<georgeowell>
I have never had any problems self hosting :)
<georgeowell>
Debian stable and Sandstorm. Auto updates on both :)
<georgeowell>
We would just scale vertically if we needed to
<ocdtrekkie>
coffeebird[m]: Is it self-hosting if it's on Digital Ocean?
<ocdtrekkie>
When I switch to self-hosted Sandstorm, it'll be somewhere I physically control, which adds a lot of extra challenges/things that can go wrong.
<coffeebird[m]>
hm. no?
<coffeebird[m]>
we had hardware to run it on, but wasn't really secure, and could easily be unplugged by a startled cat
<ocdtrekkie>
And when I count the full range of possible problems I can have, I have to include that one of my PCs lit on fire before. :P
<ocdtrekkie>
Mostly the key thing for me is that I :can: self-host it. As long as I can self-host it, I'm reasonably comfortable trusting Kenton to do it for me until there's some reason for me to not.
samba_ has quit [Quit: WeeChat 2.0.1]
Zertrin has quit [Read error: Connection reset by peer]
Zertrin_ has joined #sandstorm
jemc has joined #sandstorm
samsmith123 has joined #sandstorm
<samsmith123>
good evening
<samsmith123>
anyone here?
<simpson>
Sure, but I probably can't answer your other questions.
<kentonv>
I'm here, and I probably can answer your questions. ;)
<samsmith123>
thanks kenton
<samsmith123>
i upgraded my ec2 instance for more memory and now the mongo.log says i have insufficient space
<samsmith123>
the ec2 instance also changed the ip address if that is of any help
<TimMc>
samsmith123: You moved the Sandstorm instance to a new, larger instance size?
<samsmith123>
so the sandstorm service is not running
<samsmith123>
@TimMC yes
<kentonv>
samsmith123, it looks like Mongo needs more free disk space
<TimMc>
(as opposed to having the install be on an EBS volume or something)
<kentonv>
(I just saw your support message)
<kentonv>
Mongo is silly, it wants like 500MB free on the disk or it won't even start
<samsmith123>
i have 54GB free
<kentonv>
hmm
<kentonv>
Mongo seems to think otherwise. Is it possible that /opt/sandstorm isn't on the right disk?
<samsmith123>
the sandstorm folder is in /opt/sandstorm
<samsmith123>
the host ip at the top of the mongodb file is the old ip, how can i change it to the new one?
<samsmith123>
nvm is the same
<TimMc>
samsmith123: If you run `df -h` on the instance do you see anything unexpected in terms of small disks, disks nearly full, etc.?
<kentonv>
samsmith123, sorry, I'm not sure how to help at this point. For whatever reason, Mongo thinks that the hard drive containing /opt/sandstorm/var/mongo doesn't have enough free disk space... you'll have to figure out why that is
<TimMc>
(The reason I asked about EBS was in case you had forgotten to mount the EBS volume or something like that.)
<samsmith123>
can i move the sandstorm folder to a different directory? one with more space available?
<kentonv>
samsmith123, yes. You'll have to update the symlink in /usr/local/bin/sandstorm and also the init files (systemd or init.d, under /etc) to point to the new location.
<kentonv>
alternatively, you could set up your mount table so that /opt/sandstorm is on a different drive
<kentonv>
but either of these might be tricky if you aren't familiar with Linux system adminsitration
<samsmith123>
yeah that would be tricky for me. i deleted a backup file i had which was taking 1.3GB
<samsmith123>
after that i ran the service start and its running.
<samsmith123>
i will eventually run out of memory got all team on this server so will need to find a way to use the available memory on my ec2, it says 54GB overall but not sure why this folder onl7 has 7
<kentonv>
I have to head out now, sorry. Good luck figuring it out!
<georgeowell>
Just added a new user to our sandstorm and they apear as active but they just aren't there when I try and add them via username to a collection or grain
<georgeowell>
Ok now I can see them but can't invite them by username. Email seems to work fine.
<georgeowell>
hmm seems to work ok now. Must take a while to propogate or something.
pie__ has joined #sandstorm
pie_ has quit [Ping timeout: 264 seconds]
jemc has joined #sandstorm
ocdtr_web has joined #sandstorm
<ocdtr_web>
FYI it looks like Oasis may be down again.
<ocdtr_web>
Kenton's real job project is at the top of HN today though, so his response time may be slow.
<ocdtr_web>
I did send him an email to make sure he was aware, but as soon as I saw that on HN, I kinda figured it might be a little while until he got the chance to look at it today. :/