| To: | pcp@xxxxxxxxxxx |
|---|---|
| Subject: | [Bug 1136] pmlogreduce type conversion conflicts with naive multi-archive same-type assertion |
| From: | bugzilla-daemon@xxxxxxxxxxx |
| Date: | Thu, 11 Feb 2016 02:19:34 +0000 |
| Auto-submitted: | auto-generated |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <bug-1136-835@xxxxxxxxxxxxxxxx/bugzilla/> |
| References: | <bug-1136-835@xxxxxxxxxxxxxxxx/bugzilla/> |
|
Comment # 3
on bug 1136
from Ken McDonell
(a) yes, the nirvana solution for pmlogreduce would be to convert counter,bytes -> instant,bytes/sec (b) we'd need to widen _every_ 64-bit counter for this to be a viable option, and that just does not feel right (c) widening was added to pmlogreduce because the longer the time interval that the archive spans the higher the probability of a counter wrap, and in the absence of (a) the simplest way to deal with this is to expand 32-bit counters to 64-bit counters I would prefer to fix this properly, not Band-Aid the "32-bit counters being widened to 64-bit counters thanks to pmlogreduce" problem. Needs some more soak-n-think time at my end ... You are receiving this mail because:
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | pcp updates: kenj+lukas+marko merges, Nathan Scott |
|---|---|
| Next by Date: | pcp updates - fix pcp-iostat QA issues and use SIGPIPE SIG_DFL, Mark Goodwin |
| Previous by Thread: | [Bug 1136] pmlogreduce type conversion conflicts with naive multi-archive same-type assertion, bugzilla-daemon |
| Next by Thread: | new pmiostat qa failures, Ken McDonell |
| Indexes: | [Date] [Thread] [Top] [All Lists] |