Bug 29773

Summary: server disapear from the network - nothing in logs
Product: Base System Reporter: Pawel Fornalski <pkf>
Component: confAssignee: Dag-Erling Smørgrav <des>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: Unspecified   
Hardware: Any   
OS: Any   

Description Pawel Fornalski 2001-08-16 13:40:01 UTC 
We (i am owner) and my Adminisrator have very strange and very dificult for us problem. Since several moths our server looses contact with the internet. We have ABIT KT7-RAID mainboard and Intel Ether Express 100+ network adapter.
this server temporarely looses signal and after serveral minutes, sometimes hours get back to work. In logs and problem machine there is no sign after this. Once we were able to be on the console when it happened and we ping'ed to some adress. we run tcpdump and it looked like the tcp packets got into black hole.
our server have rather big traffic (about 100000 hits a day) and this problem anoys me very much. 
it is not a problem of the network because there is 2 more servers connected to the same switch and they work. both are Free BSD too.

can you help me and my administrator with solving the problem??
please give a suggestion what can it be.

How-To-Repeat: by e-mail on pkf@hip-hop.pl (me) or craimz@hip-hop.pl (administrator)
Comment 1 Kris Kennaway 2001-08-19 10:39:14 UTC 
On Thu, Aug 16, 2001 at 05:36:41AM -0700, Pawel Fornalski wrote:

> We (i am owner) and my Adminisrator have very strange and very dificult for us problem. Since several moths our server looses contact with the internet. We have ABIT KT7-RAID mainboard and Intel Ether Express 100+ network adapter.
> this server temporarely looses signal and after serveral minutes, sometimes hours get back to work. In logs and problem machine there is no sign after this. Once we were able to be on the console when it happened and we ping'ed to some adress. we run tcpdump and it looked like the tcp packets got into black hole.
> our server have rather big traffic (about 100000 hits a day) and this problem anoys me very much. 
> it is not a problem of the network because there is 2 more servers connected to the same switch and they work. both are Free BSD too.
> 
> can you help me and my administrator with solving the problem??


You'll probably have more luck asking this question on
net@FreeBSD.org: the GNATS database is for problems which can be
replicated by FreeBSD developers, and this one doesn't include enough
details for anyone to be able to solve it at this point.

Kris
Comment 2 Jeroen Ruigrok van der Werven freebsd_committer freebsd_triage 2001-11-15 17:37:15 UTC 
State Changed
From-To: open->closed

Try cvsupping.  The fxp driver underwent massive changes and small 
bugs were fixed in the stack.
Comment 3 Dag-Erling Smørgrav freebsd_committer freebsd_triage 2001-11-15 20:39:42 UTC 
State Changed
From-To: closed->feedback

This PR will stay in feedback state until we hear back from the originator 
or it times out.
Comment 4 Dag-Erling Smørgrav freebsd_committer freebsd_triage 2001-11-15 20:39:42 UTC 
Responsible Changed
From-To: freebsd-bugs->des

I'll handle this.
Comment 5 Dag-Erling Smørgrav freebsd_committer freebsd_triage 2001-11-27 13:34:12 UTC 
State Changed
From-To: feedback->closed

PR contains no useful details, and originator does not respond to inquiries.