Blockchain

I had a unclean shutdown less than a day ago and got this warning afterwards:

WARN [03-26|10:10:29.944] Unclean shutdown detected booted=2021-03-17T13:16:42+0000 age=1w1d20h

As it would be better to have the "age=" part closer to the shutdown @holiman suggested to update the marker like every 5 minutes

The backup storage format being defined very simple and dump, which is expected, there are a few bad cases people need to be careful with.

1. on the backup side, two backup coordinators can run at the same time and write to the storage at the same time, creating duplicate, or even worse, conflicting (if the two coordinators read different chains) backups.
2. on the "read" side, backup CLI tool

Abstract

We need an option in soltest to split all test cases into N batches and run only the i-th batch to take full advantage of CircleCI's parallelism option and make soltest/soltest_all jobs give us faster feedback.

Motivation

The way we parallelize soltest today is to have [soltest_all.sh](https://github.com/ethe

When deploying fabric in enterprise's intranet (without the ability to access internet), users maybe upload chaincode builder images and chaincode runtime images to a private docker registry which requires authentication when pulling images.
I think fabric can support this scenario by adding configuration in chaincode section of core.yaml, such as:

chaincode:
    registry:
        

• I've opened a support ticket before filing this issue. (Ticket: 521615)

Issue

The default recommended configuration for mainnet is not provided in truffle-config, so it leads to misconfiguration.

I used the old style configuration parameter, gas/gasprice, which works fine for both the local

Setup

Setup a project with sqlite using the following up.sql:

create table example (
    id integer not null primary key,
    calculated integer generated always as (id*2)
);

Versions

• Rust: 1.56.0 nightly
• Diesel: 1.4.7
• Database: Sqlite
• Operating System Linux

Feature Flags

• diesel: sqlite

Problem Description

Sqlite does

Creating a new issue based on lbryio/lbry-sdk#877

• Update ensure_directory_exists() to check that the directory is writable by the current process.
• If it's not, return a useful error
• Add tests to cover the case when
  • download directory does not exist
  • download dir exists but is not writable
  • dir exists an

Problem

The transaction simulation logic does not check whether the simulation was successful before looking at the post_simulation_accounts length. This causes inaccurate errors to be returned. Specifically, if the simulation fails and accounts are requested, the RPC response is Too many accounts provided; max {} instead of the simulation error. The number of accounts provided was corre

Background

Inspired by C-Otto/rebalance-lnd#241.

As part of pathfinding, lnd might encounter edges that have a policy specifying a zero CLTV delta (possibly because of no/outdated gossip?). Every time an edge like this is used in the pathfinding code, a warning is logged: https://github.com/lightningnetwork/lnd/blob/290b78e700021e238f7e6bdce6acc80de8d0a64f/routi

A parametrized option to set the blocksize would be useful to experiment with. As quorum is used in a private setting I think this option should be available

Issue Description

The instructions provided on failed jobs (example) don't clearly explain where to find the generated patch file. This can be a problem for anyone who is not familiar with how Github Actions pages are organized.

The general idea is:

From the details page, click on "Summary", then scroll down to the "Artifacts" secti

After X minutes of being open plus Y seconds of idleness (i.e. don't reload while user is moving mouse, etc.), reload the homepage or category pages to show the latest content.