Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 6 Dec 2001 10:04:10 -0500
From:      Bill Vermillion <bv@wjv.com>
To:        stable@FreeBSD.ORG
Subject:   Re: stable-digest V5 #360
Message-ID:  <20011206100410.C12127@wjv.com>
In-Reply-To: <bulk.11497.20011206064930@hub.freebsd.org>; from owner-freebsd-stable-digest@FreeBSD.ORG on Thu, Dec 06, 2001 at 06:49:30AM -0800
References:  <bulk.11497.20011206064930@hub.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
> Re: Whats with this -> sendto: No buffer space available


> Date: Thu, 06 Dec 2001 18:28:48 +1100
> From: Kal Torak <kaltorak@quake.com.au>
> Subject: Re: Whats with this -> sendto: No buffer space available
> 
> Doug Silver wrote:
> 
> > It happens all the time on one of my nameservers, but only when amanda is
> > trying to back it up.  Then it says this:
> > 
> > 
> >>Dec  4 23:00:43 HOST syslogd: sendto: No buffer space available
> >>Dec  5 00:00:36 HOST syslogd: sendto: No buffer space available
> >>
> > 
> > It does not affect the machine other than this message.
> 
> Hmmm, no affect???
> 
> When ever I get this message the machine can no longer send any data
> to the network on that interface... The interface has to be reset...
> 
> Perhaps the cause of your message is due to lack of mbufs for the load
> its getting, you should probably increase them...
> 
> The problem the rest of us have is different... I am almost 100% sure
> this is being caused by a problem upstream, the data cant get out so
> the sendto buffer fills and seems to lock up the interface... I have a
> feeling this may be specific to PPP connections, anyone finding this
> locks there interface on a non PPP connection??? (pppoe is still ppp
> could this even be specific to pppoe??)
> 
> Anyway to my way of thinking this is an error in the handeling of this
> kind of situation and needs to be looked at... Is there anything that
> I can log or dump next time this happens that would be helpfull in fixing
> this???

Well my no buffer space available came from a DSL.  Not pppoe.
It's actually ppoa - BUT - all that is handled by the DSL modem.
I have private IPs.   I had a WinME machine doing something, I had
this machine doing something, and my other BSD machine was
performing a DL from an FTP site in Europe or possibly to one of my
servers. [My DSL is 512Kb down while the servers I admin are
100Mbit uplinks into an OC192 backbone and I often DL to there and
then just perform a final transfer to here as I contol both ends of
the link and and wont have delays/timeouts]

The machine with the FTP got the no buffer space available and it
and this machine are running 4.4.  This machine was just recently
reloaded with 4.4 from CD as I stupidly trashed my HD by accident.

Since the machine that died was performing pure IP to IP [as it
looked to itself] there was no pppoe involved.

Bringing the interface down and backup fixed it.  

If it makes any difference the image 
> 
> 
> End of stable-digest V5 #360
> ****************************
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with unsubscribe freebsd-stable-digest in the body of the message
> 

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011206100410.C12127>