home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gatech!destroyer!gumby!yale!yale.edu!ira.uka.de!chx400!sicsun!disuns2!disuns2.epfl.ch!simon
- From: simon@lia.di.epfl.ch (Simon Leinen)
- Newsgroups: comp.protocols.time.ntp
- Subject: Re: ntp v2 on SGI Irix 4.0.x
- Message-ID: <SIMON.92Jul24114128@liasg1.epfl.ch>
- Date: 24 Jul 92 09:41:28 GMT
- References: <1992Jul23.160745*Harald.Eikrem@delab.sintef.no>
- Sender: news@disuns2.epfl.ch
- Organization: DI-LIA -- Ecole Polytechnique Federale de Lausanne
- Lines: 38
- Nntp-Posting-Host: liasg1.epfl.ch
- In-reply-to: Harald.Eikrem@delab.sintef.no's message of 23 Jul 92 14:07:45 GMT
- X-Md4-Signature: c7d6d4568c56d66162ca4bca820e94db
-
- In article <1992Jul23.160745*Harald.Eikrem@delab.sintef.no>
- Harald.Eikrem@delab.sintef.no writes:
-
- Actually, after many attempts to get xntpd up and running, I gave
- it up. It went straight down without any messages whatsoever.
- Then, after I made some fixes to the source files, mainly in signal
- handlers, it seemed like I could make use of ntpdate. Irix
- appearently resets signal handling to SIG_DFL upon traps.
-
- That's the default System V behaviour. You can add -D_BSD_SIGNALS to
- DEFS in the configuration file to get around that.
-
- A very good place to start would be the patches to xntp (v2) posted by
- Scott Henry of SGI some time ago. You can FTP them from
- liasun3.epfl.ch as /pub/net/ntp/xntp-SGI.diffs.
-
- As far as I can tell, the same problems appear with ntp v3, but
- that one I could not even get going on SunOS machines, so I figured
- I had to live with ntp v2 for the time being.
-
- I found xntp.v3 quite easy to install on SunOS. Maybe you picked up a
- bad release?
-
- Now, running ntpdate from cron on the Iris'es seemingly works on
- most machines except one, which has Irix 4.0.4. The others are
- Irix 4.0.1 and 4.0.2. Have anyone else fiddled with ntp and
- Iris'es out there?
-
- Yes, we have xntp.v3 running on our two Indigos (IRIX 4.0.4) for a few
- days, but I don't trust my modifications of the loopfilter code. The
- machines do seem to keep good time *as far as I can tell* which is not
- very far because I don't know how to interpret the statistics files.
-
- It would be great if Scott could adapt his patches tp xntp.v3 and
- submit them for integration into the distribution! The SGI boxes
- should be very good for high-precision timekeeping.
- --
- Simon.
-