Time
S
Nick
Message
03:46 jri joined #dataverse
04:01 jeffspies______ joined #dataverse
05:55 dataverse-user joined #dataverse
06:31 knikolla joined #dataverse
06:31 karissa joined #dataverse
07:25 knikolla joined #dataverse
07:25 karissa joined #dataverse
08:20 jri joined #dataverse
08:57 jri_ joined #dataverse
10:10 jri joined #dataverse
10:50 jri joined #dataverse
12:18 jri joined #dataverse
13:52 donsizemore joined #dataverse
15:05 pdurbin
donsizemore: morning. The new `vagrant up` is gonna "just work", right?
15:09 * pdurbin
kicks it off
15:24 yoh joined #dataverse
15:32 donsizemore
@pdurbin mine did for me, and @pameyer's did once he tweaked it =)
15:32 donsizemore
@pdurbin or rather, mine did for @pameyer once he tweaked it
15:32 pdurbin
it's weird, `curl http://localhost:8888/api/info/version ` is showing `curl: (52) Empty reply from server`
15:33 donsizemore
@pdurbin oh, i always did a 'vagrant ssh' to poke around
15:33 donsizemore
@pdurbin lemme do a pull on that branch and try the updated Vagrantfile here
15:34 pdurbin
`curl localhost:8080/api/info/version` works fine if I ssh in first
15:36 yoh joined #dataverse
15:39 pdurbin
donsizemore: could it be a firewall? I'm such a noob with el7... the commands seem to have changed. I'm used to `service iptables status`.
15:40 pdurbin
I'm looking at this: HOWTO: Use iptables with CentOS 7 https://community.rackspace.com/products/f/25/t/4504
15:41 donsizemore
@pdurbin i just dropped @pameyer's fixen in place, running on my end
15:41 donsizemore
@pdurbin since we're switching base boxes we'll need to see what is and isn't being launched by systemd
15:42 pdurbin
makes sense
15:42 donsizemore
@pdurbin since it's a vagrant box i'd disable firewalld and any iptables config by default
15:42 pdurbin
sure, it's just for testing
15:43 donsizemore
@pdurbin on centos7 by default you'd say "$ sudo firewall-cmd --list-all"
15:43 pdurbin
sudo? Am I still allowed to just `su -` to root? :)
15:44 donsizemore
yis
15:44 pdurbin
I ran the sudo version. "ports" is blank. The output looks like this: "ports: "
15:45 donsizemore
@pdurbin yeah, it's doing the centos default dhcpv6-client ssh thing
15:46 pdurbin
"services: dhcpv6-client ssh"
15:46 donsizemore
@pdurbin let me pop in some command dataverse ports and i'll send in an updated copy
15:46 donsizemore
my bad, i never test vagrant from the host machine
15:46 pdurbin
Ok. Thank you!
15:47 pdurbin
Meanwhile, I'm tweaking the Installation Guide.
15:50 pameyer joined #dataverse
15:51 pameyer
pdurbin, donsizemore: I was seeing the same thing with vagrant ports not being accessible from the host system
15:51 pameyer
tried turning off firewalld; no luck
15:52 pameyer
things worked inside the guest though
15:55 jri joined #dataverse
15:57 donsizemore
@pameyer i can poke at that. i'm hoping for a quiet afternoon =)
15:57 pameyer
donsizemore: those are nice (when they happen) ;)
15:58 jri joined #dataverse
16:03 donsizemore
@pameyer i'm in the blissful state of "waiting on someone else" for most everything i've got going on... and those someones are mostly out until january
16:03 jri_ joined #dataverse
16:04 donsizemore
so @pameyer disabling and stopping firewalld made dataverse accessible for me from the vagrant host machine. (Ubuntu 16.04.3, vagrant 1.8.1, virtualbox 5.0.40-dfsg-0ubuntu1.16.04.2)
16:06 donsizemore
@pameyer http://i.imgur.com/o46y9d4.gif
16:06 donsizemore
@pameyer which host OS are you using?
16:06 pameyer
donsizemore: I probably only did one of those
16:06 pameyer
OS X host; hasn't usually been a problem
16:10 pdurbin
I'm looking at http://bytefreaks.net/gnulinux/how-to-startstop-or-enabledisable-firewalld-on-centos-7
16:13 pdurbin
If I run `sudo systemctl status firewalld` I'm able to get at http://localhost:8888/api/info/version from my Mac. So we just need to run that inside Vagrant?
16:16 pdurbin
donsizemore: ah, I see you fixed it. You pushed a commit. Thanks! Next problem: `vagrant destroy` doesn't "just work" anymore. I get "A box must be specified." I'm not sure why
16:17 pdurbin
`vagrant halt` works
16:18 pdurbin
well, it's running at least
16:19 donsizemore
@pdurbin poop. I thought vagrant was just ticked at me because I updated the Vagrantfile while I had accidentally left a VM running
16:22 jri joined #dataverse
16:25 pdurbin
I just pushed a fix: https://github.com/IQSS/dataverse/pull/4385/commits/8232cbbe8fc323a1fa0d134c437ebd9c9baa9abd
16:26 pdurbin
whoops, but now `vagrant up` is broken. sorry
16:28 pdurbin
ok, fixed it. sorry
16:33 pdurbin
donsizemore: your firewall fix worked for me. Thanks!
