era@niwot.scd.ucar.edu (Ed Arnold) (08/11/90)
When someone said the IBM RS/6000 is a "sun killer", they meant it literally. We have an RS/6000 that had a filesystem on a 4/280 running 4.0.3 remotely mounted, and when it blasted NFS UDP packets to the 4/280 at maximum speed, all of the nfsd-s (eight of them) froze into permanent wait states. This required a full reboot of the 4/280. The 22 client users were ever so pleased. Has anyone run into this problem? If so, have you been able to get a solution/patch from Sun? Please reply via e-mail to me, will summarize if I receive any info. Ed Arnold * NCAR * POB 3000, Boulder, CO 80307-3000 * 303-497-1253(voice) 303-497-1137(fax) * era@ncar.ucar.edu [128.117.64.4] * era@ncario.BITNET era@ncar.UUCP * Edward.Arnold@f809.n104.z1.FIDONET.ORG
era@ncar.ucar.edu (Ed Arnold) (08/14/90)
Thanks to the following persons for pointing out that I really ought to apply the "4.0.3 NFS jumbo patch" before giving up on mounting files from a 4/280 on an RS/6000 (which regularly hangs the 280's nfsd-s): emv@math.lsa.umich.edu (Edward Vielmetti) sinclair@strider.sps.mot.com (Brian Sinclair) jayl%sputnik@bit.UUCP (Jay Lessert) ehrlich@colossus.cs.psu.edu (Dan Ehrlich) This is a BIG patch (about 35 separate bugids) whose Patch-ID# is 100091-01. It's available via anonymous ftp from psuvax1.cs.psu.edu:~ftp/pub/sun-fixes, files Jumbo.NFS.Patch.README and jumbopatch.tar.Z, if you don't have sun support. -- Ed Arnold * NCAR * POB 3000, Boulder, CO 80307-3000 * 303-497-1253(voice) 303-497-1137(fax) * era@ncar.ucar.edu [128.117.64.4] * era@ncario.BITNET era@ncar.UUCP * Edward.Arnold@f809.n104.z1.FIDONET.ORG