Skip to content

Conversation

@instagibbs
Copy link
Contributor

@instagibbs instagibbs commented May 6, 2019

First, only start scheduling checks if the initial check on load succeeds. This initial check is a sanity-check that makes sure all daemons are spun up and communicating correctly. Second, for the initial check only we want to spin-wait for the mainchaind to finish warming up, otherwise elementsd may fail rather than wait a couple seconds for bitcoind to start up, for instance, which can occur when automating the processes.

Previously we were spin-waiting for all instances of MainchainRPCCheck, which means if your bitcoind falls over mid-run for whatever reason and is in warmup state for a long time(say, updating UTXO db), these checks will pile up on each other, spamming your bitcoind with useless RPC calls. Post-init it's completely fine for the call to simply fail and try again in N minutes.

resolves #608

@stevenroose
Copy link
Contributor

ACK ad43f9c

@stevenroose stevenroose merged commit ad43f9c into ElementsProject:master May 8, 2019
stevenroose added a commit that referenced this pull request May 8, 2019
ad43f9c Only spin-wait MainchainRPCCheck on initial loading with warmup code (Gregory Sanders)
9d4097d Only schedule mainchain RPC checks after initial check succeeds (Gregory Sanders)

Pull request description:

  First, only start scheduling checks if the initial check on load succeeds. This initial check is a sanity-check that makes sure all daemons are spun up and communicating correctly. Second, for the initial check only we want to spin-wait for the mainchaind to finish warming up, otherwise elementsd may fail rather than wait a couple seconds for bitcoind to start up, for instance, which can occur when automating the processes.

  Previously we were spin-waiting for all instances of MainchainRPCCheck, which means if your bitcoind falls over mid-run for whatever reason and is in warmup state for a long time(say, updating UTXO db), these checks will pile up on each other, spamming your bitcoind with useless RPC calls. Post-init it's completely fine for the call to simply fail and try again in N minutes.

  resolves #608

Tree-SHA512: 6a02f1b150a8631e36da63a33f9775fbb2cf17aa749e4895b6a947ba605e00ce9fda8349bf7a556e01d9fccaece7387914fb6bb78d7b47c9f272bbcf7f3ff919
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

MainchainRPCCheck shouldn't spin wait on non-init call

2 participants