Actually this isn't accurate. Unless testnet has a mixture of v20 and v19 nodes (replicating production state), this problem would not have been caught. Remember the bug was that one of the HF20 consensus rules became active by accident before the scheduled date.
That's only good when it shows us that "normal everyday data" seems to work.
I was hoping we deliberately send "bad and tainted" data to the nodes, and eonwarped makes a good point.. Having a mixture of v20 and v19 nodes on testnet is best, especially to duplicate the transition period.
On a testnet, you really do need to go wild, and test every possible scenario when possible.
It's impossible to do that... but if you can get "close to impossible", you'll get "close to seeing if the code is nearly perfect" too. :)
Thank you for this detailed update Drakos. We have to trust in our witnesses to make the best decisions regarding what happens with the underlying code.
I don't know if this is publicly available information or not, but I would be interested to know which of our top 20 witnesses (if any) were NOT a part of the effort to get things back up and running September 17th. We all know some witnesses are more attentive than others, I believe those who are not part of keeping things up and running should be held accountable in the witness voting.
Thank you for having alarms set and making your witness nodes a priority. It does not go unnoticed and is appreciated!!!
It's a very valid question. As much as I would love to give individual credit to those who put efforts on that day, I think it's best that they write about their experience themselves and let their supporters be the judge. The timezone, work schedule, family matters or other engagements, they all affect the availability of each witness, which doesn't make them any lesser than others. Also, an event such as that, however critical it was, should not be the only metric in evaluating the overall community/technical input of a witness. The bottom line, each witness contributes in their own ways whenever humanly possible.
There are certain features I really want from HF20...and other ones I'm not too keen on...but, even if it has seemed like forever waiting for it, perhaps some extra time should be taken to evaluate it...if it's needed.
It's the same v20 that we are familiar with just with some fixes that we are really happy to have before HF will be triggered.
I think it's a little soon to fork. Would like to see more testing done on HF20 code first.
Let's throw more "bad and tainted" data at it on testnet... one more time. If we had done that... this last outage would not have happened.
Actually this isn't accurate. Unless testnet has a mixture of v20 and v19 nodes (replicating production state), this problem would not have been caught. Remember the bug was that one of the HF20 consensus rules became active by accident before the scheduled date.
The testnet is actually getting a copy of the transactions that are getting submitted to the mainnet automatically transmitted.
That's only good when it shows us that "normal everyday data" seems to work.
I was hoping we deliberately send "bad and tainted" data to the nodes, and eonwarped makes a good point.. Having a mixture of v20 and v19 nodes on testnet is best, especially to duplicate the transition period.
On a testnet, you really do need to go wild, and test every possible scenario when possible.
It's impossible to do that... but if you can get "close to impossible", you'll get "close to seeing if the code is nearly perfect" too. :)
Thank you for this detailed update Drakos. We have to trust in our witnesses to make the best decisions regarding what happens with the underlying code.
I don't know if this is publicly available information or not, but I would be interested to know which of our top 20 witnesses (if any) were NOT a part of the effort to get things back up and running September 17th. We all know some witnesses are more attentive than others, I believe those who are not part of keeping things up and running should be held accountable in the witness voting.
Thank you for having alarms set and making your witness nodes a priority. It does not go unnoticed and is appreciated!!!
It's a very valid question. As much as I would love to give individual credit to those who put efforts on that day, I think it's best that they write about their experience themselves and let their supporters be the judge. The timezone, work schedule, family matters or other engagements, they all affect the availability of each witness, which doesn't make them any lesser than others. Also, an event such as that, however critical it was, should not be the only metric in evaluating the overall community/technical input of a witness. The bottom line, each witness contributes in their own ways whenever humanly possible.
Thanks for being an awesome pirate witness!
Thank you for the update ! Take care @drakos
Posted using Partiko iOS
Thanks for the hardwork. Being a witness is not easy and you are surely deserve to be top 20.
I'd vote for a delay. Best to get things right than make a mess.
Is there a list of what each witness favours? Perhaps we can influence a delay by changing our witness votes.
Looks like it's going to happen, unless a critical bug happens or is discovered within the next 24h.
https://steemian.info/witnesses
There are certain features I really want from HF20...and other ones I'm not too keen on...but, even if it has seemed like forever waiting for it, perhaps some extra time should be taken to evaluate it...if it's needed.
FORK, but closer to the day of
Thanks for info @drakos, resteemed
😎👍👍