I know that zikzak and goreo and probably others I’m forgetting spent a lot of time looking at different hosting arrangements, so I trust their judgment on how we got here. Our current plan is described like this:
If anyone wants to go through the Digital Ocean pricing and suggest an alternative, I’m open to it:
You seem like a tiny D
Your face is a tiny!
I wonder what that 100GB/TOR1 is doing?
TOR1 refers to the Toronto data center hosting the droplet. Not sure about the 100gb offhand.
Edit: oh it’s marked as 160gb (disk size for that droplet type) plus 100gb. Think that’s an extra 100gb volume that was added. Maybe for log storage or backup storage or something?
Do we get any support with the hosting company? I didn’t see that extra 100GB disk in the df results above.
Doesn’t look like they have any add ons for storage. The next tier doubles storage for twice the price.
Would be interesting to know about the extra 100 gig though. That would solve everything if it were accessible.
Sounds like the add-on drives are way slower than the 160GB one that comes with the droplet. These guys mention they only put uploads on them and that it still performs okay. Not sure about the mechanics of doing that with our setup though. Making use of added storage volumes - hosting - Discourse Meta
I think the volume would appear like a second disk/device in the OS.
The general idea would be that you’d take a specific directory that fills up (say image uploads), and put that on the second volume instead.
Guide to leveraging an additional DO volume in discourse.
Something like this may have already been completed by a past admin, I can’t recall.
So the site went down again, and I feel pretty confident it’s because of disk usage.
I ran the launcher cleanup as suggested here, and that recovered about 2-3GB. But I think that’s going to be a very temporary solution, since we’re still at about 99% usage.
We had a couple of backups online, and I was hoping that deleting one of those backups would do the trick, but it doesn’t seem to have had any effect. (Maybe the backups are stored somewhere else–like AWS–that doesn’t affect our actual Discourse usage. I don’t know, I’m pretty much out of my depth here.)
I’ll try digging around for additional fixes, but man it would be great if we had someone capable to help out.
I’m not tech capable so this is where my thoughts go: can we solve this by upgrading oir hosting plan? Last i heard we had abundant funds but i have not kept up with any of the site governance discussions.
Update:
Further on down in this thread, someone recommended to look for leftover backups:
This was different from what I was seeing on the admin control panel in terms of backup files:
Previously, I was manually deleting those backups in the control panel, but doing so didn’t free up any space.
I think these “leftover” backups that omarfilip was talking about might have been failed backups of some sort? Anyway, when I looked in that /standalone/backups/default directory, I found two very large files:
Deleting these two files (which I think were unsuccessful backups, since they don’t show up in the admin control panel) actually did free up a bunch of space. We’re now at 73% usage of the 160GB, which should mean that we’re fine for the near future.
I’ll post an updated funds situation today or tomorrow. But I’m really hesitant to upgrade even more (unless it’s really inexpensive). I feel like our rather small site shouldn’t even cost as much as it currently does (~$68/month).
Thanks for digging into this and possibly resolving it spidercrab.
How much space is taken up with uploaded media like pics and whatnot? Current maximum is 8mb per upload. Also, if I link some photograph which is hosted somewhere else does it make a local copy for some reason?
Edit: Doesn’t seem to. I seem to recollect more than one occasion when I’d posted media of some kind and my post sort of auto-updated a few moments after posting. Maybe I dreamt it.
73% usage still seems a little high to me. There are probably other things that could be cleaned up as well but I’m no expert (old logs, etc.)
From recent conversation, it seems like this extra 100GB drive isn’t being used? If so, that’s something we should pursue since I think it’s an extra $10/month for it. 
Does anyone know if this DO Metrics agent is installed on the droplet? It would be great to get this setup and setup an alarm for something like 90% disk usage to get a heads up and remediate before the site goes down.
I am willing to help with these tasks, I just don’t really know when I’ll have the free time
Slight update - it’s at 65% usage, with 55GB of 155G listed as available. But that doesn’t really move the needle if you’re concerned.
Absolutely no idea. Would be happy to have anyone dig into this to figure it out…
I would give my pinkie finger if I could get my phone storage down to 65%, seems like a win to me.
Spidercrab seems to be doing a good job, but has stated that “it would be great if we had someone capable to help out” and it would be nice to have some redundancy.
@suzzer99 any interest in joining the “keep the forum alive” team? Anyone else want to?
I’m glad to help. I told Spidercrab he could contact me on this stuff.
But just FYI this stuff isn’t really my area of expertise. I’ve meticulously avoided learning much low level sysadmin stuff my entire career. Those guys are always grumpy and stressed out and look like they’re going to keel over of a massive heart attack at 55.
