quorum | permissioned implementation of Ethereum supporting data | Blockchain library

I've got two EC2 instances running Artemis in the master-slave replication setup and I always perform rolling upgrade by shutting down the slave, upgrading it and starting it again. Then I do the same with the master. This no longer works for me while upgrading from 2.17 to 2.18. After upgrading one of the nodes I always get this message on that node:

I am trying to run the HBase shell. The shell get starts but when I put any command it gives me this error:

I use MassTransit and RabbitMQ quorum queue for application integration. I send message using

I was reading through docs and encountered this part:

So let's say I have Strong consistency set as default (for writes). And I want to override it for some queries.
It would matter only if I select between Strong, Session and Eventual consistencies, because these are the ones with distinct Quorum Reads.
Changing between Strong and Bounded Staleness or Consistent Prefix and Eventual wouldn't matter, because it's only for reads - writes are still configured with Strong consistency.

Am I correct?

Edit: Though the sheer answer of @Mark doesn't satisfy me, it's complete for me with the comments, so I'm marking it as a solution. To simplify, the answer is that effectively there's actually only 2 options in query relaxation: Strong and Consistent Prefix.

In cosmos DB, one replica-set consists of 4 replicas. It has one leader and 3 followers (including a forwarder).

I assume that all updates go through the leader first and the leader is the first one to be updated when updates come in the replica-set.

If my assumption is correct, a reader does not have to read many replicas. It only needs to check the leaders in each replica-set to figure out latest updates.

For example, Cosmos DB documentation says that in case of eventual and consistent prefix consistency models, quorum reads is accessing one single replica. As far as I understand it, the single replica can be any replica in a region. But, if a reader needs to read a update from a replica, does it have to be a leader replica?

I have a high availability cluster with two nodes, with a resource for drbd, a virtual IP and the mariaDB files shared on the drbd partition.

Everything seems to work OK, but drbd is not syncing the latest files I have created, even though drbd status tells me they are UpToDate.

I have a git repository I am attempting to move to github from bitbucket. This is something I have done before without much difficulty, both using the github importer tool and manually sending up a repo. In this particular case, however, my repository seems to have an issue in its history, which causes github to fail. Notably, when I run git fsck, I get:

Consider that we are running Raft on 3 machines: A, B, C and let A be the leader. There is a network partition that splits C, from A, B. Call the current term t. A and B remain on term 2, with no additional messages besides periodic heartbeats. At this time, C enters candidate state and increments term to 3, votes for itself, times out, and repeats. After say 10 cycles, the network partition is resolved. Now the state is A[2], B[2], C[12]; C will reject AppendEntries RPC from A as the term 2 is less than its current term, 10; C cannot assemble a quorum and will continue to run the leader election protocol as a candidate, and become increasingly more divergent from the current term value of A and B.

The question is then, how does Raft (or Raft-derived implementations) handle this issue? Some thoughts I had included:

1. Such a situation is an availability issue, rather than a safety violation. Ignore and let human operators handle by killing or resetting C
2. Exponential backoff to decrease the divergence of C per elections
3. Have C use lastApplied instead of currentTerm as the basis for rejecting or accepting the AppendEntries RPC. That is, we trust the log as the source of truth for terms, rather than currentTerm value. This is already used to ensure that C would not win as per the Election Restriction, however the paper seems to indicate that this "up-to-date" property is a grounds for not voting for C, but is not grounds for C to acquiesce and reset to a follower.

Note: terminology as per In Search of an Understandable Consensus Algorithm (Extended Version)

I have a Google sheets with 65 lines set up for data. In every row some of the cells use dropdown list and others use an if formulas based on the selections of the dropdown choices. The final cell (CQ), through the use of an if statement in the cell, populates as follows;

A) Returns "" if all other calculated cells in the row are blank B) Returns "DATA INCOMPLETE" if all the other calculated cells in the row do not meet specific criteria C) Returns "COMPLETED" if all calculated cells meet the specific criteria.

My sheet currently has data in 10 lines however it is trying to generate 65 PDF files. In functionNMPRBulkPDF it is suppose to stop when it reaches a blank row however that is not happening. I believe this is occurring because it thinks a row is not blank because either:

A) it sees the dropdown box as having data or B) it thinks the if formula is actual data even if the return is blank ("")

I am not sure which.

What I would like to be able to do is to use last cell in the row (CQ) to determine if the row should have a PDF file created for it. If the entry is "" or "COMPLETED" I would like to skip the row and go on to the next one. Or in other words only print the ones with "DATA INCOMPLETE" in that last cell (CQ)

I have this working well other than this issue so any help will be highly appreciated. Please have patience with me as this is my first foray into Google Script and I am totally self taught.

Here is my script