<paulproteus>
I am installing Vagrant+VirtualBox+vagrant-spk+git to test things on Windows.
<ocdtrekkie>
I was figuring the existing note was enough now that this is a niche sidebar article as opposed to the main dev article, which it was originally.
<ocdtrekkie>
New users will hopefully have read the vagrant-spk tutorial, and understand that Sandstorm runs on Linux, somewhere long before they get to this article. ;)
<paulproteus>
I mostly agree except for deep links to this page.
<ocdtrekkie>
That is probably a good change though in #882.
<paulproteus>
Not as an alternative to what we have now, but as a way to generate reasonably complete and easy-for-devs-to-write reference documentation.
gopar has quit [Remote host closed the connection]
mquandalle has quit [Quit: Connection closed for inactivity]
larjona has joined #sandstorm
mnutt_ has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 260 seconds]
mnutt_ has quit [Quit: mnutt_]
mort___ has joined #sandstorm
mort___ has quit [Client Quit]
mort___ has joined #sandstorm
Aric has joined #sandstorm
<Aric>
Is there a way to do httpd side optimizations on a WP grain? (gzip, browser cache, etc.)
mort___ has left #sandstorm [#sandstorm]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 250 seconds]
decipherstatic has quit [*.net *.split]
decipherstatic has joined #sandstorm
mnutt_ has joined #sandstorm
mnutt_ has quit [Quit: mnutt_]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 256 seconds]
mort___ has joined #sandstorm
xet7_ has quit [Ping timeout: 252 seconds]
Aric has quit [Quit: Connection closed for inactivity]
mort___ has left #sandstorm [#sandstorm]
xet7_ has joined #sandstorm
mort___ has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 250 seconds]
mort___ has quit [Quit: Leaving.]
mort___ has joined #sandstorm
mnutt_ has joined #sandstorm
mnutt_ has quit [Quit: mnutt_]
larjona has quit [Quit: Konversation terminated!]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 260 seconds]
mort___ has quit [Ping timeout: 246 seconds]
larjona has joined #sandstorm
mort___ has joined #sandstorm
bb010g has quit [Quit: Connection closed for inactivity]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 250 seconds]
mort___ has quit [Quit: Leaving.]
jadewang has joined #sandstorm
simonv3 has joined #sandstorm
amyers has joined #sandstorm
amyers has quit [Read error: Connection reset by peer]
amyers has joined #sandstorm
NOTevil has joined #sandstorm
mort___ has joined #sandstorm
<paulproteus>
OK now my localhost Sandstorm speaks HTTPS. Now I should make my localhost Sandstorm actually respect config options that I so carefully designed last week.
bb010g has joined #sandstorm
jadewang has quit [Remote host closed the connection]
<paulproteus>
zarvox: I heard you say "Oh!" as if you had one more thing to say after the call but had already clicked to hang up!
<paulproteus>
So anyway feel free to ping me, and possibly oops on my part.
jadewang has joined #sandstorm
jadewang has quit [Remote host closed the connection]
jadewang has joined #sandstorm
gopar has joined #sandstorm
simonv3 has quit [Quit: Connection closed for inactivity]
jadewang has quit [Remote host closed the connection]
amyers has quit [Ping timeout: 268 seconds]
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
jadewang has joined #sandstorm
NOTevil has quit [Quit: Leaving]
<dwrensha>
investigating the jenkins failure...
<dwrensha>
eh... works locally. Let's see what happens with the next queued build.
bb010g has quit [Quit: Connection closed for inactivity]
<dwrensha>
looks like offer templates and ApiSessions are broken right now. Almost certainly my fault.
<zarvox>
dwrensha: So just to be clear about what broke there: offer templates generated a token and the text, but the token in the offer template was no good when we tried to use it?
<dwrensha>
no, there was a match failure in the newApiToken() Meteor method
<dwrensha>
... when `rawParentToken` was undefined
<zarvox>
Ahhh. And the tests caught it?
<dwrensha>
no, I caught it
<dwrensha>
we should have some tests for offer templates and API sessions, but currently we don't
<zarvox>
Ahh, okay.
<paulproteus>
Whoa I have not forgotten everything I ever knew about C++.
<paulproteus>
Surprising but useful.
<dwrensha>
I'm puzzled about the exact cause of the problem. It seems that somehow moving the function into a package caused Meteor to translate `undefined` to `null`.
<dwrensha>
fortunately, the code that I had to change is directly adjacent to a note about "Meteor bug #3877", which seems relevant.
larjona has joined #sandstorm
larjona has quit [Quit: Konversation terminated!]
amyers has quit [Read error: Connection reset by peer]
jadewang has quit [Remote host closed the connection]
datakid has joined #sandstorm
<datakid>
morning. I'm following the packaging instructions from the docs, and it's failing on "vagrant-spk up" with the issue "default: SSH auth method: private key" timeout. I can't see any provate keys, and I didn't ever make a private key. Where is the private key creation part of the howto?
<paulproteus>
Hi datakid
<paulproteus>
That's an implementation detail of vagrant-spk
<paulproteus>
You might run into this if you can't start VirtualBox VMs in general.
