* juri_ particles.
gopar has joined #sandstorm
dwrensha has quit [Ping timeout: 240 seconds]
gopar has quit [Quit: Leaving]
jadewang has quit [Remote host closed the connection]
jadewang has joined #sandstorm
leee has joined #sandstorm
leee has quit [Remote host closed the connection]
larjona has joined #sandstorm
jadewang has quit [Remote host closed the connection]
tmpproteus has quit [Ping timeout: 264 seconds]
kentonv has quit [Ping timeout: 260 seconds]
tmpproteus has joined #sandstorm
jadewang has joined #sandstorm
kentonv has joined #sandstorm
jadewang has quit [Ping timeout: 250 seconds]
_whitelogger____ has joined #sandstorm
ocdtrekkie has joined #sandstorm
_whitelogger___ has quit [Ping timeout: 240 seconds]
jinnko has quit [Ping timeout: 240 seconds]
cbaines has quit [Ping timeout: 240 seconds]
itscassa|away has quit [Ping timeout: 240 seconds]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 244 seconds]
cbaines has joined #sandstorm
itscassa|away has joined #sandstorm
_whitelogger____ has joined #sandstorm
jinnko has joined #sandstorm
dwrensha has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 244 seconds]
zeroish has joined #sandstorm
mort___ has joined #sandstorm
larjona has quit [Quit: Konversation terminated!]
snolahc has quit [Quit: Disconnecting]
snolahc has joined #sandstorm
itscassa|away has quit [Ping timeout: 240 seconds]
itscassa|away has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 265 seconds]
<dwrensha> argh SMTP configuration errors
<dwrensha> Exception while invoking method 'testSend' Error: connect ETIMEDOUT
<dwrensha> ^ it takes several minutes for this error to show up, and until then the client is frozen
<dwrensha> trying to debug my connection problem here
<dwrensha> ...
<maurer> Hrm. Evidently my version of node is either too new or too old for capnp-node
<dwrensha> 0.10.x is what you need
<dwrensha> you likely have 0.12.x
<dwrensha> "nvm" might be able to help you
<maurer> Ah yeah, I have 0.12
prosodyContext has quit [Remote host closed the connection]
fkautz has quit [Remote host closed the connection]
jksonc has quit [Remote host closed the connection]
bpierre___ has quit [Remote host closed the connection]
<dwrensha> If I'm sending a message over SMTP, does my server need to be accepting incoming connections?
<dwrensha> I'm having trouble with outgoing SMTP on a local dev server
<dwrensha> (going through mailgun)
<maurer> Normally no. It's possible some services may do that, but SMTP does not normally involve you listening for a connection.
<maurer> (if you're the sender)
<maurer> dwrensha: Have you tried manually telnetting to the target and sending what you think should be sent, or using sendmail or somesuch to verify that you can send mail outside the context of the webservice?
<dwrensha> ^ this program succeeds if I run it from my publicly accessible server
<dwrensha> but fails if I run it on my local NAT'ed VM, with Error: connect ETIMEDOUT
<dwrensha> I suppose the NAT could be misconfigured somehow
<dwrensha> outgoing ssh and http work find from that VM
<dwrensha> *fine
<maurer> dwrensha: Is your nat on a home internet connection?
<maurer> dwrensha: Your ISP could be filtering, thus the suggestion of checking telnet
<dwrensha> OMG if it's comcast's fault I'm going to be so angry
<dwrensha> I suppose there are two NAT's happening here
<dwrensha> (though I admit that I'm no expert on network configuration)
<dwrensha> there's the thing set up by my cable modem
<dwrensha> and then there's a private network I set up for my virtualbox VM
<dwrensha> so that it can have a static IP only accessible locally
<maurer> dwrensha: Comcast recommends using 587, and http://blog.mailgun.com/25-465-587-what-port-should-i-use/ says that mailgun endorses using this too
<maurer> tl;dr it's probably comcast's fault, switch 25 to 587 and it will probably work
<dwrensha> that did the trick! thanks
<dwrensha> but whyyy
<maurer> The answer to why is that 25 is what usually gets used by spambots
<maurer> so to avoid getting their ips filtered, ISPs will frequently filter it
<dwrensha> maurer: you're telling me this whole exercise was an elaborate captcha?
<dwrensha> :\
<maurer> Basically
<dwrensha> let's make sure that nobody tells the bots about port 587
larjona has joined #sandstorm
<maurer> Do I need to have a successful build before ekam uses a cache?
<maurer> Waiting 10-15 minutes to figure out if I've fixed a build error is getting really old
<dwrensha> maurer: you should look at what `make continuous` does in Sandstorm's Makefile
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 264 seconds]
jksonc has joined #sandstorm
bb010g has quit [Write error: Connection reset by peer]
kamalmarhubi has quit [Remote host closed the connection]
fkautz has joined #sandstorm
bpierre___ has joined #sandstorm
NOTevil has joined #sandstorm
bb010g has joined #sandstorm
prosodyContext has joined #sandstorm
kamalmarhubi has joined #sandstorm
jadewang has joined #sandstorm
<kentonv> dwrensha: Almost all ISPs block outgoing port 25 for spam reasons.
<dwrensha> should we change our default to 587 then?
<kentonv> you can set any port in the smtp:// url
<kentonv> IIRC 587 is for "email submission" which is actually a subtly different protocol
<dwrensha> whaa
<kentonv> but sendgrid accepts mail on port 2525
<kentonv> for this reason
<kentonv> since they do their own spam management
<kentonv> "email submission" is for, say, sending mail from your gmail account
<kentonv> but sandstorm wants a relay -- a server that will send from any address in the domain
tmpproteus has quit [Ping timeout: 244 seconds]
mort___ has quit [Quit: Leaving.]
mort___ has joined #sandstorm
mort___ has quit [Client Quit]
isd has joined #sandstorm
<XgF> kentonv: 587 is still SMTP...
