17:01:57 <hellais> #startmeeting
17:01:57 <MeetBot> Meeting started Mon Nov 23 17:01:57 2015 UTC.  The chair is hellais. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:01:57 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
17:02:07 <hellais> greetings
17:02:20 <hellais> 'ya all know the drill, who is here?
17:02:24 <landers> i am
17:02:31 <anadahz> me too
17:02:36 <M-VE> likewise
17:02:40 <poly> hello all
17:02:46 <hellais> purrfect
17:03:03 <hellais> so what have you been up to last week or how long back in the past you wish to go
17:04:38 <poly> alright, I'll go
17:05:15 <poly> after dealing with a series of OS-related crashes, I've finished working on javascript based regexes for network meter
17:05:50 <poly> this solution, thought of by sbs and myself, has allowed us to get rid of lots of awkward escaping characters in the program
17:06:09 <poly> my understanding in that sbs will review the PR in the next few days for merging
17:06:12 <poly> EOF
17:06:36 <landers> i've been working on the ooni-sysadmin and ooni-api things. running one of the ooni-pipeline tasks to move some archival data nearer towards the hadoop tasks and eventually the public-facing api/db
17:06:58 <landers> got a few tasks related to that this week EOF
17:07:49 <M-VE> we've been working on a network of probes and our own backend to monitor censorship during the upcoming venezuelan elections. to limited success. EOF
17:10:01 <poly> M-VE. What's causing the troubles? Anything we can do to help?
17:10:11 <hellais> I have been helping joe get started into doing sysadmin and pipeline tasks. I also have been doing a lot of let's say "Project management" related activies as well as figuring out funding and proposal related matters.
17:11:22 <landers> (i'm joe)
17:11:46 <hellais> EOF
17:11:55 <anadahz> continue working for the OONI hardware infrastructure. Worked on the completion
17:11:57 <anadahz> of a contract deliverable, reviewing the PRs for getting some PTs tests implemented
17:12:10 <anadahz> EOF
17:12:42 <hellais> great!
17:12:58 <hellais> I would also be curious to learn more about what we can do to help M-VE with their deployment.
17:13:55 <M-VE> I can explain if it's ok
17:14:49 <hellais> M-VE: please do :)
17:14:50 <M-VE> we got a ooni backend to run. tried following the instructions but they didn't work, so we did a few things to make it run
17:15:01 <M-VE> fix commands in the instructions, forward additional ports and change ports in the config
17:15:53 <M-VE> it runs, but it shows this when we run a test deck on a probe
17:16:07 <M-VE> 2015-11-13 17:07:44+0000 Matching <_sre.SRE_Pattern object at 0x2747270> with dns_consistency | test_name
17:16:08 <M-VE> 2015-11-13 17:07:44+0000 invalid-input-hash
17:16:21 <M-VE> 2015-11-13 17:07:44+0000 Matching <_sre.SRE_Pattern object at 0x2747270> with http_requests | test_name
17:16:22 <M-VE> 2015-11-13 17:07:44+0000 invalid-input-hash
17:16:28 <hellais> M-VE: ah, that is because you need to configure your policy
17:16:48 <M-VE> it works with some tests, and not with others
17:16:55 <M-VE> how would I go about that?
17:17:44 <hellais> https://github.com/TheTorProject/ooni-backend/blob/master/data/policy.yaml
17:17:48 <hellais> you need to edit this file
17:18:16 <hellais> (the one specified in your conf file)
17:19:15 <anadahz> s
17:19:22 <M-VE> interesting, will try that
17:19:26 <M-VE> also
17:19:39 <M-VE> could you provide some information about annotations?
17:19:44 <M-VE> on ooni-probe
17:19:48 <hellais> you can also remove a policy alltogether
17:20:01 <hellais> by setting it to none in the config file
17:20:10 <hellais> if you could add this to the docs that would be epic too
17:20:51 <hellais> anadahz: have you read the emails from greenhost support?
17:20:52 <M-VE> yeah, I have a couple of pointers suggestions for the documentation
17:21:18 <hellais> they disconnected our machine and created another one and the data was not migrated
17:21:31 <anadahz> hellais:just read them
17:21:32 <hellais> I think we should drop using greenhost altogether and move either to the greek VPS or to AWS
17:21:53 <hellais> it seems like they are not a very reliable host
17:22:06 <M-VE> we had some issues
17:22:14 <hellais> we had running on that the bridge reachability collector
17:22:15 <anadahz> hellais: yes it makes more sense to move to a different host
17:22:16 <M-VE> a HK machine went down for hours
17:22:32 <hellais> M-VE: with Greenhost infrastructure?
17:22:38 <M-VE> yep
17:22:56 <hellais> yeah we also have not been particularly pleased with it's performance
17:23:16 <hellais> luckily grnet is sponsoring us some vps' there and we got a AWS research grant to cover ec2 costs
17:23:28 <hellais> so far we haven't had any issues with these two providers
17:25:01 <anadahz> hellais: I could deploy something on a dedicated server, I need an OS though.. if anyone has an idea of what then I could deploy this in some hours (today).
