| To: | Greg Banks <gbanks@xxxxxxx> |
|---|---|
| Subject: | Re: [pcp] braindump on unified-context / live-logging |
| From: | Nathan Scott <nathans@xxxxxxxxxx> |
| Date: | Tue, 14 Jan 2014 21:38:10 -0500 (EST) |
| Cc: | pcp developers <pcp@xxxxxxxxxxx> |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <52D5ED4C.9060107@xxxxxxx> |
| References: | <20140108013956.GG15448@xxxxxxxxxx> <20140110190525.GA28062@xxxxxxxxxx> <0a923e$520gar@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> <52D4666E.7030601@xxxxxxx> <21204.32676.163457.438142@xxxxxxxxxxxx> <52D49301.2000403@xxxxxxx> <440877259.2327550.1389747502556.JavaMail.root@xxxxxxxxxx> <52D5ED4C.9060107@xxxxxxx> |
| Reply-to: | Nathan Scott <nathans@xxxxxxxxxx> |
| Thread-index: | 4gSugbwFxh/1bgIMXxAHoRJ5ZnPPUA== |
| Thread-topic: | braindump on unified-context / live-logging |
----- Original Message ----- > ... > Ken did something to the guts of pmlogger which made it write the files > in the correct order, with an fsync(). It worked and I think it was > checked in, but I'm not sure. We couldn't use it because our design > relied on a stock PCP and there was no way to ship an update. There are no fsync's in either libpcp or pmlogger, but there is a series of fflush's which might be what you're thinking of here (I hope!). > Plus it only solved half the problem, the lag being the other half. > > Plus again, it meant that any metric we wanted to present an average for See plus plus? Worst... joke... ever. > ... > Hope this helps. Very much - thanks! cheers. -- Nathan |
| Previous by Date: | Re: [pcp] braindump on unified-context / live-logging, Greg Banks |
|---|---|
| Next by Date: | Processing of pcp_3.8.10_i386.changes, Debian FTP Masters |
| Previous by Thread: | Re: [pcp] braindump on unified-context / live-logging, Greg Banks |
| Next by Thread: | Re: [pcp] braindump on unified-context / live-logging, Dave Brolley |
| Indexes: | [Date] [Thread] [Top] [All Lists] |