Adding command requests_per_hour #34
Conversation
valleedelisle
force-pushed
the
requests_per_hour
branch
2 times, most recently
from
ca13b48
to
ba3eb5f
Compare
Refactored the time calculation function to be more dynamic
valleedelisle
force-pushed
the
requests_per_hour
branch
from
ba3eb5f
to
d95967f
Compare
| minutes=tm.minute % aggregate, | ||
| seconds=tm.second, | ||
| microseconds=tm.microsecond, | ||
| ) |
There was a problem hiding this comment.
@valleedelisle would you mind explaining the logic of this adding and removing? 🤔 I get the seconds and microseconds removal, as to zero them, but the aggregate puzzles me a little bit.
I was playing with the request_per_minute.log file and I see that what counts as a request that belongs to a minute goes from the half a minute before to the half a minute after, i.e.
Requests logged starting from 12:12:30 until 12:13:29 will all belong to 12:13:0 minute. Is what you expected to do?
At least, we could mention that it is not bound to the exact minute/hour to avoid users getting confused?
There was a problem hiding this comment.
@gforcada I based this part from this stackoverflow question. You're right when you say that :30 to :29 are going to be in the same minute. I didn't realize it would change the actual behavior maybe it would be better to keep the same behavior. The reason why I picked this is because we can pass any number of minutes to round up/down to.
There was a problem hiding this comment.
No actually it is fine, usually when I found solutions via stackoverflow I put a comment on top with the link to it, so one does not need lengthy explanations 👍
|
@valleedelisle thanks for the pull request! 💯 I would like to get a comment on the logic of it, but otherwise it looks good! 👍 |
|
@valleedelisle sorry I made a new release of haproxy_log_analysis with a complete rewrite of the commands and the main logic. Sadly this means that this pull request can no longer be merged as it is 😞 On the up side, the commands now are more easier to write and perform much better 👍 I can try to recreate your additions here on a new pull request adapted to the new code base. Would it be ok with you? 🤔 |
|
@gforcada Nice! I'll look into this, thanks for maintaining this project. Honestly, I was able to achieve what I had to do and extract the stats I needed so I kind of lost interest into this for now. Feel free to take what I did and adapt it however you think it fits best. Thanks again for the great job! |
|
Done at #36 |
Refactored the time calculation function to be more dynamic