17:25:53 <landers> anadahz: it seems to be a bunch of docker containers and some data that needs to be rsync'd
17:26:13 <anadahz> landers: yes but still we need an OS for that!
17:26:34 <landers> i think some reconnaisance is needed
17:26:53 <anadahz> landers: reconnaisance?
17:27:04 <landers> intel gathering
17:27:10 <landers> not ready for today, i mean
17:27:44 <anadahz> hm..
17:27:57 <hellais> anadahz: we are on it with joe, don't worry about that for the moment
17:28:07 <hellais> it would be good if you could focus on the raspberry pi and ifcci stuff
17:28:10 <anadahz> landers: hellais nice ;)
17:28:23 <hellais> we also have a problem with brnvk
17:28:35 <anadahz> :(
17:28:37 <hellais> we discussed today that given he cannot dedicate some time to the ooni explorer project we should find another hire
17:29:07 <hellais> I believe the best way to proceed is to make some chunks of the task available to the community
17:29:17 <hellais> people can begin hacking on them and if they do well we can hire them
17:29:43 <hellais> brennan is going to help out with that and has reached out to a community of designers and data viz people
17:29:55 <hellais> https://github.com/opensourcedesign
17:33:43 <anadahz> M-VE: would you like to work on the OONI explorer?
17:35:06 <M-VE> sounds interesting, but we'd need a raincheck. The elections are in less than 2 weeks and we don't have a baseline for the election coverage
17:36:06 <M-VE> I have a few additional questions if you don't mind: - Can you provide an example of usage of annotations on ooni-probe? I haven't worked on this personally but they've had issues
17:37:49 <anadahz> M-VE: I have never used annotations but i'm not sure if/or in which tests are working
17:38:20 <M-VE> is there any other way to send custom data?
17:38:36 <anadahz> M-VE: have you tried to use the annotations parameter?
17:38:51 <anadahz> M-VE: on any test?
17:38:52 <M-VE> that's what they've been using, I believe
17:38:56 <hellais> -a, --annotations=  Annotate the report with a key:value[, key:value] format.
17:39:03 <M-VE> yes, that
17:39:38 <hellais> what question do you have about them? They have not been used in quite some time
17:39:49 <hellais> or I haven't at least
17:39:50 <M-VE> but it doesn't work for them. sorry I can't give more details than that... that's why I'm asking for a full example
17:40:04 <M-VE> we want to send several pieces of information
17:40:05 <M-VE> city
17:40:08 <M-VE> isp
17:40:13 <M-VE> among others
17:40:22 <hellais> I will look into this
17:40:38 <hellais> I just ran it and it seems to not work
17:40:45 <hellais> can you file an issue on github?
17:40:59 <M-VE> ok
17:41:00 <hellais> https://github.com/TheTorProject/ooni-probe/issues/new
17:41:01 <M-VE> also
17:41:13 <hellais> In general also for questions you should open tickets on ooni-probe or ooni-backend
17:41:21 <M-VE> good to know
17:41:36 <M-VE> then the next question will be there then
17:41:47 <M-VE> (about the config changes I had to use to get the backend running)
17:41:58 <hellais> that is the best way to avoid you having to wait on us being on IRC and make sure it get's taken care of
17:42:09 <hellais> also about the policy it would be good to file an issue
17:42:10 <M-VE> awesome. thanks for all the help
17:42:50 <M-VE> will do
17:42:56 <hellais> are there more topics for discussion?
17:46:43 <hellais> so I guess next steps?
17:47:51 <hellais> I will be working with joe and vasilis on moving the explorer and ifcci projects forward
17:48:11 <hellais> mainly my tasks will be related to knowledge transfer and doing the plumbing in the API to get it towards production ready
17:49:01 <hellais> EOF
17:51:02 <landers> migrate off greenhost, move things through the pipeline, get the api up.
17:51:09 <landers> got these all scheduled for this week
17:51:12 <landers> EOF
17:56:00 <anadahz> completing the contract work, helping joe and arturo
17:56:22 <anadahz> polishing lepidopter documentation
17:56:29 <anadahz> EOF
17:57:24 <anadahz> M-VE: add me in the loop WRT to your deployments
17:58:23 <M-VE> you mean now or as it progresses?
17:59:02 <anadahz> M-VE: both
17:59:35 <anadahz> M-VE: btw did you have any issues with the RasPis and lepidopter?
17:59:42 <M-VE> we have probes in a few cities, rpis. they're not collecting anything yet, but we have ssh access to them
18:00:33 <M-VE> anadahz: we couldn't get lepidopter to run, not even with the image provided. error I can give you later. we used minibian
18:00:41 <M-VE> errors*
18:01:45 <hellais> all this should go into issues on github
18:01:50 <anadahz> M-VE: yes please do so!
18:01:58 <M-VE> so, probes already in place but not gathering data... and collector running but not receiving
18:02:05 <hellais> unless there are more people wanting to report back I am going to consider this meeting done 2 minutes
18:02:08 <M-VE> yeah, agreed
18:02:16 <hellais> don't let this stop you from stopping to talk though
18:02:18 <M-VE> (on the issues(
18:02:19 <M-VE> )
18:02:20 <hellais> it will just not be logged
18:02:41 <hellais> #endmeeting