We had a very good and smooth HF26. There were no big issues or alarming issues after the HF. However, after a few hours or after a day or so, there was an error identified in the witness schedule. For some reason, the backup witnesses were not included in the schedule as how they are supposed to be included. Some of the backup witnesses expressed their thoughts on this and finally the issue was identified and a fix was made ready.
It took nearly 2 weeks to complete this and apply the fix. The fix is ready and it is again going to be another hard fork and it will be HF27. Also, this is going to be a very small code chain. Most likely it is just a one-liner change if I'm not wrong. I'm not sure if there will be any official announcements on this yet but the witnesses are aware of this and they have started upgrading their node to v1.27.0 already. The code can be found here.
https://gitlab.syncad.com/hive/hive/-/tags/v1.27.0
How did this affect the backup witnesses?
For the regular users, there was absolutely no impact at all. Only the backup witnesses had this issue. They did not produce enough blocks as how they should be based on their ranking. The scheduling was having a problem and when someone is picked, they would continuously get blocks within the same day or sometimes even within the same hour which is unusual. Some witnesses benefited because of this and some did not get even the bare minimum number of blocks.
Among the witnesses I have been voting, there are a few of them who have not produced blocks for several days. My witness @balaz was also part of this list till yesterday waiting to produce a block for nearly 8 days. But luckily yesterday I started producing blocks and I produced a decent amount of blocks in just one day. Though the number of blocks I produced yesterday did not compensate for the number of blocks I should have actually produced, I would say it was pretty decent.
Compared to the top 20 witnesses, the other backup witnesses earn less amount for running a witness node. This problem really created some heated conversations among the witnesses. From what I observed, people who are running a node under Rank 60 will have their ROI. Anyone who is running from 60-70 rank might possibly be able to get their investment back. But anyone above 70 ranks may not even get the server cost back. But I believe these 2 weeks shouldn't be a big deal and after the HF is applied, the schedule should be back to normal.
How to upgrade?
For the witnesses who are running HF26, the upgrade is very easy with a few steps. If you are running a Hive In A Box node from Privex, the commands are very simple. We have to just install and restart the node.
./run.sh install
./run.sh restart
These two commands should take care of the upgrade. People who are running a node on their own servers can follow the below commands to upgrade your witness node to the latest HF version. I was using Marky's post as my guide to upgrading to HF26. So those who are using the same guide can follow the below commands.
# Checkout the latest tag
cd hive && git fetch && git checkout tag/v1.27.0
# Grab the modules
git submodule update --init --recursive
# Create a build directory for the new version
cd ..
mkdir -p ~/build-v1.27.0 && cd ~/build-v1.27.0
# Run make
cmake -DCMAKE_BUILD_TYPE=Release -G Ninja ../hive
# compile
ninja
# Move executables to bin folder
ninja install
# Confirm version
hived --version
# The above command should give you version 1.27.0
# Now go back to your screen
screen -r
# Stop and start
hived -d /root/hive-data
The above commands should do the magic and we have to wait for a new block to produce to see the updated version number. There is no replay required for this change. The HF is scheduled to happen on 1666612800; // Monday, 24 October 2022 12:00:00 GMT. Still, 3 more days to go. The witnesses have already started upgrading their nodes. If you have any questions, you can reach out to Hive discord.
If you like what I'm doing on Hive, you can vote me as a witness with the links below.
 |
 |
 |
 |
 |
 |