Aha, well that sounds like nodes are full, without enough waiting in the wings.
Mem usage per node is still low, ~200mb on average. And honestly now that much more put atm on average. Hmmm. Gonna dive into the logs some and see what I can see.
Sounds like it’s reporting full without that being the case, though 
.
I’m unable to cat the
I am still getting this error…
willie@gagarin:~$ curl https://raw.githubusercontent.com/maidsafe/safe_network/master/resources/scripts/install.sh | bash
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 7237 100 7237 0 0 25005 0 --:--:-- --:--:-- --:--:-- 25041
=> Downloading Safe CLI package from 'https://github.com/maidsafe/safe_network/releases/download/0.58.4-0.57.1-0.50.4/sn_cli-0.50.4-x86_64-unknown-linux-musl.tar.gz'...
=> Saving to '/tmp/tmp.E8dnCJrhee/sn_cli-0.50.4-x86_64-unknown-linux-musl.tar.gz'...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 685 100 685 0 0 2255 0 --:--:-- --:--:-- --:--:-- 2260
Warning: Failed to create the file
Warning: /tmp/tmp.E8dnCJrhee/sn_cli-0.50.4-x86_64-unknown-linux-musl.tar.gz:
Warning: No such file or directory
0 9.8M 0 852 0 0 1318 0 2:10:10 --:--:-- 2:10:10 1318
curl: (23) Failure writing output to destination
=> Unpacking Safe CLI to '/home/willie/.safe/cli'...
tar (child): /tmp/tmp.E8dnCJrhee/sn_cli-0.50.4-x86_64-unknown-linux-musl.tar.gz: Cannot open: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now
chmod: cannot access '/home/willie/.safe/cli/safe': No such file or directory
=> Adding statement to '/home/willie/.bashrc' profile to have Safe CLI binary path in the $PATH
=> Profile '/home/willie/.bashrc' already contains a statement to set Safe CLI in the $PATH
I will grab the binaries manually.
Same as @Josh …
willie@gagarin:~$ time safe files put Desktop/cockerel-southside.jpeg
FilesContainer created at: "safe://hyryyryyn7e7somp39di84bzyfnbem4mm6un5kfbrqhead8kt1ixt638wfneoeuy?v=hgip1akd6y1eg5cnzxjfqayrb7yyyobuttozu3uwcoggpcnoq1aho"
+---+---------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------+
| E | Desktop/cockerel-southside.jpeg | <ClientError: Error received from the network: InsufficientAdults { prefix: Prefix(0), expected: 4, found: 1 } for cmd: MsgId(34d5..94a0)> |
+---+---------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------+
real 2m54.901s
user 0m9.381s
sys 0m0.188s
I am still waiting, not yet been accepted.
So “not enough adults” means the network is full and there are not enough new nodes to join… or do I misunderstand.
Should do aye.
It doesn’t say anything about this tbh. So the join process could have stalled.
But none of that should mean lost data or slow respones. There’s something else afoot there.
Yeh I can see there’s some error in the logic around chunk storage.
➤ Chunk holders of 617e6c(01100001).. are empty adults: {04e332(00000100)..} and full adults: {}
There should be more than one here.
(empty is misleading there. just means non-full).
Just more info as I’m going.
I’ve found where we transition from 4 chunk holders to less, and we have a spate of ProposeOffline there, seemingly due to connectivity check fails.
Which makes me think we’re cleaning those up and failing nodes when we should be retrying.
For the curious:This log file line 24413 is the last instance of 4 chunk holders, and you can see what’s afoot after that.
theissue.log.zip (571.6 KB)
I’ll just be confirming this, and then seeing if my simple fix for node conn retries I added this morning might sort this (that’s pending a merge just now, so not in the current release)
Aha, yup, a failed msg send from nodes will trigger the Cmd::HandlePeerLost, so mayybe that fix will sort it. waiting on the merge there and then will trigger out a new release
A search for empty adults shows this happening every couple of hundred lines or so
dont you need the filecontainer xor url for get?
You mean for @Sascha’s .txt? He posted it.
It was a general question… so yeah FileContainer xor url
usually people post the safe cat ‘file xor url’ > ‘file’
so you cant safe files get if thats the norm
Yes, you need it afaik, but as i mentioned i dont use it much. That will change now that I know that I can see download progress.
I have a ~700mb folder of random files being uploaded and it is there purring allong for 4 hours… anyone has success with put?
Been AFK for an hour or so - Came back in to find Im still “Retrying after 3 mins”.
Also
willie@gagarin:~$ safe files put -r ~/projects/maidsafe/olympus/
FilesContainer created at: "safe://hyryyryynt8eokije8inp7j9bzz71iy4j53oakwyijkbq87be6rk9w3jx6tuyeuy?v=huy3nchwspkx69b6kczf7t613qaqgqkhpk4xrty6hcn76mnff8pny"
|---+-----------------------------------------------------------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------|
| + | /home/willie/projects/maidsafe/olympus/index.html | safe://hy8oyryyb6zip1tkwgs5ijzksou4khbu5yophgwnuj5afno9kc31npsigntwy |
+---+-----------------------------------------------------------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------+
<ClientError: Error received from the network: InsufficientAdults { prefix: Prefix(0), expected: 4, found: 1 } for cmd: MsgId(…)> for all the actual content though I can cat te index.html itself just fine.
Aye, with every chunk there @JPL .
I’m going to bring this down, as I think it’s voted most of its adults off now. If i get another release through I’ll try and reup and we can see if that’s had any impact.
@chriso would you mind having a look at this when you have a sec? 
I see this consistently on my own desktop.
Also seen it on an AWS instance and a n00b I was helping a couple of weeks back reported similar.
All on Ubuntu20.04.
Hmm, I don’t reproduce it initially, although that was on an Ubuntu 20.10 VM. I’ll see if I can get an Ubuntu 20.04 VM setup.
The progress indicator was always there. In fact, it used to be a progress bar rather than just text, but we had to drop the use of that crate.