BUT I WANT FOLDERS
And you will get folders, my little tumshie.
But just not todayā¦
NO! FOLDERS NOW!
Anyway
You have a 2c8fae27d5e93d9c7a2c883a4aae8a5f7433b17c34fc2204db90d991b95cb23c
You cant get a folder cos you didnāt finish all your PR.
Cos Iāve got a HeartOfStone.mp3 2c8fae27d5e93d9c7a2c883a4aae8a5f7433b17c34fc2204db90d991b95cb23c
Node up
Uploads working
guess.html a208ac0e6a1bd009cee9b0d41cdb798153fc4d231bde4af372fbd62cea3d1c78
Chunks seems not, but what is this total_mb_received":23.69142
Successfully stored file ācock01.jpgā to d724a3ff9b2746e042fa1e8db4c4e0ec5769097042911cc8a34197edc1342e47
Not much to add here as a non-tech guy, apart from the fact that i truly appreciate all the work all of you are putting in the testing of a network I truly believe in! Thanks
Still no chunks for my node.
Looks like something is wrong with decentralization again.
But at least network did not crashed - I can still download file, which was uploaded yesterday.
Did anyone complile arm binaries? Or would the arm-darwin build work on Ampere arm-ubuntu?
I did up- and download a 600MB file yesterday, and it still downloaded fine just a minute ago.
Now Iāll try to upload something like 1,2GB.
Small 200K image file
9274f136f58a1e1f20cd4c475250811737dd88459853fa45240d07a4ae99fa31
1.8 GB file uploaded in 180m37s
which is 1.46 Mbps upload rate
on a connection thatās 1 Gbps up
ubuntu-22.04.2-live-server-amd64.iso
565cf00c400853ce08549b66270c7640a4edab4e51b38b020f7d209a001537bb
1975971840 B * 8 / 10837 s = 1.46 Mbps
safe files download 565cf00c400853ce08549b66270c7640a4edab4e51b38b020f7d209a001537bb
took 107m46
which is 2.44 Mbps
I compiled for ARM on an AWS T4g.medium Instance. It works well and I have chunks. Not a huge quantity but there is no issue with ARM. I doubt the Mac binary would work on AWSās ARM but itās worth trying! If I get time Iāll try that as well and report.
Edit: OOps! Replied to the wrong message!
Edit: Actually, no, right message, just quoted the wrong person. Iām no good at this modern stuff.
My laptop had worn out power adapter, and I tried to change it to a new one in the middle of the 1.2GB upload. The laptop shut down immediately on unplugging the old one. It had been uploading for about an hour at that point. I restarted the upload now, letās see what it says. It might not have enough time to finish before I have to shut down again.
Letās add some pussies then
b3188b9cd9735b0fcc43ee626f89552dbeb59233fffff0eedec564b6946842d7
(Looks like Iām lucky today - that fffff
must be rare thing)
My node is running on the smallest droplet available to see how it goes.
512 MB Memory / 10 GB Disk
[2023-06-08T10:15:58.697854Z TRACE sn_logging::metrics] {"physical_cpu_threads":1,"system_cpu_usage_percent":0.3992016,"system_total_memory_mb":488.30463,"system_memory_used_mb":210.70233,"system_memory_usage_percent":43.149773,"network":{"interface_name":"eth0","bytes_received":12923,"bytes_transmitted":30768,"total_mb_received":667.36505,"total_mb_transmitted":163.15547},"process":{"cpu_usage_percent":0.1996008,"memory_used_mb":29.597696,"bytes_read":0,"bytes_written":16384,"total_mb_read":62.451714,"total_mb_written":92.45081}}
root@SAFE:/tmp/safenodedata# du -sh record_store
4.0K record_store
@qi_ma what is this total_mb_received":667.36505,"total_mb_transmitted":163.15547}
as you can see above record_store is empty.
It feels like weāre turning a corner!
Hopefully these testnets will be more accessible to less technical folk soon - would love to jump on board but its a bit too technical at the mo - would need a guide for 5 years olds
I think thatās system network activity but not sure. Iām curious about some of the metrics too - maybe take a look in the sn_metrics crate and see what is being logged.
Iām also interested in the bytes read/written values for āprocessā but havenāt looked myself yet to see what they are. Bytes written is almost* always zero but read fluctuates up and down
*I just saw bytes written briefly go up to 600,000 (and read also went up to about half that) for a few seconds, so that was probably a replication and I conclude that bytes read/written refers to process (node) network activity.
I may update the vdash
display now I understand better what the metrics mean.
thatās regarding your network
(eth0) interface during the period. will include all in/out flows across your programs.
meanwhile total_mb_read
and total_mb_written
, are only about your own safenode process usage. However it also contains logging access and transmission to network, in additonal to chunk files creation
.
Whats the general consensus so far? A success? Or still very much a work in progress? or in between ?
Thanks