Re: Solaris ARP bug?
From: Barry Margolin (barmar_at_alum.mit.edu)
Date: 02/03/04
- Next message: Scott Howard: "Re: Help! OBP update -> Additional Info"
- Previous message: Tom Taylor: "Re: Help! OBP update -> Additional Info"
- In reply to: Michael Steele: "Re: Solaris ARP bug?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 03 Feb 2004 03:25:27 GMT
In article <20ATb.168032$nt4.752197@attbi_s51>,
Michael Steele <michael+nospam@netsteele.com> wrote:
> As far as I know, the Ethernet specification requires that when a host
> sends an ARP RESPONSE, all hosts listening on that wire should cache the
> entry.
Is ARP really covered in the Ethernet specification? Is the description
there different from RFC 826?
RFC 826 says that a host should update an *existing* ARP cache entry
when it receives any ARP packet, and should create one when it receives
an ARP query for itself.
Sun has gone beyond this, caching all ARP information that it receives.
However, information from unsolicited messages has a shorter timeout, so
that they won't overflow the ARP cache.
-- Barry Margolin, barmar@alum.mit.edu Arlington, MA *** PLEASE post questions in newsgroups, not directly to me ***
- Next message: Scott Howard: "Re: Help! OBP update -> Additional Info"
- Previous message: Tom Taylor: "Re: Help! OBP update -> Additional Info"
- In reply to: Michael Steele: "Re: Solaris ARP bug?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|