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
blockperf should output the details of the block propagation data to the console or, if explicitly requested by the operator, to the journal log.
By default - when blockperf is started as a systemd service - the journal log should only show that blockperf was started successfully as a service, and with which configuration it is running (networkmagic, port, mqtt target server).
In addition only errors and warnings (e.g. mqtt endpoint timeout, or incomplete parsing data for a block) should be logged.
The text was updated successfully, but these errors were encountered:
Yes, good point. Currently the way the output is generated is a bit messy.
So in a default setting, it should only write to stdout a handfull of initial configs/settings it has parsed and is going to run with. Plus anything of warning/error level that can happen during runtime.
Can you elaborate a little on what the user should be able to configure? Or how the configuration could look like? There is a --debug flag that one can provide that prints out alot of extra stuff that propably noone other then us will need for now.
But there certainly is also a middle ground which provides what you describe. But again, i am currently not having a good idea on how the user could specify that? Would it also go to stdout (thus journald), some extra file?
I would suggest to have a (default) console, and a special service mode.
If the script is started in console mode, it should show all information it currently does.
In service mode, on the other hand, when no one is paying attention, all this data should not fill the journal logs.
In this mode, the script should only output information about startup/shutdown, as well as warning and critical messages for the parsing and transfer operations, if any
blockperf should output the details of the block propagation data to the console or, if explicitly requested by the operator, to the journal log.
By default - when blockperf is started as a systemd service - the journal log should only show that blockperf was started successfully as a service, and with which configuration it is running (networkmagic, port, mqtt target server).
In addition only errors and warnings (e.g. mqtt endpoint timeout, or incomplete parsing data for a block) should be logged.
The text was updated successfully, but these errors were encountered: