Closed tvolk131 closed 5 days ago
Will revert if it is causing issues. :shrug: :rofl:
MQ found some LN test problems, worth looking into before kicking it back into the MQ imo.
Hmmm most of the CI errors are some form of the following out-of-disk-space:
fedimint-workspace-lcov-ci> 2024-05-22T22:56:47.489529Z ERROR connect_fed{payload=ConnectFedPayload { invite_code: "fed11qgqpw9thwvaz7te3xgmjuvpwxqhrzw33xymnsdf0qqqjqmlee0af0aap6fyvmslxhn94k9tsms6wtel7sq32z747qdhra0uuej7vxh" }}: ln_gateway::rpc::rpc_server: error=Failed to open the database: Error opening rocksdb: IO error: No space left on device: While appending to file: /build/test-14034218261564233250GBHOqh/gateway-16131950401517601773/6ff9cbfa97f7a1d248cdc3e6bccb5b1570dc34e5e7fe8022a17abe036e3ebf9c.db/000000.dbtmp: No space left on device
Whatever's causing this flakiness appears to be outside of the control of the code being changed in this PR.
I audited the code for
open_channel()
to ensure thatopen_channel_between_gateways()
matches it 1-to-1. I then re-enabledopen_channel_between_gateways()
for devimint (which was disabled here due to being slow and flaky). I measured the startup time forjust mprocs
10+ times on my local machine before and after the change and with the changes in this PR it never failed and the time difference was within margin of error.