16:33 jri joined #dataverse
16:34 pdurbin
It would be nice if `vagrant destroy` just worked but at least I documented a workaround in the Vagrantfile.
16:38 pameyer
hmm… I wonder if I broke that with my changes yesterday
16:38 pameyer
ah - looks like maybe not
16:40 pdurbin
I bet I'm not the only one who likes to run `vagrant destroy` after spinning up a Vagrant environment. It would be nice to fix this. I'm not sure how.
16:42 pameyer
I'm with you - but `vagrant destroy` does mean that I'll be due for another wait when I need to spin it back up
16:42 pameyer
I'll take a look
16:43 pdurbin
Sure, I just mean that I'll do destroy when I'm really done for a whle.
16:57 pameyer
`vagrant destroy` works as expected for me ac496b14be3de3cf26533810fc81cded12465717
16:58 pameyer
vagrant 1.7.4
17:01 pameyer
so I think somebody fixed it
17:03 jri joined #dataverse
17:32 pdurbin
Maybe, but I'm on a newer version: Vagrant 1.8.7
17:40 pameyer
how did it not work for you?
17:43 pdurbin
It fails with "A box must be specified."
17:46 pameyer
huh
17:46 pdurbin
it's a head scratcher
17:46 pameyer
I just got the usual confirm destroy (which `destroy -f` should get rid of)
17:50 pdurbin
pameyer: try upgrading your Vagrant :)
17:50 pameyer
let me break something right before vacation? :0
17:51 donsizemore
@pdurbin @pameyer I promise I'll look at the broken vagrant destroy (but I also had a lunch date) - back now
17:53 pdurbin
pameyer: I'm kidding. Don't change anything right before winter break. I linked to this tweet in Slack: https://twitter.com/chrisalbon/status/943342608742604801
17:53 pdurbin
donsizemore: even if we don't fix it, I'm fine with merging it. I'm close to being done updating the guides.
17:54 pdurbin
Huh, nothing about EPEL in the pull request yet.
17:58 donsizemore
@pdurbin don't merge it until i can look into the destroy dealy? (and yes, consider the rest of december "read-only friday!"
17:59 pdurbin
donsizemore: I'm sure this won't get merged until 2018. No rush.
18:04 pdurbin
donsizemore: ok, I just pushed my doc changes. Please pull the latest.
18:04 donsizemore
@pdurbin done!
18:24 pdurbin
donsizemore: you install R from EPEL on el7, right? I assume it works fine.
18:28 donsizemore
@pdurbin yes, and i've blindly updated each time they bump the version =)
18:28 pdurbin
good, good
18:29 pdurbin
donsizemore: what version of R are you running?
18:38 donsizemore
@pdurbin R-3.4.2-2.el7.x86_64 in production and elsewhere
18:38 donsizemore
@pdurbin p.s. so far updating vagrant to 2.0.1 and manually removing borked VMs in virtualbox haven't squared away the destroy issue. i was hoping it just had its eyes crossed
18:39 pdurbin
donsizemore: ok, same R version I just got in Vagrant. Thanks.
18:41 pdurbin
donsizemore: yeah, me too. I was also looking at some global thing: vagrant global-status
18:44 donsizemore
i removed the VMs from VirtualBox, then did a "vagrant global-status --prune" which made the entry go away, but a subsequent "vagrant up" put me back in the same state. i forgot to rm -rf .vagrant tho
18:47 pdurbin
I guess if someone notices we ask them to open an issue. :) Again, my workaround is in the Vagrantfile. Uncomment a line, run `vagrant destroy`. Done. :)
18:48 donsizemore
@pdurbin so you're having it default to debian?
19:12 pdurbin
donsizemore: no...
19:12 pdurbin
Did I mess something up?
19:19 pdurbin
donsizemore: oh! Whoops! I grabbed the wrong line, it looks like.
19:24 pdurbin
donsizemore: pushed. good catch. thanks
19:36 donsizemore
@pdurbin just checking if that was intentional - sorry, was outside delivering a cheesecake!
19:40 * pdurbin
wants cheesecake
20:10 pdurbin
Whelp, I'm taking off in an hour or so. Thank you to everyone for all of your participation all year and I'll see you in 2018.
20:11 pdurbin
andrewSC bjonnh bret dzho jeffspies______ jhand JonathanNeal karissa knikolla larsks mdunlap pameyer sivoais yoh ^^
20:11 pdurbin
The official happy holidays note: https://groups.google.com/d/msg/dataverse-community/RF0koZzQGFs/Wue7AqiQCgAJ
20:36 knikolla
pdurbin: happy holidays!
20:37 pdurbin
thanks, you too
21:00 donsizemore joined #dataverse
21:27 pameyer
happy holidays everyone
21:29 pdurbin
+1
21:29 pdurbin left #dataverse
21:33 jri joined #dataverse
22:16 dataverse-user joined #dataverse
