Github looking extra spicy!
What make you think like that?
Just from what was said in the last update about a new test network being available again at some point and the amount of things going on in there. I could just be jumping to conclusions and being overly optimistic.
Kinda sounds like the opposite. A ton of changes doesn’t really sound testnet ready ![]()
Its a LOT readier <— is that even a word? than it would had the changes not been made…
Either way, all good, the team will get there ![]()
Or testnet is close!
Just fired up a baby-fleming with the latest patch Churn on gha infra by joshuef · Pull Request #1849 · maidsafe/safe_network · GitHub
Stores large files rather quickly
2Gb in well under a minute ![]()
willie@gagarin:~/projects/maidsafe/safe_network$ time safe files put ~/2gbflash.img
FilesContainer created at: "safe://hyryyryzm8ue74rtbzbm7u9qza6biyqrkega7xy48u18eycsix67bhfmjsanra?v=hmmuirwnxzmfcz8fksp58cjeu8efeby4zegqhqcdg66hubkuhxczo"
+---+---------------------------+--------------------------------------------------------------------+
| + | /home/willie/2gbflash.img | safe://hyfeynyjzkducq5wb98yqxgegzcbtphh87ktm9dszc1p9w3y3uiuojjta8o |
+---+---------------------------+--------------------------------------------------------------------+
real 0m29.757s
user 0m45.250s
sys 0m2.722s
Ah that’s nice. We’ve not been poking the big-file-bear yet. But that’s great to see. How was your local mem/cpu load during this?
It took a bit of a pounding…
I killed it to run again with a larger no of nodes but then the phone went…
I’ll try again shortly with 30 nodes - unless you want different test?
willie@gagarin:~/projects/maidsafe/safe_network$ RUST_LOG=sn_node=trace safe node run-baby-fleming --nodes 30
even faster
nice one @joshuef
![]()
willie@gagarin:~/projects/maidsafe/safe_network$ time safe files put ~/2gbflash.img
FilesContainer created at: "safe://hyryyry19saud6bu3mygopq7rc4kawrq1p1sdwb4szs9o9xehwtds6h4euenra?v=hk9be5ngy8esh8j48cs9dp78i7wotywcj3c6nyfjay1zgxpbzezdy"
+---+---------------------------+--------------------------------------------------------------------+
| + | /home/willie/2gbflash.img | safe://hyfeynykq7edm8djnpnk9ang4g9au1gemt7zmt5i4gu7imad3a48ek1w3fe |
+---+---------------------------+--------------------------------------------------------------------+
real 0m22.629s
user 0m44.434s
sys 0m2.354s
but a 2.3Gb iso file took nearly 6mins…
willie@gagarin:~/projects/maidsafe/safe_network$ time safe files put ~/7600.16385.090713-1255_x86fre_enterprise_en-us_EVAL_Eval_Enterprise-GRMCENEVAL_EN_DVD.iso
FilesContainer created at: "safe://hyryyry1rarnkm8pb9ff7mjyb39e3xe54bhoxur3owheb1b935qjq97agocnra?v=hobpxmnmpfict6fx9t1dmcxzj1e3433418miri1gstxz6npgrtguo"
+---+--------------------------------------------------------------------------------------------------------+--------------------------------------------------------------------+
| + | /home/willie/7600.16385.090713-1255_x86fre_enterprise_en-us_EVAL_Eval_Enterprise-GRMCENEVAL_EN_DVD.iso | safe://hyfeynypeb75dagjc8whap6qjaqn9rub8jzn8mxiquc4zsegmiw7nbgcupr |
+---+--------------------------------------------------------------------------------------------------------+--------------------------------------------------------------------+
real 5m43.219s
user 3m28.111s
sys 0m38.580s
more testing required …
At the moment, nothing specific required. I was just curious of the overall state of affairs there. I’m hopeful we’ll get deeper into larger files “soon”
Same here 2gb files are going up nicely this has made my day ![]()
Results on bare metal i5 with 4Gb ram
cat nohup.out
FilesContainer created at: "safe://hyryyryuw176jc554cgubuerk1jrocfujmimky4fexuah91tpzk4biireyhnra?v=h5o811cfy97jrrqcek36spbxm9pbh5kwnx5u8q6d36wmii5h9ny3o"
+---+------------+--------------------------------------------------------------------+
| + | ./crow.mkv | safe://hyryyyyccadu49upygdfzicz9fpsxune1m9fk9gr3ssjcdmk8mo9nrxymaa |
+---+------------+--------------------------------------------------------------------+
185.66user 44.06system 9:33.65elapsed 40%CPU (0avgtext+0avgdata 5207616maxresident)k
4539800inputs+40outputs (112major+2109599minor)pagefaults 0swaps
BegBlag in a bawhair over 2.5 seconds!!!
willie@gagarin:~/projects/maidsafe/safe_network$ time safe files put ~/BegBlaga.mp3
FilesContainer created at: "safe://hyryyryz4skn9fhfzmnwi7iqpaok596bte5tudiypxbh6ewuphd65zfki7ynra?v=hufp7y8arojmxem573z88u45zzktk8mdxc6eds8uugrh7tj3ix1fy"
+---+---------------------------+--------------------------------------------------------------------+
| + | /home/willie/BegBlaga.mp3 | safe://hygyy1ymx4y4rbbociyr9ayrrbe9aw76mf586qx4u9jo5r45z3m338i1x4y |
+---+---------------------------+--------------------------------------------------------------------+
real 0m2.540s
user 0m1.618s
sys 0m0.366s
Is this a record?? ← No a record is a round black thing with a hole in the middle ← dad joke
I’m here all week, try the veal.
However…
Early this morning I was posting about a 2Gb file saved in ~30 secs but noted a 2.3Gb took ~6mins - Thinking about it, that was not so odd. The 2Gb file was a saved image of an old USB stick and on poking a bit I found it 60-70% open space plus a verrrry old Ubuntu distro and the 2.3GB was a rammed-full, compressed to the gills Windows iso - kinda understandable – sorry for getting youse all excited
This PR is in preparations for the next testnet. We’ll try for a single very large section with joins enabled.
This is implemented by increasing the recommended section size to a very large constant (10000).
OH HELL YEAH! nodes from home? infinite? I am gonna join with 10000 nodes myself! ![]()
Yea… let’s get our lines together and break this before it even moves! ![]()
we can all not go mental with this one
It’s great if folk can join, but idealy imagine it costs resources to join and the network will not allow unlimited joining like this in real life.
So if folk could run a node but really try and keep it online, then it would be really helpful. If folk put nodes off and on rapidly, it will not tell us much right now. So a wee appeal to treat it like the real thing and start a few nodes and keep them going as normal. Switch them off by all means but not off/on/off/on/off/on type patterns ![]()
![]()
![]()
![]()
Bug found and is being fixed. Expect some more releases ![]()
