[Beowulf] ntpd wonky?
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Gerry Creager gerry.creager at tamu.eduTue Feb 17 09:30:49 PST 2009
- Previous message: [Beowulf] Re: Tracing down 250ms open/chdir calls
- Next message: [Beowulf] Tracing down 250ms open/chdir calls
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Greg Lindahl wrote: > On Thu, Dec 11, 2008 at 11:01:31PM -0800, Bernard Li wrote: > >> Have you tried other pools eg. pool.ntp.org? > > That is stable for me. So it's not me, it's Red Hat's pool that's > wonky. > > I see that CentOS switched to using ntp.org in 5.2, which I didn't > automagically get thanks to rpm creating ntpservers.rpmnew, even > though I hadn't modified the ntpservers file. Mmf. After a network performance class at the Internet2 Joint Techs meeting in early Feb, I ditched the centos/redhat pools, and have a bunch of discrete entries in my ntp.conf file. The entries I'm using are seen below: server owamp.hous.net.internet2.edu server owamp.chic.net.internet2.edu server owamp.atla.net.internet2.edu server now.okstate.edu server terrapin.csc.ncsu.edu server otc1.psu.edu server time-c.timefreq.bldrdoc.gov server utcnist.colorado.edu ntpstat is showing a fairly stable 0.5 ms jitter with a 37 ms offset from an IRIG-locked system at owamp.hous.net.internet2.edu The recommendation from the I2 geeks was to have at least 5 servers of known stratum 1 capability. I'm planning to put my own GPS-timed system into my network to accommodate all my clusters. gerry
- Previous message: [Beowulf] Re: Tracing down 250ms open/chdir calls
- Next message: [Beowulf] Tracing down 250ms open/chdir calls
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
