17:00:38 #startmeeting openstack security group 17:00:39 Meeting started Thu Dec 18 17:00:38 2014 UTC and is due to finish in 60 minutes. The chair is hyakuhei. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:40 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:43 The meeting name has been set to 'openstack_security_group' 17:00:46 o/ 17:00:46 o/ 17:00:51 o/ 17:00:52 o/ 17:00:57 o/ 17:00:58 o/ 17:00:58 o/ 17:01:08 Good turnout for a pre-holiday meeting :) 17:01:30 :) 17:01:36 well the rumour got out that you were back chairing hyakuhei :P 17:01:46 nice 17:02:04 haha thanks tkelsey 17:02:05 hey all 17:02:09 Hey nkinder nice to see you! 17:02:12 yo nkinder 17:02:22 Yeah, finally a thursday without a conflict! 17:03:09 woo, nice 17:03:16 hi 17:03:23 Ok, what's the agenda for today then ? 17:03:36 allright so I missed a couple of meetings but I'd be curious about: 1) meeting date for OSSG mid-cycle 2) did we get any progress on that calibration exercise? 17:03:51 good points tmcpeak 17:04:07 1) The meetup will not run concurrent with Barbican in the same location 17:04:31 They (Rack) couldn't get the funding for SFO and are holding it in Austin or San Antonio I forget which 17:04:33 There were also some template changes discussed for OSSNs that we should discuss here 17:04:38 Austin 17:04:56 We currently have space reserved at the Geekdom in SF for the _same_ week as Barbican 17:05:04 oh cool 17:05:11 I'm waiting to hear back if we can more that reservation for the week before or week after 17:05:17 although I guess that's no good for some 17:05:23 * bdpayne joins 17:05:57 hyakuhei: so that's the week of Feb 16th currently? 17:06:07 So those who want to attend both can do 17:06:10 nkinder: Yes. 17:07:00 hyakuhei: I can't do the first 2 days of the previous week 17:07:17 flying back from CZ on the 9th 17:07:25 Ok I'm asking for availability for both 17:07:35 but the rest of that week, or the week after are OK with me 17:07:49 I want you to be there if you can, you bring a lot of value 17:07:50 hyakuhei: collect availability on an etherpad? 17:07:51 maybe a similar approach to last time (availability marks on etherpad) could work 17:08:05 tmcpeak: jinx 17:08:09 ;) 17:08:55 ah shucks... you guys! 17:09:06 hyakuhei: does an etherpad already exist? 17:09:28 No sir I've been running around trying to coordinate the dates :( 17:09:38 is geekdom the only option for location? or is it just priced right? 17:09:40 https://etherpad.openstack.org/p/ossg-kilo-meetup 17:09:48 tmcpeak: price is free 17:10:19 hard to beat that 17:11:04 they make it up in volume. 17:11:39 lol cool 17:11:57 So to tmcpeak's second point - OSSA metrics calibration 17:12:22 I'm asking the community to take a look at the metrics I created, take some previous OSSAs and score them 17:12:31 hyakuhei sounds like the location is a lock then? is there still a desire for me to check it out? getting up there is kind of a pain, but I could do it if needed. 17:12:42 And see if the metric that comes out matches the perceived issue 17:12:54 bdpayne: Yes please, I'd really like it if you could check it out 17:13:05 come on bdpayne: cal train is *great* 17:13:07 If we can't move the date we'll have to overlap Barbican 17:13:08 ok, I'll see what i can work out 17:13:18 actually tmcpeak could possibly check it out ? 17:13:27 or are you gallavanting? 17:13:44 gallavanting 17:13:44 perhaps... how close are you to the space tmcpeak? 