You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jun 11, 2024. It is now read-only.
Postgresql default max number of connection is 100, so the SDK should comply with the max number of connection
Actual behavior
Currently max connection is 95 * 3 modules, which is 275.
Therefore, for some case, it will hit the max number of connection.
07:11:24.880Z ERROR lisk-framework: remaining connection slots are reserved for non-replication superuser connections (err.code=53300)
error: remaining connection slots are reserved for non-replication superuser connections
at Connection.parseE (/Users/x/xxx/node_modules/pg/lib/connection.js:554:11)
at Connection.parseMessage (/Users/x/xxx/node_modules/pg/lib/connection.js:379:19)
at Socket.<anonymous> (/Users/x/xxx/node_modules/pg/lib/connection.js:119:22)
at Socket.emit (events.js:189:13)
at Socket.EventEmitter.emit (domain.js:441:20)
at addChunk (_stream_readable.js:284:12)
at readableAddChunk (_stream_readable.js:265:11)
at Socket.Readable.push (_stream_readable.js:220:10)
at TCP.onStreamRead [as onread] (internal/stream_base_commons.js:94:17)
07:11:24.883Z INFO lisk-framework: Cleanup controller...
07:11:24.883Z ERROR lisk-framework: Reason: Error: DB Connection Error
Steps to reproduce
Starting the application with default setup
Which version(s) does this affect? (Environment, OS, etc...)
2.0-
The text was updated successfully, but these errors were encountered:
I had a same error in v2.2.0, but there was no trouble in v2.1.0.
There were three instance at v2.1.0. however from v2.2.0, there are four since controller has own connection.
It's just a guess but, is there a possibility that the change might have caused it?
(I tried to do initial DB creation with v2.1.0, then run v2.2.0. In that case, no error occurred.
also I tried to halt http_api module connections, then that worked fine.)
Sorry, that was my misunderstanding.
I suppose the problem has nothing to do with the number of instance, but a timing of loadBlockChain execution .
In v2.1.0, API module and Network module are loaded before loading Blockchain.
However in v2.2.0, it seems loading those modules are called after loading Blockchain.
so in that case, if max setting is 95, full of connections are used by Chain module.
I agree, that default value was carry forward when every thing was using single db object. Now since load been splitted between modules so default max connection can be reduced.
Expected behavior
Postgresql default max number of connection is 100, so the SDK should comply with the max number of connection
Actual behavior
Currently max connection is 95 * 3 modules, which is 275.
Therefore, for some case, it will hit the max number of connection.
Steps to reproduce
Starting the application with default setup
Which version(s) does this affect? (Environment, OS, etc...)
2.0-
The text was updated successfully, but these errors were encountered: