rclone seems to log all non-fatal retry errors to stdout instead of stderr. Is this normal behaviour? According to the docs (https://rclone.org/docs/#logging), all logs goes to stderr.
The log I posted below is the results of running rclone inside a script with the stderr redirected, so the terminal is only displaying stdout output. I couldn't post the logs with '-vv' because I initially didn't use that and I'm not sure when I will encounter a retry error again.
What is your rclone version (output from rclone version)
go version: go1.14.4
Which OS you are using and how many bits (eg Windows 7, 64 bit)
Microsoft Windows [Version 10.0.18362.900] 64-bit
Which cloud storage system are you using? (eg Google Drive)
The command you were trying to run (eg rclone copy /tmp remote:tmp)
Yes, my question is why was this message (in OP) sent to stdout instead of stderr? Even non-'Error' level messages (like that 'Notice' level message about config files) get logged to stderr, so why is this particular 'Error' level message being sent to stdout? Is there anything about this in the docs?
I understand some output gets to go to stdout, but a log message isn't part of the normal output of a command right? Progress status being sent to stdout is understandable, but not an error message.
The error message I posted is explicitly a log message. And supposedly, by default, rclone logs to standard error.
Even the notice message about config files got sent to stderr after all, so I'm pretty confused.
you are using this flag, which will display log messages. https://rclone.org/docs/#p-progress
Any log messages will scroll above the static block. Log messages will push the static block down to the bottom of the terminal where it will stay.
i am not very interested in non-fatal retries, which are very common.
i just need to know if the rclone command completed without a fatal error.
i do not rely on stdout and --progress to determine success or failure of a rclone command.
i do the following
--log-file=c:\path\to\log\file.txt and --log-level=DEBUG and parse the log file
Yes, I'm already checking the exit codes, I might try using the log file option too. It's just that this unexpected behavior tripped me up and got me worried, so I wanted to clarify a few things. I hope that this -P behavior does not happen to actual fatal errors though (haven't encountered one so far)