I am new to sprites, so I might be doing something wrong.
I created the acf-dev-templ sprite this morning. It’s in the DFW region. When I try to connec td it was hanging earlier today. Now it just times out:
$ sprite list
eric-ayers: 0 / 2 running 1 / 2 warm (2 total)
┌──────────────────────────┬──────────────┬───────────────────────────────────────────────┬─────────────┬──────────────┐
│ NAME │ STATUS │ URL │ CREATED │ LAST RUNNING │
├──────────────────────────┼──────────────┼───────────────────────────────────────────────┼─────────────┼──────────────┤
│ acf-dev-templ │ warm │ xxx.sprites.app │ 2h ago │ 49m ago │
│ dailyspaceimage-dev-tmpl │ cold │ yyy.sprites.app │ 3 days ago │ │
└──────────────────────────┴──────────────┴───────────────────────────────────────────────┴─────────────┴──────────────┘
$ sprite console -s acf-dev-templ
Error: failed to start sprite command: failed to connect: read tcp 172.20.25.250:33006->169.155.48.226:443: i/o timeout
Something else that seems unusual: Although my acf-dev-templ sprite shows up with sprite list, I don’t see it from the web dashboard.
Hm… That looks like the older Machines platform’s dashboard, whereas Sprites have their own, I think:
I don’t use them myself (yet, 
), but my guess would be that you go to sprites.dev and then Sign In.
Oh yes, it does look different from the Sprites platform. Still, the instance in question has been marked as ‘warm’ all day long. Is there any way to see a log when the command sprite console
fails or why it failed to transition from warm to cold?
I tried restarting it, got an error. Went away for an hour and it’s cold finally, so whatever the problem was, it’s resolved.
As mayailurus said, the two dashboards are separate. I’m glad you were able to find your sprites in the sprites dashboard.
As for not being able to connect, were you ever able to connect to that sprite? Maybe right after you created it?
@aezell_fly thanks for the response.
Yes, I was able to connect to the sprite. I cloned my repo and was busy installing prerequisites. My project uses docker so I was probably at about the point where I was working on that when I got locked out. I created a new sprite to do the same thing and have been successful with it to date for developing, running tests, etc. It’s a pretty beefy project that uses bazel and docker (to run some GCP emulators.) Everything works, it’s just much slower than even my 7 year old laptop running WSL. The bottleneck seems to be disk I/O from what I can tell. Since I’m using them to run agents in the background, I can live with it and save my local linux development servers for my interactive work. I thought about just leaving it around but was concerned it might eat up my credits when warm (In retrospect, it didn’t seem to)
The only time a sprite will incur cost is when it is actively running. The difference between warm and cold is startup time when you reconnect and whether or not certain processes will resume from where they stopped. Neither warm or cold count as active usage and so won’t incur costs.
If you get curious and find out what kinds (or size/speed) of disk I/O seems problematic, we’d love to hear about it to see if we can improve things for you.
