok, so then how can I track down which neutron process is crashing and causing IPtables to be reloaded ?
21:50 albionandrew joined #fuel
22:18 e0ne joined #fuel
22:25 xarses joined #fuel
22:45 mutex I am considering changing the password for some of the ones that are not autogenerated by fuel
22:45 mutex or
22:45 mutex like
22:45 mutex murano:swordfish
22:45 mutex heat:heat
22:46 mutex any idea where they are set in puppet ?
22:48 xarses joined #fuel
23:03 richardkiene_ joined #fuel
23:04 jhurlbert joined #fuel
23:05 jhurlbert when we try adding a new compute node to OpenStack with Fuel, we get this error: (/Stage[main]/Ceph::Conf/Exec[ceph-deploy config pull]/returns) change from notrun to 0 failed: ceph-deploy --overwrite-conf config pull node-3 returned 1 instead of one of [0] at /etc/puppet/modules/ceph/manifests/conf.pp:44
23:06 jhurlbert I've been able to add 2 other compute nodes after this one failed, but this particular node continues to fail.
23:46 jhurlbert joined #fuel
23:47 mutex ok so looks like I need to update astute.yaml
23:48 mutex just not sure how this is generated from the fuel db and placed on the node
23:48 mutex ah, mcollective
23:59 vk joined #fuel

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