Report warning if space for dump on disk is smaller than requested #30
Conversation
kobliha
commented
Jun 17, 2015
kobliha
commented
- Uses new API implemented here Implemented "triggers" for calling installation proposals again when needed yast-installation#292
- FATE#317488: YaST kdump: issue warning, if space on "/" is smaller than (RAM + 4Gib)
- FATE #317488: YaST kdump: issue warning, if space for dump is smaller than (RAM + 4 GiB) - including test cases
| "expect" => " | ||
| Yast.import \"Kdump\" | ||
| Yast::Kdump.free_space_for_dump | ||
| ", |
There was a problem hiding this comment.
with fun_ref it can look like
def expectation
Yast.import "Kdump"
Kdump.free_space_for_dump
end
...
...
"expect" => fun_ref(method(:expectation), "boolean ()")
this way it is much easier to test your expectation method and easier to read it.
There was a problem hiding this comment.
@jreidinger OK, next time I'll discuss ideas with you :)
|
@kobliha Here, an award for the longest branch name so far: 🏆 |
| } | ||
| } | ||
|
|
||
| warning = Kdump.proposal_warnig |
There was a problem hiding this comment.
Well, at least it's consistent :)
|
Thank you @mvidner :) |
| } | ||
|
|
||
| warning = Kdump.proposal_warnig | ||
| unless warning.empty? |
There was a problem hiding this comment.
IMHO this is bad practice, similar to description += item unless item.empty?. Adding an empty object works as intended so we don't need the condition.
- _B, _b constant/method suffixes - better warning text (from Ken) - different text for little difference between required/available space - fun_ref in action
kobliha
force-pushed
the
report_warning_if_space_for_dump_on_disk_is_smaller_than_requested
branch
from
2f2911b
to
5507969
Compare
- After several iterations, this should finally work in running Yast, unit tests, and also for Yast Team members as it doesn't use that evil eval
|
I don't know if @mvidner has something to say, it looks good to me. |
|
|
||
| # Smaller differences are not possible to show with quite small precision | ||
| # in human-readable size formatting | ||
| if (requested_space * 0.99) < free_space && free_space < requested_space |
There was a problem hiding this comment.
I don't understand this. Why the 1%? Doesn't requested_space contain enough buffer already?
There was a problem hiding this comment.
1%+ difference can be seen in the result of String.FormatSizeWithPrecision(requested_space, 2, true) 2 is a precision
There was a problem hiding this comment.
IMHO this is overengineering, accounting for a case where we need 2000MB but have only 1999MB and the text says "We need 2.00GB but have only 2.00GB available". I think nobody will mind, and I will close such bug reports as wontfix.
Please revert this, then it LGTM.
…g reporting "Kdump warns that it needs 8.5 GiB, but 8.5 GB is available" as WONTFIX - The human-readable form of "size" always fights with rounding - 8.5 GiB does not need to == 8.5 GiB exactly - But let's live with that :)
…on_disk_is_smaller_than_requested Report warning if space for dump on disk is smaller than requested
kobliha
deleted the
report_warning_if_space_for_dump_on_disk_is_smaller_than_requested
branch
