kentonv changed the topic of #sandstorm to: Welcome to #sandstorm: home of all things sandstorm.io. Say hi! | Have a question but no one is here? Try asking in the discussion group: https://groups.google.com/group/sandstorm-dev
jfred has joined #sandstorm
frigginglorious1 has joined #sandstorm
frigginglorious has quit [Ping timeout: 260 seconds]
frigginglorious1 is now known as frigginglorious
frigginglorious has quit [Read error: Connection reset by peer]
XgF has quit [Remote host closed the connection]
XgF has joined #sandstorm
frigginglorious has joined #sandstorm
frigginglorious has quit [Ping timeout: 272 seconds]
frigginglorious has joined #sandstorm
frigginglorious has quit [Ping timeout: 260 seconds]
frigginglorious has joined #sandstorm
frigginglorious has quit [Ping timeout: 256 seconds]
nicoo has quit [Remote host closed the connection]
frigginglorious1 has quit [Read error: Connection reset by peer]
frigginglorious has quit [Remote host closed the connection]
frigginglorious has joined #sandstorm
frigginglorious has quit [Read error: Connection reset by peer]
frigginglorious has joined #sandstorm
sam_w has quit [Ping timeout: 260 seconds]
fkautz has quit [Ping timeout: 240 seconds]
fkautz has joined #sandstorm
sam_w has joined #sandstorm
<kentonv>
ugh google cloud is deprecating something or other API and it says our web server and Alpha VMs are still using the API but I don't know why they would be
<kentonv>
it's probably some google utility service on the machine doing it but who knows what or what will break if I don't fix it, maybe I won't be able to log in anymore???
<kentonv>
guess I'd better rebuild the machines from scratch
<kentonv>
machine metadata API. Basically there's an HTTP API exposed to each machine on a special IP address
<kentonv>
and the original version of the API was too SSRF-prone so they deprecated it
<kentonv>
which is a valid concern, sure
<JacobWeisz[m]>
Opinion: Every Google API deprecation is an excellent time to move off Google services.
<kentonv>
but like, these machines don't have SSRF vulns, pretty sure, and so this is just needless work for me
<JacobWeisz[m]>
That seems pretty unlikely to break Sandstorm... you think it'd possibly just murder the VM somehow though?
<kentonv>
I think it's very possible it will break SSH
<kentonv>
they have a thing where they deliver SSH keys to the machine via metadata...
<kentonv>
so that would be rather bad
<kentonv>
yeah I'm tempted to switch to Digital Ocean but I'd have to transfer 130GB of data
<kentonv>
just for Alpha... more for the web server
<kentonv>
and I'm lazy
<JacobWeisz[m]>
Like a third of Alpha is the app index grain now, I see...
<kentonv>
surprised it's not more TBH
<kentonv>
heh just the bandwidth will cost me like $15. Not that I care about $15 but it's amusing.
<JacobWeisz[m]>
I wonder... so it costs money to move stuff out of cloud services usually, due to data egress fees. ...Is that concept legal in the EU? Since you own your data and have a right to request it.
<kentonv>
heh I imagine any reasonable court would find those laws aren't intended to apply at that level.
<JacobWeisz[m]>
I don't know. Google Takeouts can be pretty huge and they'd probably take issue with Google charging for them.
<JacobWeisz[m]>
The EU lacks corporate personhood, so I imagine the right to your personal data wouldn't apply to any corporate accounts, but I wonder if a valid case could be made for personally-owned accounts.
<kentonv>
I mean, I don't know if they'd treat low-level cloud infrastructure the same as application-level personal data
<kentonv>
oof, Digital Ocean would be *much* more expensive for the RAM and disk requirements
<kentonv>
infrastructure has to just keep working forever. You cannot just go around telling people to rewrite things
<JacobWeisz[m]>
Yegge posts have always knocked it out of the park.