17:13:53 I can check it out 17:14:05 what should I find out? 17:14:13 I'm about a 20 min drive 17:14:14 ok, I filled in some basics on the etherpad 17:14:17 Size, shape, smell 17:14:22 cool, yeah no problem 17:14:23 just take some pics, get a feel for the size of the space, etc 17:14:26 smell is important, yeah :-) 17:14:30 Basically will it work - you were at the last ossg meetup :) 17:14:30 and coffee options 17:14:31 will post everything to etherpad 17:14:52 +1 coffee :) 17:14:55 does it pass the "would you want to work there for a week" test 17:14:56 I can speak to coffee options - I used to work a block away. Lot's of good coffee and close drinks 17:14:57 stuff like that :-) 17:15:02 excellent 17:15:03 bdpayne: Yup exactly 17:15:03 ok will do 17:15:05 I can take an action 17:15:14 thanks tmcpeak... you just saved me several hours 17:15:20 no worries :) 17:15:24 Superb, thanks guys! 17:15:34 ok, carry on with metrics 17:15:37 #link https://wiki.openstack.org/wiki/Security/OSSA-Metrics 17:15:58 a couple of weeks ago there was talk on standardizing around 5 or so OSSA's 17:16:10 Anyone want to take an action to take an OSSA and run it through the metrics? 17:16:11 and then whoever had bandwidth applying metrics to them 17:16:23 hyakuhei: I think you were away that week 17:16:27 was there any action on that front? 17:16:39 I've not seen it move much 17:16:42 someone was going to send out a list to review 17:16:44 I haven't seen that list 17:16:51 yeah, that's what I was wondering about, that list 17:16:57 but I still think it's a good idea 17:17:19 yeah, I like that idea 17:17:24 yeah a list was talked about, but I guess didnt get out 17:17:34 no worries 17:17:39 who wants to do it now? 17:18:24 i can probably spend a bit of time on this. i'm interested from a vmt perspective and have been working on getting a decent record of ossa data 17:18:34 thanks gmurphy 17:18:48 Excellent - thanks gmurphy 17:18:58 awesome! 17:19:02 I would, but I did one already so I think we need others for a better representative set 17:19:27 Yeah 17:19:28 so are these going to be added to the existing wiki page? 17:19:51 It looks like 2 have been gone through already there 17:19:57 Yeah 17:20:03 I've just been adding them as we go 17:20:11 what's the page? 17:20:27 bknudson: https://wiki.openstack.org/wiki/Designate/Blueprints/IPABackend 17:20:35 err, wrong paste 17:20:39 #link http://ttx.re/the-way-forward.html 17:20:42 https://wiki.openstack.org/wiki/Security/OSSA-Metrics 17:20:42 balls sorry 17:20:45 wrong link too :) 17:20:47 bahahaha 17:20:49 haha, you did it too 17:20:49 hehe 17:20:57 double fail. Time to call it a year 17:21:01 just getting random links 17:21:06 lol 17:21:16 luckily safe for work. 17:21:21 haha 17:21:32 both OpenStack related at least! 17:21:34 paste roulette anybody? 17:21:52 so i'd like to try and merge that data into our ossa repository that we've been working on - http://git.openstack.org/cgit/openstack/ossa 17:21:58 Incidentally, you should all be aware of the changes the TC is going through http://ttx.re/the-way-forward.html 17:21:59 but i can add it to the wiki too 17:22:38 hyakuhei: that brings interesting questions on which projects in this larger upcoming group will be "security-supported" 17:22:55 Exactly 17:22:58 so far we (OSSG and VMT) roughly supported the integrated release 17:23:00 I was surprised when I saw that ceilometer has essentially no security considerations. 17:23:21 hyakuhei: but we could now be more fine-grained 17:23:32 other than, apparently, don't give access to ceilometer to anybody 17:23:33 and support based on audit results 17:23:46 and response times to security issues 17:24:08 * gmurphy goes to read this post.. 17:24:18 It's a long post but worth reading. 17:24:27 Not 100% sure what I think of all the changes tbh 17:25:33 Anyway, OSSG Meetup and Metrics were the only things I wanted to bring up, not looking to add lots of actions before the holidays. 17:25:55 ttx interesting, it will take me a bit to digest this all, but it sounds like the goal is to better scope what we support? 17:25:57 One thing to mention would be Anchor. Does it make sense for it to become more OSSG oriented in the same way that Bandit is 17:26:22 it seems like a good fit 17:26:27 bdpayne: the goal is to describe more accurately what we provide 17:26:28 so we've got a couple of metrics... anybody else look at the 2 that are done and agree / disagree? 