Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

powerman: support new --diag option #163

Closed
wants to merge 8 commits into from

Conversation

chu11
Copy link
Member

@chu11 chu11 commented Mar 9, 2024

Built upon #161

Now that the new setresult statement is available in #161, it's trivial to report back why individual power operations failed.

e.g.

> pm --diag -1 t[0-15]
t[7,15]: error
Command completed with errors

Since I don't want default output to change too much I have this dispalyed via a new --diag option to powerman. Can debate option name of course.

Will be super helpful as we work towards completing #128

Problem: in the near future we would like to create a PlugList
from a List of Plugs.  It is currently inconvenient to create
such a list given the current PlugList API.

Add a function pluglist_copy_from_list() function to create a PlugList
from a List of Plugs.
Problem: In powerman device files the only "interpretation" that
is done is the output of a power status query via the
"setplugstate" statement.  Other interpretations may be done
in the future against other power operations.  Internally anything
associated with "setplugstate" is called "interpretation" (or
"interp" or "interps", etc.).  This will get confusing with future
interpretations.

Solution:  Rename any "interpreations" related code associated
with "setplugstate" to "state interp" (i.e. "StateInterp" or
"state_interp", etc.).
Problem: In many cases, there is no way for a power operation (i.e.
power on, but not power status) to inform powerman that an error has
occurred in the operation.  The user will always get a
"Command completed successfully" output and exit status 0 after
issuing a power operation.

Solution: Support a new "setresult" statement that can inform powerman
that a power operation did not succeed.  A regex can be used to determine
what output is expected of a successful power operation.  If any are not
successful, powerman can subsequently inform the user an error has occurred,
leading to a "Command completed with errors" message and exit status 1.

Some example uses:

script on_all {
	send "on\n"
	foreachnode {
		expect "([^\n:]+): ([^\n]+\n)"
		setresult $1 $2 success="^ok\n"
	}
	expect "redfishpower> "
}

script on {
	send "on %s\n"
	expect "([^\n:]+): ([^\n]+\n)"
	setresult $1 $2 success="^ok\n"
	expect "redfishpower> "
}

Fixes chaos#79
Problem: The new setresult statement is not documented ni powerman.dev(5)

Add information about the new device statement.
Problem: The new setresult statement is not covered in the testsuite.

Add tests in a new t0034-power-result.t.  Add new test device files
etc/redfishpower-setresult-all.dev, etc/redfishpower-setresult-range.dev,
and etc/redfishpower-setresult-singlet.dev which cover the "all", "range",
and "singlet" versions of various power operations (i.e. on_all, vs
on_range, vs on).
@chu11
Copy link
Member Author

chu11 commented Mar 15, 2024

it occurred to me that supporting --diag w/ setplugstate (and thus pm -q) would be trivially easy ... could be useful?

pro: it'd be useful when diagnosing problems with pm -q
con: it might be weird that lots of boards support --diag with pm -q but don't with pm -1/-0?

edit: re-pushed supporting within --query as well

Problem: When power control/query to a target fails, there is no way for
a user to know why it failed.

Add a new --diag to powerman that will inform powermand to send
diagnostic information about why a power operation failed.
This only works with setplugstate and the new setresult statement.
Problem: There is no coverage for the new --diag option.

Add tests in new t0035-diag.t tests.
Problem: The new --diag option is not documented.

Add it in powerman(1).
@chu11
Copy link
Member Author

chu11 commented Mar 19, 2024

i'm going to close this as I've recently merged this into my mega cray ex chassis PR chain. Don't want to mess up the order of all the broken up PRs.

@chu11 chu11 closed this Mar 19, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant