<mnutt_>
ocdtrekkie: have you looked into where all of the storage is going, via the grain backup? I'd imagine it'd be pretty easy to do a cleanup?
<mnutt_>
and yeah, I guess non-sqlite is pretty annoying for inspecting
<mnutt_>
pretty happy that one of the 3 people who rated Davros as "needs fix" (due to the Owncloud crashing issue) tried yesterday's update and is having success. I haven't yet had to rate anything "needs fix" yet so I don't know what the flow looks like, but maybe afterwards it could link to the app's issue tracker so that they could file and issue and get updated when a fix is present?
<mokomull>
ocdtrekkie: if it's InnoDB-backed, the files *never* get truncate()d. No idea about MyISAM since that's terribad.
<ocdtrekkie>
mnutt_: I tried to make sense of it once, but it's really not my expertise.
<ocdtrekkie>
TTRSS is supposed to purge old articles and stuff after x number of days, I don't know if it works correctly or not in the Sandstorm package. For all I know, 91 MB is just the largest the database has ever needed to be or something.
<ocdtrekkie>
It compresses super well as a backup.
<ocdtrekkie>
92.9 MB on Sandstorm, 8.3 MB as a backup file.
peterood has joined #sandstorm
isd has quit [Quit: Leaving.]
elimisteve has left #sandstorm [#sandstorm]
<digitalcircuit>
mnutt_: As a heads-up, just ran into this issue for the first time, found that someone else has it, too: https://github.com/mnutt/davros/issues/33 ("'#' in file and folder names causes problems") Maybe just a path escaping issue?
<digitalcircuit>
Otherwise, new Davros update seems really nice!
<mnutt_>
thanks! and yeah, definitely a path escaping issue. I've been meaning to figure out whether I can just special-case '#' or if I need to do something more comprehensive
* digitalcircuit
really likes the Delete All button to easily clean up the demo files.
<digitalcircuit>
Duly noted. If there's anything I can help test, let me know, though I suspect you don't need extra help. UTF-8 filenames, perhaps?
<mnutt_>
I expect the next version will have the demo files removed and empty directories will say "This directory is empty. [Upload some files] or just drag them in, or [add some demo files]"
<digitalcircuit>
mnutt_: Alright, something like that makes sense.
peterood has quit [Quit: Connection closed for inactivity]
<mnutt_>
so apparently libreoffice really, really wants /proc to exist :-(
<mnutt_>
> ERROR: /proc not mounted - LibreOffice is unlikely to work well if at all
<digitalcircuit>
mnutt_: Trying to package the web version of LibreOffice?
<mnutt_>
no, just generate some word doc previews
<mnutt_>
that'd be interesting too, though
DanC has quit [Ping timeout: 244 seconds]
<mnutt_>
I'm going to see what happens if I just build libreoffice from scratch without that check, but I don't have very high hopes
<digitalcircuit>
Preview functionality seems like a nice idea! Good luck :)
jemc has quit [Ping timeout: 250 seconds]
<mnutt_>
kentonv: would you happen to have any ideas for missing /proc workarounds?
<asheesh>
mnutt_: Hi!
<asheesh>
I have some ideas : P
<asheesh>
What are you running into?
<mnutt_>
asheesh: awesome, thanks!
<asheesh>
My favorite is to snapshot the files; let me show you what I did for Andrew Wansley
<mnutt_>
libreoffice checks "/proc/version" exists, and bails if not
<mnutt_>
I think if that was any type of file, it'd probably get past that
<asheesh>
- Other stuff, too, but probably you can do what I said actually.
<asheesh>
mnutt_: Let me know!
<mnutt_>
trying it now
<asheesh>
: )
Jan\ has quit [Ping timeout: 258 seconds]
Jan\ has joined #sandstorm
<asheesh>
(discussed more by PM)
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
DanC has joined #sandstorm
ckocagil_ is now known as ckocagil
elensil has joined #sandstorm
kentonv has quit [Ping timeout: 244 seconds]
kentonv has joined #sandstorm
ocdtrekkie has quit [Read error: Connection reset by peer]
ocdtrekkie has joined #sandstorm
amyers has joined #sandstorm
Jan\ has quit [Ping timeout: 252 seconds]
Jan\ has joined #sandstorm
amyers has quit [Ping timeout: 250 seconds]
amyers has joined #sandstorm
rolig has quit [Ping timeout: 264 seconds]
rolig has joined #sandstorm
amyers has quit [Remote host closed the connection]
rolig has quit [Ping timeout: 258 seconds]
rolig has joined #sandstorm
frigginglorious has joined #sandstorm
tantalum1 has joined #sandstorm
mnutt_ has joined #sandstorm
kentonv_ has joined #sandstorm
kentonv_ has quit [Client Quit]
synchrone has left #sandstorm [#sandstorm]
jemc has joined #sandstorm
synchrone has joined #sandstorm
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mnutt_ has joined #sandstorm
mnutt_ has quit [Client Quit]
frigginglorious has quit [Quit: frigginglorious]
frigginglorious has joined #sandstorm
mnutt_ has joined #sandstorm
Salt has quit [Ping timeout: 240 seconds]
Salt has joined #sandstorm
elensil has quit [Quit: Leaving.]
jsilence has quit [Ping timeout: 240 seconds]
jsilence has joined #sandstorm
frigginglorious has quit [Quit: frigginglorious]
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
eternaleye has quit [Changing host]
eternaleye has joined #sandstorm
mnutt_ has joined #sandstorm
wolcen has joined #sandstorm
Telesight has joined #sandstorm
rolig has quit [Quit: Quit]
aldeka has joined #sandstorm
ocdtrekkie_ has joined #sandstorm
* asheesh
waves, mornin'
rolig has joined #sandstorm
<JonTheNiceGuy[m]>
Evening :)
<JonTheNiceGuy[m]>
Or rather #TZAppropriateGreeting
<JonTheNiceGuy[m]>
I hacked out that bug I reported the other week, by reinstalling Sandstorm and not using sandcats. Sorry :( I was going to try to save the state for you to take a look through, but it had all the documents about an upcoming holiday and my wife was stressing
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<asheesh>
That's life; thanks for dealing with it in a way that worked for you.
<asheesh>
Sorry I wasn't able to help in a more timely way. I don't suppose you have a backup that reproduces the problem, or something, though? If not, then that's OK; I'll remember this exchange in case it comes up in the future for someone else.
<asheesh>
I guess the test is finding out how Google reacts.
mnutt_ has joined #sandstorm
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
wolcen has joined #sandstorm
mnutt_ has joined #sandstorm
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mnutt_ has joined #sandstorm
xet7_ has quit [Ping timeout: 252 seconds]
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mnutt_ has joined #sandstorm
wolcen has quit [Ping timeout: 276 seconds]
kxra has quit [Ping timeout: 260 seconds]
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<digitalcircuit>
ocdtrekkie: The end seems to sum it up fairly well for Google in general - "While Google Cloud offers some powerful features, they are not safe to use until Google abandons its “guilty until proven innocent, without an option to prove” threat response model."
<digitalcircuit>
...which is a good reason for Sandstorm at scale :)
wolcen has joined #sandstorm
wolcen has quit [Client Quit]
<mokomull>
Other than the people involved, there's nothing preventing cloud-hosted Sandstorm from having the same issue, no?
wolcen has joined #sandstorm
kxra has joined #sandstorm
isd has joined #sandstorm
wolcen has quit [Ping timeout: 264 seconds]
isd has quit [Remote host closed the connection]
isd has joined #sandstorm
<zarvox>
Yeah, but you can also run Sandstorm on your own bare metal, if that's a risk you want to hedge against.
<mokomull>
zarvox: Well, the article is about services which you can already run on your own bare metal, but the operator chose to use the Cloud™ instead.
<digitalcircuit>
mokomull: Good point! The advantage to Sandstorm in this case might be the ease with which you can switch from cloud-hosted to self-hosted - download grain backups, upload to your own server. A mass-export and mass-import option could make this more manageable.
<digitalcircuit>
I'm not sure if there's a way to reuse the work for Google Cloud compatible projects with other service providers.
<mokomull>
I wonder if a mass-export of Sandstorm grains would be any less cumbersome than backing up, e.g., a hundred-GB EBS volume. You're still fighting your various service providers in the time-domain.
<mokomull>
Single grains: yeah, way easier to export a document than it would be to pick that document out of the MySQL database out of the filesystem stored atop that cloud block device.
wolcen has joined #sandstorm
prettyvanilla has joined #sandstorm
prettyvanilla_ has quit [Ping timeout: 264 seconds]
isd has quit [Quit: Leaving.]
isd has joined #sandstorm
wolcen has quit [Ping timeout: 252 seconds]
wolcen has joined #sandstorm
Lionel_Debroux_ has joined #sandstorm
Lionel_Debroux has quit [Ping timeout: 276 seconds]
nwf has quit [Ping timeout: 265 seconds]
nwf has joined #sandstorm
wolcen has quit [Ping timeout: 244 seconds]
rolig has quit [Ping timeout: 265 seconds]
mnutt_ has joined #sandstorm
rolig has joined #sandstorm
Telesight has quit [Quit: Leaving.]
tantalum1 has left #sandstorm [#sandstorm]
wolcen has joined #sandstorm
xet7_ has joined #sandstorm
wolcen has quit [Ping timeout: 250 seconds]
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mnutt_ has joined #sandstorm
<ocdtrekkie>
digitalcircuit: I guess what Sandstorm mass backup would need to do to be efficient would be like we can currently do with file systems pretty easily: Only download changed or new files.
<ocdtrekkie>
asheesh: According to the author, Google reactivated his account four hours after he tweeted out the blog post.
mnutt_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
awans has joined #sandstorm
<awans>
so i'm like || <- that close to getting my distributed social bookmarking thing running on sandstorm
<awans>
i've got one-way sync working (using the hack-session HTTP get capnp call)
<awans>
but to get two-way sync working, i need to expose an API
<awans>
so imagine my sadness when on the plane home from my sister's wedding this weekend, i hit `vagrant-spk dev` one last time and got
<awans>
Error getting endpoint URL: hack-session.capnp:HackSessionContext.generateApiToken: rpc exception: remote exception: remote exception: remote exception: Error: generateApiToken() has been removed. Use offer templates instead.
<awans>
what do y'all think? is there any other option for me? I can use offer templates, but the UX is pretty janky for what I'm doing.
<awans>
(I think the best case scenario is telling the user to mysteriously copy paste a URL from one textarea to another..)
TheWhisperSP3 has quit [Ping timeout: 240 seconds]
<awans>
i understand for the git / mobile app use-case offer templates are quite elegant, but for apps that need to know their own URL (to, for instance, tell others to come say hello there), it's not awesome