On Thu, Oct 10, 2013 at 5:48 PM, Robert <[email protected]> wrote: > Thanks for the documentation thread, I had not read that doc, but > obviosuly all the answers are there. so from what I am seeing and now > understanding from the imstats, is that so far I have not seen any > discarded messages (which is a good thing) , and what I am curious about is > that I constantly get the main q size up to the maximum, does that mean > that the server cannot process those messages as quickly as its receiving > them? > > I had a chance for a quick look at the /dev/null stats ... and the numbers somehow do not sum up.
Can you please re-run this test without the deltas (so use the default cumulative stats mode). Maybe there is a bug in delta mode, or maybe memory access races wreck it that much (one reason I did originally not intend to support deltas) - or maybe I am something misreading... Rainer > > > > > ----- Original Message ----- > > From: Rainer Gerhards > > Sent: 10/10/13 11:33 AM > > To: Robert > > Subject: Re: [rsyslog] Tr : Re: perfomance tweaking (fwd) > > > On Thu, Oct 10, 2013 at 5:10 PM, Robert <[email protected]> wrote: >> >> I have a question >> >> I am going through the the stats, and there are some instances where the >> action processed goes to 500,00 + does that mean that it writes out that >> many messages in that second? > > > yup, minus a small margin for i/o in progress plus the extra vagueness you > enabled by setting it to deltas. But I'd say +/-5% that's the number that > goes to that file during the period. > > > >> also what are the resource-usage numbers? >> > > > actually, it would save all of us time if you would read the doc links I > continously post ;) > > once again: > > http://www.rsyslog.com/rsyslog-statistic-counter/ > > I really have a reason point you to them ;-) > > Rainer > >> >> Thu Oct 10 10:43:47 2013: action 1: processed=1400000 failed=0 >> Thu Oct 10 10:43:47 2013: imudp(*:514): submitted=0 >> Thu Oct 10 10:43:47 2013: imudp(*:514): submitted=0 >> Thu Oct 10 10:43:47 2013: resource-usage: utime=127966546 stime=173249662 >> maxrss=20414472 minflt=5116893 majflt=0 inblock=0 oublock=21631064 >> nvcsw=4687486 nivcsw=59354 >> Thu Oct 10 10:43:47 2013: main Q: size=1985593 enqueued=0 full=0 >> discarded.full=0 discarded.nf=0 maxqsize=20000000 >> >> Thu Oct 10 10:43:48 2013: imuxsock: submitted=0 ratelimit.discarded=0 >> ratelimit.numratelimiters=0 >> Thu Oct 10 10:43:48 2013: action 1: processed=574000 failed=0 >> Thu Oct 10 10:43:48 2013: imudp(*:514): submitted=0 >> Thu Oct 10 10:43:48 2013: imudp(*:514): submitted=0 >> Thu Oct 10 10:43:48 2013: resource-usage: utime=131556000 stime=174801426 >> maxrss=20414472 minflt=5116893 majflt=0 inblock=0 oublock=22391480 >> nvcsw=4814080 nivcsw=59376 >> Thu Oct 10 10:43:48 2013: main Q: size=1410593 enqueued=0 full=0 >> discarded.full=0 discarded.nf=0 maxqsize=20000000 >> >> >> >> ----- Original Message ----- >> >> From: Robert >> >> Sent: 10/10/13 09:46 AM >> >> To: Rainer Gerhards >> >> Subject: Re: [rsyslog] Tr : Re: perfomance tweaking (fwd) >> >> >> These are the results with 1000 for both timeREquery and batchsize, writing >> out to /dev/null >> >> ----- Original Message ----- >> From: Rainer Gerhards >> Sent: 10/10/13 08:48 AM >> To: Robert >> Subject: Re: [rsyslog] Tr : Re: perfomance tweaking (fwd) >> >> umm.. again the ML thread is broken :-( >> >> On Thu, Oct 10, 2013 at 2:43 PM, Robert < [email protected] > wrote:From the >> documentation does that mean that I should lessen my timeRequery and >> batchSize? to something lower ? >> >> A relatively large batch size is a good thing. I would be conservative with >> timeRequery in a large batch environment. As I wrote in the doc, I'd >> recommend not to go higer than 10. >> >> _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
