home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!agate!dog.ee.lbl.gov!cad4.lbl.gov!mario
- From: mario@cad4.lbl.gov (mario aranha)
- Newsgroups: comp.sys.hp
- Subject: Re: clusters
- Date: 5 Jan 1993 22:46:44 GMT
- Organization: Lawrence Berkeley Laboratory
- Lines: 26
- Message-ID: <28226@dog.ee.lbl.gov>
- References: <BzFsCM.H5u@bunyip.cc.uq.oz.au> <42692@sdcc12.ucsd.edu>
- Reply-To: mario@cad4 (Mario Aranha)
- NNTP-Posting-Host: 128.3.252.177
- Keywords: cluster
-
- >Clusters work great under HP-UX. The machines don't have to be
- >identical, but using the same series is best. You should set up
- >your computers with a small local swap disk (although net swapping
- >works ;-).
-
-
- We have ~15 300/400's and ~10 700s. Any significant comments on
- making a mixed-cluster out of these?
-
- Any stats on server swap vs. local swap ? Is local swap an order
- of magnitude faster or what?
-
- Was told HP plans to replace clusters with NFS in the near future.
- Is this true?
-
- >I just wish that the 700s would loop retry start-up after a
- >crash. After a short time out, reboot quits and you have to hike to
- >the console to restart the client-boot selection process...
- >Essentially you save about 100MB disk per machine, since a single
- >copy of the OS is shared between the cluster members.
-
-
- I noticed that some 700s came with inadequte factory-set timeouts.
- Try the following in boot-admin mode on offending clients:
-
- BOOT_ADMIN> path primary lan.server_address.24.24
-