Multiple try/catch+zombie fixes

[hlolli]

Changes

• remove unneccecary async try-catches
• Improve randomized NODE selection
• Prevent zombie processes with proper promisified recursion

---

Rebased over #96, I'll re-rebase after that gets merged.

But the problem I'm hitting is stuff like this, and for parts which can fail, should fail. The stack gets usually lost in async functions (not in Promises) and a series of them, makes debugging very tedious.

yarn start
yarn run v1.22.10
$ npm run migrate:latest && node dist/src/Gateway.js

> @arweave/gateway@1.0.0 migrate:latest
> knex migrate:latest

Requiring external module ts-node/register
Already up to date
info: [app] started on http://localhost:3000
info: [database] starting sync, parallelization is set to 2
node:internal/process/promises:246
          triggerUncaughtException(err, true /* fromPromise */);
          ^

Error: connect ECONNREFUSED 163.47.11.64:1984
    at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1133:16) {
  errno: -61,
  code: 'ECONNREFUSED',
  syscall: 'connect',
  address: '163.47.11.64',
  port: 1984,
  response: undefined
}
error Command failed with exit code 1.
info Visit https://yarnpkg.com/en/docs/cli/run for documentation about this command.

With this change I'm able to get somewhat started

yarn start
yarn run v1.22.10
$ npm run migrate:latest && node dist/src/Gateway.js

> @arweave/gateway@1.0.0 migrate:latest
> knex migrate:latest

Requiring external module ts-node/register
Already up to date
info: [app] started on http://localhost:3000
info: [database] starting sync, parallelization is set to 2
info: [database] syncing from block 0 to 730868
1/730868 blocks synced [                                                                                                    ] 0% 0.0sinfo: [snapshot] could not retrieve block at height 0, retrying
5/730868 blocks synced [                                                                                               ] 0% 343944.1sinfo: [snapshot] could not retrieve block at height 5, retrying
info: [snapshot] could not retrieve block at height 5, retrying

[TheLoneRonin]

Hey @hlolli,

I think the idea you're going for here on this PR is good and makes sense.
However, this codebase is going to be completely rehauled this coming week.

Are you available to chat about this source and Vartex sometime soon?

You can actually review: https://github.com/ArweaveTeam/gateway/tree/v1.0.1
To see some of the changes coming through.

Thanks.

[TheLoneRonin]

I'm also probably going to migrate the improvements to try/catch queries to that branch. Unless you want to open a PR to v1.0.1 with that specifically.

Thank you :)

[hlolli]

@TheLoneRonin

The issue at the time was that errors were taking place and they weren't visible. I have since discovered the culprit, which is that streams need to be handled with .on("error", cb) since the callsite where the error is thrown isn't always at the consumer side where we are trying to catch the error. Irritating when this caveat isn't known.

Let's chat soon!