Initial commit (a clone of the jsmachines project as of v1.15.3)
This commit is contained in:
commit
a5e3e6a59d
714 changed files with 130602 additions and 0 deletions
5
logs/.gitignore
vendored
Normal file
5
logs/.gitignore
vendored
Normal file
|
|
@ -0,0 +1,5 @@
|
|||
*.json
|
||||
*.log
|
||||
bugs.html
|
||||
bugs/
|
||||
users/
|
||||
20
logs/README.md
Normal file
20
logs/README.md
Normal file
|
|
@ -0,0 +1,20 @@
|
|||
Logs
|
||||
---
|
||||
This is where Node website activity is logged, and where per-user data is stored (temporarily).
|
||||
|
||||
Copies of the remote AWS server logs are saved here as well, whenever you run [awsget.sh](../bin/awsget.sh)
|
||||
or [awspush.sh](../bin/awspush.sh). See the [Bin](../bin/) directory for details.
|
||||
|
||||
Also, to minimize the number of random root directories, I've moved all the per-user data into this folder
|
||||
as well. [users.log]() contains all the user IDs that are allowed to have per-user machine states and disk
|
||||
images on the server.
|
||||
|
||||
Per-user support is still in the experimental phase, and security is woefully inadequate. At the very least,
|
||||
every user should also have a "secret key" (the hash of which would also be stored in users.log), not to mention
|
||||
login and logout pages, key request and recovery pages, https support, and so on. That's an entire infrastructure
|
||||
I'm not interested in supporting yet, so for now, the integrity of per-user support relies on the use of
|
||||
obscure user IDs, and the users of those IDs avoiding open, unencrypted networks.
|
||||
|
||||
Storing user data on the same server as the web server is also temporary, because this (or any other folder)
|
||||
can be completely wiped out every time a new version is pushed to AWS. So we will need a completely different
|
||||
strategy for preserving per-user data.
|
||||
Loading…
Reference in a new issue