Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
removed vote for OL, as they failed to maintain proper price feeds.Thank you for the vote of confidence. I hope OL will get everything corrected soon and get back to active witnessing though.
added @sahkan witness instead
We will likely do conversion from BTS->USD->CNY to attain the feed price, but I'm not sure if the witnesses have this all figured out yet.The price feed will depend on Bitshares liquidity vs CNY. If there is a reliable exchange that has BTS:CNY pair we will use that, otherwise we will have to calculate the fair value of BTS in CNY based on the current USD:CNY pair.
wss://node.testnet.bitshares.eu
When the blockchain got stuck a few months ago I remember Dan Larimer saying something about some data might get lost because of corrupted memory. I'm also aware that the full public nodes are using more and more ram every day.
There's a new memory technology from intel called optane which can be used as either RAM or storage is also non volatile in that data isn't lost when it loses power like ram does.
What do you techies make of it?
https://www.engadget.com/2017/09/07/intel-optane-explainer/
Looks right, I voted for your witness. I checked on my witness and it was stuck on"I found the issue, will fix it and have it back online today.Code: [Select]762234ms th_a application.cpp:523 handle_block ] Got block: #10791565 time: 2017-09-07T01:12:42 latency: 234 ms from: init3 irreversible: 10791555 (-10)I have to restart and resync.
2017-09-07T03:39:51 p2p:accept_loop accept_loop ] 0: 0 node.cpp:4192
2017-09-07T03:39:51 p2p:accept_loop destroy_connection ] in destroy_connection() for message_oriented_connection.cpp:286
2017-09-07T03:39:51 p2p:accept_loop destroy_connection ] in destroy_connection() for message_oriented_connection.cpp:286
2017-09-07T03:39:51 p2p:accept_loop accept_loop ] 0 exception: unspecified
Too many open files
{"message":"Too many open files"}
asio asio.cpp:60 error_handler
Unable to accept connection on socket.
{}
Looks right, I voted for your witness. I checked on my witness and it was stuck on"I found the issue, will fix it and have it back online today.Code: [Select]762234ms th_a application.cpp:523 handle_block ] Got block: #10791565 time: 2017-09-07T01:12:42 latency: 234 ms from: init3 irreversible: 10791555 (-10)I have to restart and resync.
762234ms th_a application.cpp:523 handle_block ] Got block: #10791565 time: 2017-09-07T01:12:42 latency: 234 ms from: init3 irreversible: 10791555 (-10)I have to restart and resync.
As you might have noticed already, the seed node broke and the network for some reason split into two parts with me signing blocks on my majority "fork". This is interesting but I expected this to happen as we basically have a pretty weak p2p network. Those of you that are running the network, please resync the whole chain if you get errors while pushing blocks.
The seed node should be up again any moment.
To prevent this from happening again, I would like to request more SEED nodes to be added to the `testnet` code base.
Please consider setting up a seed node on the testnet (by opening up the p2p-node setting to 0.0.0.0:PORT) and send a pull request for this part of the code
https://github.com/bitshares/bitshares-core/blob/testnet/libraries/app/application.cpp#L163-L166
or send a mail to
testnet@lists.bitshares.foundation
Thanks!!
Just today there is a downtime of the seed and public api nodes .. I will deal with it tomorrow and see if i might have messed up due to too little seed nodes .. Lets seeI checked on my witness, it's running and producing blocks but no other nodes are producing. I wonder if Tom's witness is up...
Hello everyone. I would like to know the hardware requirements needed to build an efficient full BitShares node.
What size are all the blockchains of BitShares and what RAM is needed to load it efficiently?
I definitely agree. Mine are on two different dedicated servers.Just a note, it's best to NOT produce blocks on a full node, for example, use it as a public API server in the same time, you may miss blocks when the API service load getting high, thus impact overall performance of the network. If you want to serve more, run another node and best on another VPS/hardware.Poloniex doesn't need to run a full node. They should run node with these values configured:Indeed, I added it at the end of the config.ini. But I want to run a full node on my wallet so I will just remove it.Code: [Select]track-account = "1.2.20067"
partial-operations = true
Then it will require around 2GB of RAM.
By the way, if you configured your node with "max-ops-per-account" correctly (then it won't be a "full node" any more), it should use around 4GB of RAM but not more than 20GB, so perhaps there is something wrong with your config file. Please note that the new parameters should be added above the first logging section, but not at the end of config.ini file.
Thank you, witness is up!I am few billion votes away from beating the inits, can you please throw some votes my way? We have only 3 testers atm and dev4btc is voted in but not producing.voted
