Ah, you mean the name
of the node?
This is flushed to the config file and got read back once restart.
Actually I can’t, but I can tell that after 3 minutes it hasn’t joined or started a new trial.
I just do not understand how node can survive restart (process stop/start) with reduced age if its “name”/key/whatever is stored only inside RAM, which is cleared with process stop.
If it is not cleared, then where it is located (on HDD)?
(For network to be able to identify node, node by itself should know how it is identified, which means that such information should be stored somewhere)
According to the log there have been no attempts to reconnect for much more than 3 minutes, but I see something running here:
sascha@Knut:~$ ps aux | egrep '.safe/'
sascha 137879 0.0 0.0 31456 4792 ? Sl Apr15 0:08 /home/sascha/.safe/authd/sn_authd start --fg --listen https://localhost:33000 --log-dir /home/sascha/.safe/authd/logs
sascha 180180 0.1 0.1 98892 30244 pts/1 Sl 11:34 0:01 /home/sascha/.safe/node/sn_node -vv --hard-coded-contacts ["165.227.229.100:49408","188.166.174.88:54107","178.62.122.173:58245","138.68.135.139:49440","139.59.162.138:52910","165.227.229.103:59359","165.227.229.33:12000","165.227.232.158:50965","138.68.130.204:50087","165.227.229.104:33663","165.227.229.55:32936","138.68.146.170:32859","165.227.229.59:47925","165.227.224.69:34758","178.62.74.247:51223","178.62.32.210:39347","138.68.180.216:49021","165.227.229.131:53334","165.227.229.96:46868","165.227.224.102:34793","188.166.175.92:33968","165.227.229.148:47735","188.166.172.140:37214","165.227.229.129:57822","46.101.17.63:46664","196.244.192.46:60606","138.68.182.161:34595","165.227.229.71:54735","138.68.167.124:43149","165.227.229.57:45221","178.62.10.198:51741","139.59.184.250:50196","165.227.225.237:60110","165.227.229.150:33874","165.227.226.230:33552","46.101.72.129:37483","139.59.177.49:42192","46.101.61.67:59088","165.227.229.142:60960","165.227.229.70:48326"] --root-dir /home/sascha/.safe/node/local-node --log-dir /home/sascha/.safe/node/local-node

If it is not cleared, then where it is located
right, I cross checked the sn_node code, seems the routing key does not flushed to the HDD config file.
will double check with other guys and get this corrected asap.
thx for the finding again.

If it is not cleared, then where it is located (on HDD)?
This is a bug, it should be non volatile but is written as a None
value. Nice find again @Vort @qi_ma has confirmed the bug now and we can sort that.
@Vort is leading bughunts. I wonder if @StephenC and @JimCollinson could BGF look at rewards for finding bugs and more for supplying fixes? So a new never seen before bug in the node (not client for now)? Thinking out loud!
@Vort deserves some SNT
If I try to use safe keys create --test-coins --for-cli
I get the following
error: Found argument '--for-cli' which wasn't expected, or isn't valid in this context
USAGE:
safe.exe keys create --test-coins
For more information try --help
That means the CLI did not understand the command, so using --help should show available commands. (it’s a usual cmd line tool, they shoudl mostly all do this). Keep going though, it’s worth mastering the old command line.
No matter what I try it always hangs…Tried now safe keys --test-coins
EDIT: I deleted everything again and started from the beginning, now it hangs again with safe keys create --test-coins --for-cli

error: Found argument '--for-cli' which wasn't expected
This means that you were on the older version of the CLI at that time, hence why deleting everything & starting again at least got you past this point. However it seems that the testnet has more or less ground to a halt at the moment - I think this is probably a good time to declare it offline…
Update from MaidSafe HQ
We feel that now is the right time to declare Fleming Testnet v3 offline for community use.
We may leave our Digital Ocean nodes online a little longer for staff to continue their investigations, but this testnet has now served its purpose by highlighting the next areas that we should direct our attention to, therefore no more testing from the community is requested until Fleming testnet v4 is announced.
Thanks to the whole team about this one, and special thanks to @yogesh and @lionel.faber for choosing the work over holiday! Much appreciated, though I hope this would stay as an exception
As a community member, it was nice to get this chance to participate.
I think I have seen some new people in Github contributing to the code, am I right?
Offline again?
Yes see the post just a few above

We feel that now is the right time to declare Fleming Testnet v3 offline for community use.
We may leave our Digital Ocean nodes online a little longer for staff to continue their investigations, but this testnet has now served its purpose by highlighting the next areas that we should direct our attention to, therefore no more testing from the community is requested until Fleming testnet v4 is announced.
I can’t believe I missed iterations v2 and v3. Will there be an testnet v4 coming out soon? These tests seem to be zooming by fast.
Next week. Hopefully Wednesday or Thursday. I think I read that somewhere.
No dates have been given and personally I expect it to take longer. Just a guess.
I expect Tuesday, so I hope they announce the T4 day in advance to have more nodes connected.