| To: | Nathan Scott <nathans@xxxxxxxxxx> |
|---|---|
| Subject: | Re: [pcp] pcp updates |
| From: | Ken McDonell <kenj@xxxxxxxxxxxxxxxx> |
| Date: | Fri, 05 Jun 2015 19:40:39 +1000 |
| Cc: | pcp@xxxxxxxxxxx |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <539192193.12426395.1433464810947.JavaMail.zimbra@xxxxxxxxxx> |
| References: | <556F6D57.3030303@xxxxxxxxxxxxxxxx> <1693681402.11695125.1433374119614.JavaMail.zimbra@xxxxxxxxxx> <5570E621.3090005@xxxxxxxxxxxxxxxx> <539192193.12426395.1433464810947.JavaMail.zimbra@xxxxxxxxxx> |
| User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0 |
On 05/06/15 10:40, Nathan Scott wrote: ... I *think* its OK as is, and is behaving as I think we both expect ... but I might well be missing something more subtle there? Yep, the expanded qa/948 exercises this. There are however some fallout issues from pmlogger using PM_CONTEXT_LOCAL1. the pmcd.pmlogger metrics are probably wrong and/or need some special values to indicate PM_CONTEXT_LOCAL is being used 2. pmlc cannot be used to control a pmlogger that is using PM_CONTEXT_LOCAL because pmlc tries to connect to the pmcd that the pmlogger instance is connected to. If we can devise a fix for 1. then pmlc could be taught to not try to control these pmlogger instances. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | pcp updates: merge mgoodwin (containers, deprecate hinv.map.lvname) + kenj (qa, pmlogger local context, configure updates, newlog usage message) + nathans (python qa and other qa), Mark Goodwin |
|---|---|
| Next by Date: | Re: [pcp] JSON PMDA with indom cache changes, Ken McDonell |
| Previous by Thread: | Re: [pcp] pcp updates, Nathan Scott |
| Next by Thread: | pcp updates: merge dsmith (json pmda and assorted other python fixes) + nathans (atop, pmie, local context pmlogger, qa) + paul smith (pmtime for osx), Mark Goodwin |
| Indexes: | [Date] [Thread] [Top] [All Lists] |