17:26:34 hyakuhei perhaps, I'd like to look at it a bit more... it's on my (long) list 17:26:46 bdpayne: be more inclusive on one hand, be more precise on the other 17:26:55 ttx yeah, makes sense 17:27:01 hyakuhei: humm wouldnt hurt for sure, but worth getting input from a few people on it 17:27:01 rather than use a single definitiion to match all stuff 17:29:18 tkelsey: definintely 17:29:21 "Discoverability always assumed to be 10 " -- makes sense since they're all public. 17:30:15 bknudson: yeah 17:32:55 so we've got 2 ossas scored and for some reason the score is different when they seem similar... 17:33:11 e.g., Affected Users is 4 for one and 6 for the other 17:33:24 but they both say it's same users affected (nova users) 17:34:02 perhaps some nova users are more important than others? ;-) 17:34:50 bdpayne: the second is nova users on the node. The first is nova API in general 17:35:01 ahh 17:35:06 well that kind of makes sense 17:35:20 personally I think that "OSSA 2013-012" was worse than "OSSA 2014-038". 17:35:41 but "Nova fails to verify image virtual size" is probably harder to exploit. 17:36:04 bknudson: worth commenting on the discussion probably :) 17:36:05 since you need to get the bad image in the system first. 17:36:43 This is exactly the conversation we need to have when we have a few more to calibrate against :) 17:37:03 hyakuhei: +1 17:39:25 yeppers 17:39:35 are we supposed to update the discussion part in the wiki with comments? I don't think wiki is the best place for a discussion. 17:39:56 We'll discuss on irc or a google hangout or something once we've got 5-6 17:40:10 bknudson: humm your probably right, hyakuhei +1 17:40:11 I'd prefer to have discussion in one of these meetings or in the security room 17:41:36 are we just picking random OSSAs? 17:41:56 or was there a list? 17:41:59 I think that previous idea to standardize on 5 or so for all of us to review might be more effective 17:42:33 bknudson: jut pick any you have a good feel for 17:42:44 once we have a few we'll address any obvious gaps 17:44:48 ok. well i'll try to pick a few more and add them to the wiki. 17:44:54 Thanks guys 17:45:04 I'll work on this one that I reported https://bugs.launchpad.net/ossa/+bug/1354208 17:45:06 Launchpad bug 1354208 in ossa "[OSSA 2014-029] Catalog replacement allows reading config (CVE-2014-3621)" [Medium,Fix released] 17:45:19 probably will go for a few different flaw classes etc. 17:45:30 I assume we aren't having a meeting btw? 17:45:34 next week I mean 17:45:45 that's a safe assumption 17:45:51 (just trying to figure out by when I should recon geekdom) 17:47:14 Yeah, this will be the last pre-xmas meeting 17:47:23 week after that is new years day 17:47:34 yeah, I vote for 2 weeks off 17:47:41 +1 17:47:45 +1 17:47:58 ok :) 17:48:14 we can check out geekdom together the week after new years 17:49:11 anything else for today? 17:49:18 nothing from me 17:49:44 nope 17:49:45 nope. 17:49:47 i just wanted to throw out the link for the pad i'm working on with ideas for the sahara sec doc 17:50:00 elmiko: ah cool 17:50:02 #link https://etherpad.openstack.org/p/sahara-security-guide-notes 17:50:09 elmiko: awesome 17:50:19 it's a little sparse currently, but i'm hoping to have something ready for review early in january 17:50:35 elmiko: sounds good 17:50:40 i'm having a little difficulty in determining/developing what is the default position of the project with regards to security 17:51:12 but i'm using the OSSG guide and a little common sense to develop a starting opinion, then i figure we can hash it out from there 17:51:38 i welcome any comments, questions, criticism, in the pad. it really helps out =) 17:52:19 that's all i had 17:53:07 Thanks elmiko we can always help with specific questions 17:53:48 awesome, i'll try to work towards something a little more concrete for the SG 17:54:16 :D 17:54:26 Anything else to discuss today peoples? 17:56:06 nothing here 17:56:14 nope 17:56:17 Happy Holidays all 17:56:19 nope 17:56:25 nkinder: and you :) 17:56:33 happy holidays! have a good one all 17:56:44 have fun all =) 17:56:58 o/ 17:59:49 Thanks all, happy holidays! 17:59:55 #endmeeting