Hedera int with TXM fixes & config #14129
Conversation
| LogPollInterval = '10s' | ||
| MinIncomingConfirmations = 1 | ||
| #turning off the liveness detection, since the Websocket connection is not reliable, can be turned on if needed | ||
| NoNewHeadsThreshold = '0' |
There was a problem hiding this comment.
@dhaidashenko : Does this comment and setting look good?
If Hedera has a 2 second block rate, we assume heads to come regularly.
Websock not being reliable doesn't seem to imply we should set this to 0.
Also, how close are we to doing liveness checks via Polling?
There was a problem hiding this comment.
HeadTracker uses websocket connection to get new heads. If there is no new heads, TXM's confirmer won't mark txs as confirmed and will eventually get stuck. It's highly recommended to enable this check, especially if the ws connection is not reliable. It will force to create a new connection, thus we'll have a chance to auto resolve the issue.
There was a problem hiding this comment.
Since we have only 1 working RPC, we cannot mark that as out-of-sync, especially with un-reliable Websocket endpoints, we set it to zero, so it waited till auto recovered, this again will be switched back on once we get 2 more fully functioning RPC
|
|
||
|
|
||
| [NodePool] | ||
| #Disabling sync threshold check for testing |
There was a problem hiding this comment.
Seems like this was disabled for testing. Should be set it to something meaningful now?
There was a problem hiding this comment.
Yes, it's better to set it to something meaningful now. If we are running on single RPC it won't affect it's health.
There was a problem hiding this comment.
again with single RPC & unreliable sync times, setting this throws a lot of error & warning & marks it out-of-sync.
Rather than setting this to a really high number, we disabled it until we get more stable RPCs to work with
Co-authored-by: Jordan Krage <[email protected]>
|
Requires Dependencies
Resolves Dependencies