Menu
▾
▴
Re: [Ptpd-devel] PTPd log
|
From: Wojciech O. <woj...@ow...> - 2013-11-07 02:15:08
|
Fabrizio, In the pre-2.3 versions you can rate limit the log messages to 1 per n seconds with the -V option (or use logrotate, I think you need to send ptpd a SIHGUP for it to reopen the log file after the log has been cycled). It also supports the TICTOC PTP SNMP MIB and PTP management messages, so those mechanisms could be used to poll the offset from master. Right, I didn't answer - my answer to many ptpd-related questions right now is wait for 2.3 and upgrade. Regards, Wojciech - Wojciech Owczarek On 6 Nov 2013, at 23:16, Fabrizio Giordano <Fab...@ri...> wrote: > Thanks for your reply Wojciech, > > I’ll try version 2.3.0 and see if it works better for me. > > I’m aware that ptpd2 also logs events to syslog, but what I’m interested in is actually the information contained in the “statistics log” (option –f). In particular, for any given moment, I need to know what’s the offset between the slave system clock and the master clock (and by the way I’m not 100% sure which value among the 4 listed there is the one I’m looking for). > > But anyway, since you didn’t answer, I assume there is no way to limit the verbosity and there is no other way to retrieve the information contained in that file, is there? > > Thanks > > Fabrizio > > > From: Wojciech Owczarek [mailto:woj...@ow...] > Sent: Wednesday, November 06, 2013 14:40 > To: Fabrizio Giordano > Cc: ptp...@li... > Subject: Re: [Ptpd-devel] PTPd log > > Hi Fabrizio, > > The log that ptpd <= 2.2.2 maintains is nothing but a dump of protocol engine runtime variables - I don't think it's useful for much more than graphing and troubleshooting (i.e. not for "normal" operation [warning: personal opinion]. What people don't usually notice (and no wonder, because -f is meant to be "the" log file) is that ptpd maintains a proper event log, but by default does so via syslog. > > In ptpd 2.3.0+ (current svn), there is a separate log file (well, or syslog if you so choose), with built-in log rotation, which maintains the full event log, and the old "-f" is clearly called a statistics log. There is also the status file /var/run/ptpd2.status / or whatever you choose to name it, which provides a constantly updated at-a-glance view of ptpd's status. This includes the offset from master and much more - enough for proper monitoring and troubleshooting of most typical problem scenarios. The statistics log can be disabled / re-enabled and rate limited, all on the fly. > > ptpd 2.3.0 is full of feature enhancements, but we will need to document them and howto them. > > I strongly encourage all users to build the soon to be released svn code with --enable-statistics and --enable-ntpdc and scroll through "ptpd2 -H" to see all the config options and get an idea of all the features coming. > > > Regards, > Wojciech > > - > Wojciech Owczarek > > > > On 6 Nov 2013, at 18:58, Fabrizio Giordano <Fab...@ri...> wrote: > > Hello folks, > > I have a question about the log information written to file when the option –f is used with ptpd2. > Is there a way to reduce its verbosity? It writes a lot of information and the file grows pretty quickly. > > Also, I’m interested in the information written to that file (namely the offset with the master). Is there any other way to retrieve that information other than reading that log file? > > Thank you > > Fabrizio > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > _______________________________________________ > Ptpd-devel mailing list > Ptp...@li... > https://lists.sourceforge.net/lists/listinfo/ptpd-devel |
