| To: | Ken McDonell <kenj@xxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: PCP Network Latency PMDA |
| From: | fche@xxxxxxxxxx (Frank Ch. Eigler) |
| Date: | Thu, 19 Jun 2014 21:57:10 -0400 |
| Cc: | pcp@xxxxxxxxxxx |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <53A352FF.9090906@xxxxxxxxxxxxxxxx> (Ken McDonell's message of "Fri, 20 Jun 2014 07:15:43 +1000") |
| References: | <53A34A47.3060008@xxxxxxxxxx> <53A352FF.9090906@xxxxxxxxxxxxxxxx> |
| User-agent: | Gnus/5.1008 (Gnus v5.10.8) Emacs/21.4 (gnu/linux) |
kenj wrote: > [...] Depending on the overhead of this instrumentation, you may > wish to consider using an additional control variable and pmStore(1) > to allow a user to enable or disable the collection [...] FWIW, I wish we leaned less on this particular technique. The pmda infrastructure has the ability to track individual pcp clients coming and going. So, a pmda can/should activate collection on demand, not based on a contentious and non-obvious global variable. - FChE |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [pcp] pmServiceDiscoveryInterrupt() commit a8b87e2 et al., Nathan Scott |
|---|---|
| Next by Date: | RE: pcp+graphite, take 2, Ken McDonell |
| Previous by Thread: | Re: [pcp] PCP Network Latency PMDA, Ken McDonell |
| Next by Thread: | Re: [pcp] PCP Network Latency PMDA, Nathan Scott |
| Indexes: | [Date] [Thread] [Top] [All Lists] |