You will never able to overwrite the files. It is immutable data. Instead, you will have to update the link to the latest version. People can still connect to old version if they wish. If you don’t want people to connect to old version, just delete the files. But again, it doesn’t exactly delete. The link will still be there, but they won’t able to see the data inside of that link.
Yes of course… but I’m emulating a noob and normal user… and they will see the option to delete and immediately next to it the option to add files and wonder why they cannot just do that all in one action.
Working without any issues so far on 3 different machines (3 different IPs)
- Laptop running arch
- VM running a VPN client
- VM on the wiki server
Great work guys.
EDIT: One issue though: Restarting the vault instance will create a new dir in /tmp while the old one is not deleted.
1 Like
You are running several Vaults I guess? Then each map is a Vault I think.
2 Likes
Who’s working on building a page to show how many nodes are currently running?
Would that even be a possibility with all the anonymity and security built into this beast?
3 Likes
Uhm, is there no ARM release for 0.8.1? You better get it ready before @happybeing sees that 
4 Likes
Are the vault chunks themselves stored in /tmp/ or just a list of vault chunks? A lot of people mount /tmp/ on tmpfs in RAM.
OurNEM has an awesome interactive 3D globe that shows nodes all around the world as you drag it to spin.
1 Like
I am now running the Vault software, Launcher and Demo app with some webpages. I have no idea on how to interpret all the data presented in the Vault app. Lots of errors shown.
I think the idea is that most of the nodes know only about a very restricted set of the whole network… While it’s certainly possible to aggregate that information (if enough of the nodes participate in it voluntarily) I’m not sure that lines up with the philosophy of the network 
1 Like
Is that not just normal /tmp behaviour?.. I understand /tmp cleans up itself after a time… it’s always a mess.
Maybe this is the intended data loss, but a safe launcher currently is pending registering ‘eternally’ for me. Eventually error in registering is the display in its gui window. Using current binaries:
drwxr-xr-x 2 testuser users 4096 May 17 xx:33 safe_vault-v0.8.1-linux-musl-x64
drwxr-xr-x 4 testuser users 4096 May 17 xx:34 safe_launcher-v0.4.4-linux-x64
drwxr-xr-x 4 testuser users 4096 May 17 xx:35 maidsafe_demo_app-v0.2.3-linux-x64
I had that initially but it became quick after a while. Are you running a new vault alongside that?.. I wonder the vault is trying to catch up with ~12MB of data, though perhaps the launcher should be unaffected.
1 Like
A current binary of the vault is running as well. I tried both, I could shut it down of course. it currently has around 40peers connected or at least in its routing table outputting to the console logging.
Um that is awesome!!! 
4 Likes
Leaving the launcher process running and retrying the registering process still fails. Shutting down the launcher now maybe its own net connections are somehow affected or well the whole net is crowded or it is currently the intended result to fail and to generate measurements and benchmarking data.
Hm after restarting the launcher, now trying my reg process it outputs an interesting line:
CoreError::MutationFailure::{ reason: DataExists, data: StructuredData { type_tag: 0, name: 3b9f…, data: d13934…71bd2c, previous_owner_keys: , version: 0, current_owner_keys: [“66c247…376dfe”], previous_owner_signatures: [“9fbeec…95ed0a”] }}
So did my previous registration attempts actually go through even thout it said that it failed?
To me that looks like a simple Google Earth data overlay. If the data exist that can be done within seconds, but I am unsure whether the location of vaults are transparent to the network (#privacy)
Well uncertain but login giving my creds fails as well. Anyways will try again some more stuff later. Cheers.
1 Like
I wonder do the seed nodes function only to establish connections or are they carrying more work than perhaps they should?