<abliss1>
"Google Cloud is still #3... and according to some internal sources, there’s some concern that they may sink to #4 soon" -- Is he saying they're behind Amazon and... Microsoft azure? What would the third other competitor be?
<abliss1>
IBM? Alibaba?
frigginglorious1 has joined #sandstorm
frigginglorious has quit [Ping timeout: 265 seconds]
frigginglorious1 is now known as frigginglorious
<kentonv>
I imagine IBM, they are way bigger than you think
<kentonv>
you don't hear about them as much since they pretty exclusively court gigantic enterprise customers
<isd>
Reminder: office hours tonight
<abliss1>
yeah, i don't hear anything about azure either, but i'm willing to believe they're big in enterprise.
<abliss1>
speaking of office hours... sorry to have drifted away without saying anything, but I kept hoping that the Remote Learning pain would ease and I'd get some more time to contribute to sandstorm. But now I have a new job starting next week, so I should probably excuse myself in a slightly-more-formal way. Sorry I never got around to doing all those things I said I'd do. :/
michaeln3 has quit [Remote host closed the connection]
michaeln3 has joined #sandstorm
michaeln3 has quit [Ping timeout: 256 seconds]
<kentonv>
I'm going to have to take alpha down for a few minutes for migration
<kentonv>
going down now
<kentonv>
it's back up
<kentonv>
that wasn't so bad
<kentonv>
except I can't get the smtp port open
<isd>
abliss: no worries. Best of luck in the new job.
<TimMc>
abliss1: Are you on the teacher or the student side of the mess?
<kentonv>
to be honest I cannot tell whether SMTP is open or not, because just about everyone blocks *outgoing* SMTP so I cannot test it from anywhere. Sheesh.
<JacobWeisz[m]>
Oracle is probably another silently large cloud. They have Zoom for one.
<JacobWeisz[m]>
And allegedly soon TikTok...
<kentonv>
probably no one cares about incoming SMTP to alpha... right?
<abliss1>
TimMc: parent of a kindergartner and a 2nd grader. Onsite 3 mornings / wk, remote the rest of the time. Exhausting.
<JacobWeisz[m]>
Probably not, kentonv
<JacobWeisz[m]>
Should anyone be using Alpha as their mail server? I think the answer, from a technical sense, is no.
<kentonv>
I think we only ever had one app that could receive e-mail and it was bad
<JacobWeisz[m]>
I mean I think we had Roundcube and Mailpile, but yeah, the experience was not there.
<kentonv>
hah totally forgot about mailpile (especially since I think it was never on the "new" app market?)
<JacobWeisz[m]>
Beyond the fact that probably nobody should be doing production work on Alpha in 2020.
<JacobWeisz[m]>
Correct, it never got polished up with all the metadata and stuff.
<JacobWeisz[m]>
And the lead guy on it has come and gone from working on it for ages. Roundcube Next died too, for all the hope modern open source mail was gonna have.
sam_w has quit [Ping timeout: 260 seconds]
<kentonv>
did roundcube next ever deliver anything? I seem to recall Sandstorm gave them $10k...
sam_w has joined #sandstorm
<JacobWeisz[m]>
I mean, I think there's a repo.
michaeln3 has joined #sandstorm
frigginglorious has quit [Read error: Connection reset by peer]
michaeln3 has quit [Ping timeout: 256 seconds]
frigginglorious has joined #sandstorm
<isd>
kentonv: can confirm SMTP is reachable on alpha from my VPS
frigginglorious1 has joined #sandstorm
frigginglorious has quit [Ping timeout: 265 seconds]
frigginglorious1 is now known as frigginglorious
frigginglorious has quit [Read error: Connection reset by peer]
<XgF>
<JacobWeisz[m]> I wonder... so it costs money to move stuff out of cloud services usually, due to data egress fees. ...Is that concept legal in the EU? Since you own your data and have a right to request it. <-- organisations are allowed to charge reasonable handling fees
<ill_logic>
Sorry for missing office hours twice - I don't want to miss this debate.