How fast? I’m not online 
I seem to have been able to put this 700 MB file, but as usual, verification has failed.
Does the Timeout setting affect puts as well a cats?
sascha@sascha-Librem-15-v3:~/Safe_scripts$ time safe files put ../Safe_put/Eraserhead.avi
FilesContainer created at: "safe://hyryyryynr9wogz5qusaxrrtbd3y6xyhq15b5qaz5zzx3inp6jgcrobdxn9soeuy?v=hgip1akd6y1eg5cnzxjfqayrb7yyyobuttozu3uwcoggpcnoq1aho"
E ../Safe_put/Eraserhead.avi <Content may have been correctly stored on the network, but verification failed: safe://hy8ayyyybqf6yujtf8pjswrhtz3iftzfjnzek9yq4af975jhg3gtrekzjw66y>
real 25m35,448s
user 10m27,676s
sys 4m3,766s
1 Like
I had put failing, and succeeding increasing timeout, so I’d say yes.
To get eraserhead I had to increase to 1500:
SN_CLI_QUERY_TIMEOUT=1500 safe cat safe://hy8ayyyyb
qf6yujtf8pjswrhtz3iftzfjnzek9yq4af975jhg3gtrekzjw66y
> Eraserhead.avi
2 Likes
Why not increase Timeout to a million? It doesn’t slow anything down, does it? If I understand correctly, it’s just a threshold for when a put/cat stops trying.
1 Like
About 10 mins for me for Superfly 
2 Likes
Just tried superfly again with the timer :
time safe cat safe://hy8ayqyyb1jiokniu9oou954rra7msjxxus9yg1zwytf3py4eh1xyt1ifch5y > Superfly_1972.mp4
safe cat safe://hy8ayqyyb1jiokniu9oou954rra7msjxxus9yg1zwytf3py4eh1xyt1ifch5y
40,90s user 25,41s system 75% cpu 1:28,41 total
Not sure what it means (total time = 1’28’'41 = user + system + something else ?) but it was definitely less than 2 minutes. Fast enough !
My connection speed is around 100Mb/s
2 Likes
This 375 bytes file was because I catted the container, not the file itself! 
But putting and catting are working fine. Larger puts always throws this “verification failed”, but they are there.
1 Like
Getting closer, try 2500 timeout
My node, still not accepted, fell over with the following error at 18:46. I only just noticed and started it again.
Encountered a timeout while trying to join the network. Retrying after 3 minutes.
Error:
0: Cannot start node (log path: /home/safe/.safe/node/local-node). If this is the first node on the network pass the local address to be used using --first
1: Cannot connect to the endpoint: Failed to query our public address from peer 165.227.227.142:12000
2: Failed to query our public address from peer 165.227.227.142:12000
3: We did not receive a response within the expected time
Location:
sn/src/bin/sn_node.rs:315
Backtrace omitted.
Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
1 Like
I’m trying to put a folder of 8,36GB. I’m sick of network not getting full and accepting new nodes.
1 Like
Does this Timeout thing affect node joining too? I just increased to 3600, and I’m trying to join.
Sadly I only have a 10 Mb/s connection at the moment. There is some really strange stuff going on with my routers/connections.
I don’t think so, my node has been trying to join for a day…
1 Like
Uploaded this in 14 seconds :
safe cat safe://hy8ayqyybshnfyrcwp59f9g9hfq84tzq97ebdkduyrjzimjtz8b3w19d9xyqo > safenetwork.mp4
(40Mo)
Ok let’s feed this very strong baby!!
7 Likes
Just like human babies, this hungry and wonderful baby doesn’t let me sleep. It is doing OK, but I just have to get up every now and then and just… observe.
After about half an hour my giant recursive put is doing fine. There is about 200 files of about 80MB each. Memory usage seems to be maxing around 410MB, then it goes down again and upload rate is pumping between 0 - 25 Mbs, mostly around 10Mbs. This is on Windows machine.
Go, my baby, you can do it!
11 Likes
Absolutely gutted I’m away for the weekend if she is still up on Tuesday I promos ill get on this train
7 Likes
You could try uploading a lot of junk to try and fill the network faster. Create a bunch of these and then safe put away…
dd if=/dev/urandom of=rand_junk.bin bs=1G count=10
2 Likes
I kind of die a wee bit when I see us uploading any junk to just get our own nodes on. I think we should restart this one with a node size of maybe 5Gb? However the easier we make it to join for no cost then we will see nodes churning like mental things as folk join get happy and restart again. So we get further from reality. Perhaps as a stress test, it makes sense though, but there will be a limit to how fast data can replicate if the churn rate is very high.
11 Likes
Seems the harder it is to get in the better. I imagine that we would be more likely to make sure that once we are in we do all that we can to stay in, so making sure your node has a stable connection, reliable hardware no screwing about with unnecessary restarts of the node etc.
Win for the network in my opinion.
7 Likes
Yes, sounds like a good idea.
I think we’ve now passed the 20gb test absolutely brilliant 
12 Likes
Does anyone know how many splits have occurred for the latest 20GB run? It would be nice to see a run with many splits (>8)… that might be rather hard to test unless vaults are very small (<5GB)…
2^8 * 35 = 8960 nodes
3 Likes