Big
thanks to @andybets for finding THIS POST and making note of it in @benjojo's question directed at Stinc about what the hell is going on with all these RPC nodes going down.
It seems that I was not exaggerating, and there was a reason why I was unable to get a steemd RPC node to sync, either on a 10 core, 50Gb memory 1.2Tb SATA3 SSD VPS, or, even, on my own PC, which is in most respects even faster: ryzen7 1700, 32gb of DDR4 2400 memory, and a 256Gb Samsung NVMe SSD. 6 months ago this machine would have ripped through the replay in a couple of hours. The Xeon VPS could do it in about 8-9 hours. I finally gave up after watching it spend over 3 days on the VPS and the same on my machine.
Nobody who should have known better, was telling the truth. They all said it must be just me, because they are all lying at the same time. No, because I'm the only one having problems. Ok, if that's true, then what does this mean:
Call for help
Full rpc steemd nodes are a challenge to run. Currently, they require expensive servers with 128GB of RAM, and regular baby-sitting. To my knowledge, there are no open solutions for managing steemd clusters . Such software would be much appreciated, not only by developers, but also exchanges (currently, both Bitfinex and Poloniex have STEEM&SBD deposits/withdrawals frozen, as they are failing to tame resource hungry steemd [2]).
https://steemit.com/steem/@furion/updates-on-steem-python-steemdata-and-the-node-situation
I don't know what to say, guys. You know I had been saying right back in November last year that the rate of growth of the resource use of steemd was going to be a problem within 6 months. Here we are, 8 months later, and if nothing is done THIS MONTH, at the latest, there will be no value in anyone's steem, because nobody will be able to buy, or sell it.
An apology would be nice, but I don't expect such professional liars as yourselves to ever come clean about this.
I am just glad to finally confirm that it isn't just me. Because I was getting little help or corroboration from anyone. Real friendly bunch you are, especially the witnesses. Gandalf (@gtg) is the only one who seems to be able to make steemd obey him. That's probably because he's running it on a cluster of 256Gb memory, NVMe dual raid arrays, on a multi gigabit conneection, and it is probably costing him over $500/month to do it. But don't expect him to publish his real costs, because, if you can't run it, it must be because you are incompetent.