[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
The following reply was made to PR kern/49186; it has been noted by GNATS.
From: Tom Ivar Helbekkmo <tih%hamartun.priv.no@localhost>
Subject: Re: kern/49186
Date: Tue, 09 Sep 2014 19:14:36 +0200
I have been advised that I ought to add the following data to this bug
My system is giving me occasional "No buffer space available" messages.
While it is mostly quiescent, these are from openvpn, and I've accepted
that as being caused by a fast local net and a slow uplink over which
the UDP based openvpn runs. When I load down the system with more work,
these messages get more frequent, and are joined by error messages from
other subsystems, in particular dhcpd, which fails to respond to client
packets, saying "send_packet: No buffer space available". The location
of the error is quoted in the console messages as dhcp.c:1561 and 3556.
(Those are just the two on my console terminal right now; there may well
be more.) Note that this happens whether the RAID write-back cache is
enabled or not; the change is that when the cache is enabled, I get the
long network hangs (not with every mbuf allocation failure by far, but
quite often), and with the cache disabled, I don't get noticable hangs.
It doesn't matter how beautiful your theory is, it doesn't matter how smart
you are. If it doesn't agree with experiment, it's wrong. -Richard Feynman
Main Index |
Thread Index |