<datakid>
I've seen that there are a number of postings on SO etc about this, but their is no clear solution. I am sure that it's not a firewall problem - mostly because the system is telling me what the problem is, namely that there is a connection timeout immediately after mentioning a key I never created. Which, therefore, seems like the most likely candidate
<paulproteus>
What OS are you on?
<paulproteus>
And can you open up VirtualBox's graphical interface at all?
<datakid>
paulproteus, ok, thanks. While I understand it is there fore not a sandstorm issue, maybe it should go into the "requirements" list
<paulproteus>
If you can open up VirtualBox's graphical interface, then we can try to debug further that way.
<paulproteus>
datakid: Agreed. Would you file a bug against github.com/sandstorm/vagrant-spk ?
<datakid>
I am on Ubuntu trusty, it's a headless VM in the cloud
<paulproteus>
Oh, I see.
<paulproteus>
Yeah, on a VM in the cloud, you won't be able to spawn new VMs.
<datakid>
I think I have ssh -X though, so I could launch the VB interface? How do I do that?
<paulproteus>
And we (currently, at least) use one VM per Sandstorm app you are developing.
<paulproteus>
So basically, it won't work, and you should use vagrant-spk on e.g. your laptop if possible.
<datakid>
oh
<paulproteus>
Having said that, maybe we should come up with a better answer than "No" for your sort of situation.
<paulproteus>
BTW hi, and thank you for dropping by and asking your question.
<paulproteus>
Would you be willing to try it on your own machine?
<datakid>
Sure
jadewang has joined #sandstorm
<paulproteus>
Sweet.
<paulproteus>
BTW I notice your IP address starts with 128. Are you at a university somewhere?
jadewang has quit [Remote host closed the connection]
<datakid>
paulproteus, can I just get clarification wrt "on a VM in the cloud, you won't be able to spawn new VMs" - why is that?
<datakid>
paulproteus, yep Uni of Melbourne,
<datakid>
all uni's start with 128?
<paulproteus>
Cool! re: Uni of Melbourne
<paulproteus>
All 128.* is universities, I believe.
<datakid>
I didn't know that, thanks
<paulproteus>
Some universities are elsewhere, like MIT is just 18.*
<datakid>
they were early on the uyptake
<paulproteus>
The place I went to university is 128.220.
<datakid>
if not first on the board I'd imagine
<paulproteus>
Well 128.220.*
<datakid>
John hopkins?
<paulproteus>
As for why no VirtualBox on a cloud VM, the VirtualBox developers seem to require CPU extensions that cloud VMs don't make available.
<paulproteus>
Ya
<datakid>
sure
<paulproteus>
As for why we require VirtualBox, it's a kind of ironic tragedy that comes from me and zarvox trying to make the development tools cross-platform.
<datakid>
ok, I'll do it locally - I presume that the resulting image is generally deployable anyway
<datakid>
I also have one other question
<paulproteus>
Yup, totally should be generally deployable.
<paulproteus>
Great!
<datakid>
While installing on Ubuntu Trusty, why is the default VM debian/jessie?
<paulproteus>
Also basically hi, it's always super fun to talk to Sandstorm users, I hope you'll consider subscribing to the https://groups.google.com/forum/#!forum/sandstorm-dev email discussion list and staying in touch before you vanish into the happy ether with your questions all answered.
<paulproteus>
The vagrant-spk machinery creates one Linux virtual machine per Sandstorm app you're developing.
<zarvox>
datakid: Try running vagrant-spk up from a shell that doesn't have access to your SSH agent
<datakid>
I don't mind - I love debian and am soon moving my laptop back to it - just seems weird if you are relying on underlying OS for some of the
<datakid>
system in order to keep it small
<paulproteus>
We're actually using the debian/jessie image's underlying OS, ignoring whatever real OS you have installed.
<paulproteus>
That way, you can run the vagrant-spk tool on Windows or Mac OS and get reasonable results too.
<datakid>
sure
<zarvox>
I experienced a weird bug that I didn't have a ton of time to track down when my shell had access to an ssh-agent with my own keys loaded
<datakid>
zarvox, is that even possible? I'm not sure I can get a shell that doesn't have access to my ssh agent given that the system is in an OpenStack cloud
<zarvox>
Oh. Umm. I think there are some environment vars you could clear? One sec.
<paulproteus>
zarvox: FWIW I think that's not the problem here because we know vagrant-spk is being run on a cloud VM and so probably he can't create new VirtualBox VMs.
<zarvox>
unset SSH_AUTH_SOCK
<zarvox>
unset SSH_AGENT_PID
<zarvox>
Oh. That could also be a problem, but IIRC VirtualBox can run entirely with software emulation. It'll be slow, though.
<paulproteus>
I haven't been able to enable that mode lately, fwiw. I think it went away with during/with the 4.x series of VirtualBox.
<paulproteus>
But I'm not super-duper confident of that; there may be a way to get it back.
<paulproteus>
My inability to succeed with that is why the Sandstorm installer-tests uses qemu not VirtualBox.
<paulproteus>
(But yeah, I remain open to the notion that there is a way to make it work.)