nanog mailing list archives
Re: NIST NTP servers
From: Lyndon Nerenberg <lyndon () orthanc ca>
Date: Wed, 11 May 2016 18:12:03 -0700
On May 11, 2016, at 5:42 PM, Scott Weeks <surfer () mauigateway com> wrote: Wouldn't the buffers empty in a FIFO manner?
They will empty in whatever order the implementation decides to write them. But what's more important is the order in which the incoming packets are presented to the syslogd process. If you're listening on TCP connections, the receive order is very much determined by the strategy the syslogd implementation uses to read from FDs with available data. I.e. elevator scan, lowest/highest first, circular queue, ... In a threaded implementation, your reader workers, buffer writers, etc., are all at the mercy of the threading implementation; it's difficult to control thread dispatch ordering at that level of granularity. --lyndon
Current thread:
- Re: NIST NTP servers, (continued)
- Re: NIST NTP servers Mel Beckman (May 11)
- Re: NIST NTP servers Jay R. Ashworth (May 11)
- Re: NIST NTP servers Majdi S. Abbas (May 11)
- Re: NIST NTP servers Jared Mauch (May 12)
- Re: NIST NTP servers Mike (May 12)
- Re: NIST NTP servers Gary E. Miller (May 11)
- Re: NIST NTP servers Eygene Ryabinkin (May 11)
- Re: NIST NTP servers Gary E. Miller (May 11)
- Re: NIST NTP servers Lyndon Nerenberg (May 11)