| To: | Ken McDonell <kenj@xxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [pcp] URGENT potentially serious regression in 3.7.0 |
| From: | Nathan Scott <nathans@xxxxxxxxxx> |
| Date: | Sat, 9 Mar 2013 16:36:36 -0500 (EST) |
| Cc: | pcp@xxxxxxxxxxx |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <730602424.17617863.1362864098798.JavaMail.root@xxxxxxxxxx> |
| Reply-to: | Nathan Scott <nathans@xxxxxxxxxx> |
Ah, looks like the crazy level of diagnostics is coming from here
in pmcd...
671 #ifdef PCP_DEBUG
672 if (pmDebug & DBG_TRACE_APPL0)
673 for (i = 0; i <= maxClientFd; i++)
674 if (__pmFD_ISSET(i, &readableFds))
675 fprintf(stderr, "DATA: from %s (fd %d)\n",
FdToString(i), i);
676 #endif
The NSS variant of __pmFD_ISSET looks up the NSS socket, and that
involves a call to __pmDataIPC, which has this diagnostic on each
entry:
if (pmDebug & DBG_TRACE_CONTEXT)
fprintf(stderr, "__pmDataIPC: fd=%d\n", fd);
so, ends up being alot & pmcd will slow down cos its issuing so
many log messages.
cheers.
--
Nathan
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [pcp] URGENT potentially serious regression in 3.7.0, Ken McDonell |
|---|---|
| Next by Date: | Re: [pcp] URGENT potentially serious regression in 3.7.0, Nathan Scott |
| Previous by Thread: | Re: [pcp] URGENT potentially serious regression in 3.7.0, Ken McDonell |
| Next by Thread: | Re: [pcp] URGENT potentially serious regression in 3.7.0, Nathan Scott |
| Indexes: | [Date] [Thread] [Top] [All Lists] |