Prevent bitcoind from printing to stdout

[achow101]

Fixes #497

bitcoind now prints to console by default when -daemon is not set. Since we do not use -daemon, the output to stdout produced by bitcoind is getting caught somewhere and causing bitcoind to block. This PR sets -noprinttoconsole when we start bitcoind to suppress this behavior.

This change should be backport-able to master if necessary. Note that this change to bitcoind will be present in 0.17.0 and is not in any current release.

[droark]

utACK

Looks good at a glance. Will test later.

Personally, I wouldn't mind seeing a 0.96.5 release, just in case people wait a bit to upgrade to 0.97 (which will have a lot of under-the-hood changes). I just think it'll save us a lot of grief in the long run. I think there's a lockbox bug that also needs to be fixed? Those two are the only things I can think of offhand that I'd want to see in 0.96.5.

[goatpig]

there are a few minor bugs pilling up that could do with a .5 indeed

[droark]

tACK

Compiled the version of Core that the OP had been using and modified Armory 0.96.4 to include this patch. I then ran an auto-managed bitcoind on Linux and confirmed that everything worked properly.

As a nit, I'd like to see a fixup where the documentation is updated. Users who run bitcoind manually (e.g., me) ought to know about the changes in 0.17. Maybe add something like howtorunbitcoinmanually.md? It doesn't need to be long. It just needs the basics. This would also be a good reference for people who end up having to stop using auto-management (which, honestly, I'm kinda thinking we might want to kill one day anyway).

[achow101]

As a nit, I'd like to see a fixup where the documentation is updated. Users who run bitcoind manually (e.g., me) ought to know about the changes in 0.17

The changes to Core, including this one, (usually) don't effect manual bitcoind users.

[achow101]

Was this missed for .5?

[goatpig]

Apparently. This is a PR to dev, which 0.97. 0.96.5 was developed in testing.