Thanks so much to the entire Maidsafe team for all of your hard work!
I’m glad we have such a great community to provide money, enthusiasm, and creative support for the whole project.
And now everyone can work on a new testnet too yet.
Thanks so much to the entire Maidsafe team for all of your hard work!
I’m glad we have such a great community to provide money, enthusiasm, and creative support for the whole project.
And now everyone can work on a new testnet too yet.
I see two of them.
I see two:
Network sections information for default network:
Read from: /home/topi/.safe/network_contacts/default
Genesis Key: PublicKey(0d58..9171)
Sections:
Prefix '0'
----------------------------------
Section key: PublicKey(1687..95fa)
Section keys chain: [(PublicKey(0d58..9171), 18446744073709551615), (PublicKey(05ac..8bdd), 7), (PublicKey(1821..9904), 0), (PublicKey(1687..95fa), 5), (PublicKey(042e..c837), 1), (PublicKey(0dca..9790), 4), (PublicKey(1112..7345), 2), (PublicKey(14b1..b3b1), 6)]
Elders:
| XorName | Age | Address |
| 07e23a.. | 5 | 64.227.42.22:49976 |
| 10cbf7.. | 5 | 165.22.119.224:42907 |
| 178d5d.. | 5 | 188.166.151.208:42430 |
| 2f6116.. | 5 | 65.109.139.177:41794 |
| 482043.. | 5 | 161.35.162.137:49074 |
| 673348.. | 5 | 138.68.162.81:46787 |
| 738018.. | 255 | 134.209.16.34:12000 |
Prefix '1'
----------------------------------
Section key: PublicKey(0757..36e3)
Section keys chain: [(PublicKey(0d58..9171), 18446744073709551615), (PublicKey(05ac..8bdd), 7), (PublicKey(1821..9904), 0), (PublicKey(042e..c837), 1), (PublicKey(0757..36e3), 5), (PublicKey(0dca..9790), 3), (PublicKey(1112..7345), 2), (PublicKey(14b1..b3b1), 6)]
Elders:
| XorName | Age | Address |
| 970591.. | 5 | 209.97.183.9:49975 |
| 9d6557.. | 5 | 142.93.39.249:49476 |
| 9f31fa.. | 5 | 161.35.44.21:58269 |
| b4d09d.. | 5 | 209.97.179.216:41622 |
| bf3f89.. | 5 | 161.35.36.63:39294 |
| ccb664.. | 5 | 161.35.44.169:43499 |
| e6d9a9.. | 5 | 161.35.172.200:39601 |
Listen guys, I’m very sorry, but I have messed up here.
I didn’t upload the correct node version to main2
. If you can just bear with me here, I will get another network up.
Apologies for the inconvenience.
@ Neik At least you got home, son#1 is stuck out on the Ninian Central with a huuuuuge backlog of chopper transfers to Sumburgh - which has no power and is about to go on strike anyway so no onward flights to ABZ Could be 3 in a row - no home for Xmas for the boy.
Shit happens @chriso
Chill
EDIT: I’m killing everything and will join in again when you are happy with what you have up. I have to go cook anyway.
Iv got to bail for tonight ill check in tomorrow
Looks like I missed all the action!
Ready to go once it’s up again, but take your time to make sure everything’s ok!
I would suggest a separate post for the testnet and leave this thread only for the update related replies
I do not understand why the test parameters were changed before a successful result was obtained.
Last week there was a testnet limited to upload/download that did “pretty well” for a while before falling over. Now it is believed that bug has been analyzed and fixed.
So to me, the proper thing would be to re-do that test with the same parameters, meaning upload/download only, no new peers. Keep doing that until such time as the network can fill up entirely, repeatedly, without errors.
Then, and not before, move on to another test with new parameters, such as nodes joining and leaving, with no uploads. When that is successful, then nodes joining and leaving with uploads. When that is successful, then nodes joining and leaving with downloads. When that is successful, then nodes joining and leaving with uploads/downloads.
Anyway, that’s the type of approach I would recommend, if asked.
Yes, I think that would be good practice in general. Then it would also be easy to launch new thread with new OP and no need to edit the OP of a current thread.
No it won’t, it is new clean and empty.
I think it was alive for enough time to make conclusion that it will not fail with register limitation in place. I agree with it.
So next parameters were needed. But selection of decentralized variety was too bold on my opinion.
However nothing too wrong happened. Now something average between these two extremes can be selected.
Yeah, that’s valid, maybe this was too to ambitious, but like @Vort mentioned last testnet looked very stable. With the rate of 1 testnet / week, the risk was that if we just re-did the last testnet with the bug-fixes in place, we’d miss out on a weeks worth of learnings.
So optimizing for rate of learning rather than being 100% sure we’ve squeezed out everything from one experiment.
We’re pulling the logs from today’s original failed testnet now. We’ll post them here as well for some light reading material for folks to go over with their morning coffee
I noticed some of the instructions above point to two different network contacts URLs between the upload/downloading section and the storage node instructions section:
safe networks add main2 https://sn-node.s3.eu-west-2.amazonaws.com/testnet_tool/main2/network-contacts
safe networks add main https://sn-node.s3.eu-west-2.amazonaws.com/testnet_tool/main/network-contacts && safe networks switch main
For trying out the testnet as a storage node operator, just to verify, are we to follow the URL with ‘main’ for the network contacts, or will a fresh deploy of the testnet have a different network-contacts URL entirely for storage node operators?
Using the client or node need one config,
Today we had two testnets thats why main and main2
No, the network-contacts URL should be the same for all participants, no matter their role.
main
was pulled due to some unspecified as yet problem that meant very few could join, main2
was an unfortunate error and the wrong binary got used.
@Shu says that main
and main2
should not exists in instructions at the same time.
True from @shu
One thread, with major changes halfway through, not ideal.
As said somewhere above, we should have a thread for the Thursday Update and another for the current testnet.
one additional point:
Beyond just “valid test procedures” I think it will do the community some good to witness a testnet that is able to take a beating and accomplish the objective. and then another. and another.
If today’s does that, great… but changing the params worries me.