<XgF> The main difference is that since an MX record never specifies a port, you can't SMTP to a receiving server directly
<XgF> (err, a receiving server won't be on port 587)
<XgF> destination server! that's the noun I was looking for
<paulproteus> We're doing email submission, and I think using 587 outbound makes sense.
<XgF> Yes, Mail Submission is when you're sending something to a relay
<XgF> It basically just means "(hopefully) authenticated SMTP"
<kentonv> hmm
<kentonv> well in that case, go ahead and specify port 587 in the smtp URL.
<dwrensha> ^ I'm starting to see this exception on my sharing-ui branch.
jadewang has quit [Remote host closed the connection]
<maurer> OK, almost there, finally got sandstorm itself building, now I just need to figure out how to deal with the meteor portion
<maurer> Actually, can I get a working spk without using meteor?
<maurer> It looks like I have one built, but I don't know that it will work correctly without a running sandstorm instance
jadewang has joined #sandstorm
jadewang has quit [Remote host closed the connection]
<maurer> Hrm, looks like no, spk dev seems to require the installed sandstorm server, which I guess I kind of expected
jadewang has joined #sandstorm
larjona has quit [Remote host closed the connection]
<paulproteus> `/win 2
<paulproteus> ...
<paulproteus> maurer: I think it will unless you use the Sandstorm-specific bits, like 'spk dev', but I haven't verified this, fwiw.
<paulproteus> If you make a sandstorm-files.list by hand and then 'spk pack' it, I expect success probably.
<maurer> paulproteus: Yeah, if I thought sandstorm itself had a hard to isolate build, meteor is even worse :(
<maurer> Unfortuantely, today is the last day I can afford to spend trying to do this, so if I fail today, I'll probably have to hold off on playing with sandstorm for a few weeks
isd has quit [Remote host closed the connection]
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
ocdtrekkie has joined #sandstorm
aldeka has joined #sandstorm
larjona has joined #sandstorm
ocdtrekkie has quit [Remote host closed the connection]
aldeka has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
isd has joined #sandstorm
NOTevil has quit [Quit: Leaving]
geofft has quit [Quit: leaving]
<maurer> Hrm. Welp, looks like I've failed. Meteor evidently hasn't been able to be coherently installed since version 0.52
<maurer> :(
isd has quit [Remote host closed the connection]
<paulproteus> Wait, what, hmm?
<maurer> paulproteus: the tl;dr is I was trying to get sandstorm to run on my devbox, but I burned two days on it, and I can't afford to burn anymore
<maurer> *any more
* paulproteus nods.
<paulproteus> Sorry to hear about it.
<paulproteus> I'm curious if the Meteor 0.52 thing comes from somewhere.
<paulproteus> Maybe Meteor used to document how to install it well, and now doesn't?
<paulproteus> But anyway, sympathies. If you have tips for what we could do to improve the situation, I'd love to hear it.
<paulproteus> One answer seems to be "Allow spk to compile without a working Meteor install".
<paulproteus> Would that be useful to help with your issues?
<paulproteus> Separately it seems like "Provide a reasonable way to install Meteor from source" would be useful, though that's something that might require help from Meteor.
<maurer> paulproteus: The 0.52 thing came from meteor previously complying with npm packaging
<maurer> When it was installable as an npm package, you could actually build/install it separately
<paulproteus> I see, interesting.
<maurer> re: spk, I've successfully built it without meteor
<maurer> Honestly, if I were to have a feature request it would probably along the lines of "allow spk dev to point at a remote server"
<maurer> because then I could have installed an incomplete sandstorm and at least had the dev toolkit
<paulproteus> I want that too. Good to know.
<maurer> but I suspect that would be more complicated than it sounds
<paulproteus> I think it is; am blanking on why atm.
<maurer> It doesn't actually work, but at the point where it breaks, everything other than the shell has been produced for the bundle
<maurer> I half suspect it might work if I could magic up a packageable meteor
<maurer> paulproteus: In any case, this is largely irrelevant because nixos is a small community, but I think I've identified some hurdles you're going to hit packaging it for debian:
<maurer> * Packages are pulled in via git clone, and not just for their build products
<maurer> * Some files used from the packages are not in the release source tarballs, e.g. capnpc.ekam-rule
<maurer> * Some sandboxing systems around the build can break LD_PRELOAD of intercept.so
<maurer> * Good luck packaging meteor
<maurer> I think all the other things I had to deal with will be non-issues on debian
mcpherrin has left #sandstorm [#sandstorm]
<maurer> In any case, sorry for the spam the last few days. ttyl.
<paulproteus> No probs at all; sorry about the trouble it caused you and I hope we can get past those issues sometime soon.
decipherstatic has quit [Ping timeout: 244 seconds]
NOTevil has joined #sandstorm
<kentonv> paulproteus maurer: Technically speaking it might actually be quite easy for spk dev to point at a remote server, since the FUSE filesystem sits on top of a Cap'n Proto capability. However, performance would likely be poor.
<maurer> kentonv: I mean, if it's spk dev, I don't think that would be too terrible, right?
jadewang has quit [Remote host closed the connection]
<maurer> A bit of latency on requests when developing isn't completely insane :P
NOTevil has quit [Quit: Leaving]
<larjona> paulproteus I've sent a mail to the list with my status update about Framadate and links to spk and git repo. I'm going to bed now, it's late here. Tomorrow it will be better!
larjona has quit [Quit: Konversation terminated!]
balrog has joined #sandstorm
neynah has joined #sandstorm
<paulproteus> ++