Added a session token auth method to rf_diagnostic_data #182
Conversation
Signed-off-by: Brandon Biggs <[email protected]>
|
I added the signoff this time in my commit but the error is still there? Not sure why that's happening? |
Very strange, I don't know why... Did you use the exact commands the DCO checker recommended, or did you sign it off another way? |
|
As a general comment, I had been thinking about doing something just like this for resolving issue #174. One thing that I was considering was making this handled in a common manner so that we don't have to copy/paste the same code in the different scripts. We started using redfish_utilities/misc.py to contain other common logic. |
I did use the |
Signed-off-by: Brandon Biggs <[email protected]>
…logger script that allows for more options when logging. Updated rf_diagnostic_data with the new proposed argument and logger scheme Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: Brandon Biggs <[email protected]>
|
The DCO issue is my fault. It's expecting the hostname of the system I'm on rather than my actual email. Not sure how to fix that, but my signoffs do appear to be in the commits. |
|
@mraineri I created a new file for arguments and one for logging. Tried to put common arguments into the new arguments.py. Thoughts on this style? |
| def create_parent_parser(description: str = "", auth: bool = False, rhost: bool = False): | ||
| parent_parser = argparse.ArgumentParser(description=description, add_help=False) | ||
|
|
||
| parent_parser.add_argument( |
There was a problem hiding this comment.
Could we maintain parity with the existing --debug argument for the time being? I have a lot of folks relying on this usage now to debug everything.
There was a problem hiding this comment.
Okay, I'm back to this. Sorry it took me so long. I believe I added back the existing --debug functionality in diagnostic_data. I may have misunderstood you though.
|
Other than those few comments, this is looking really nice! |
… that is used everywhere else
Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: Brandon Biggs <[email protected]>
…logger script that allows for more options when logging. Updated rf_diagnostic_data with the new proposed argument and logger scheme Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: Mike Raineri <[email protected]> Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: GitHub Release Workflow <> Signed-off-by: Brandon Biggs <[email protected]>
…g a manager's time Signed-off-by: Mike Raineri <[email protected]> Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: GitHub Release Workflow <> Signed-off-by: Brandon Biggs <[email protected]>
… that is used everywhere else Signed-off-by: Brandon Biggs <[email protected]>
…ebox into session-token
Signed-off-by: Brandon Biggs <[email protected]>
Signed-off-by: Brandon Biggs <[email protected]>
|
Okay, sorry for all the extra commits on this, tried to get it to fix the checks. I think I was able to resolve your concerns, please let me know if not. I tried to maintain parity with the current debugging. I think I'd like to clean up the session token stuff a bit, but I'd like to close out any lingering issues with this PR since it's a bit older. Let me know what you think. |
| raise ValueError(f"Invalid debug level: {level}") | ||
|
|
||
|
|
||
| def setup_logger( |
There was a problem hiding this comment.
It looks like the --debug option isn't used when setting up the debug output. Can we add it in so that if specified, it logs to a file (and is set to "loggingDEBUG")?
|
@brandonbiggs sorry I missed the fact you were pushing updates.... There's one last nit with the debug options (mainly ensuring that if a user specifies |
2 participants
I wanted to try this on a single file before I added it to the others.