[ale] networking backpressure

JD jdp at algoloma.com
Tue Apr 9 15:08:37 EDT 2013


On 04/09/2013 02:33 PM, Mike Harrison wrote:
> On Tue, 9 Apr 2013, Jim Kinney wrote:
>> so what happens when a series of machines, each with 1G nic all try to
>> read/write to the same NFS mount filesystem,
>> also on a single 1G nic, and this goes on for hours at a time? Eventually, nic
>> buffers fill and the OS is told to "hold that thought". But for how long? Will
>> the OS continue to try
>> the write until it succeeds? Or will the constant buffer overload just cause a
>> nic to flush the buffer and hope for
>> better luck next send?

My use of NFS is a little dated ... seems it would depend on the type of mount
used - spongy, soft, or hard.
If hard, it will try again and again and will appear to lockup the machine, IME.
With soft mounts, writes aren't guaranteed.
Spongy has the good of hard and the good of soft mounts. It will give up after a
specific amount of time and continue as best it can - providing errors as
required.  NFS mounts have lots of tunable params, if I recall correctly.

Also, the specific buffer size for the involved NICs would matter too, but
filling a network buffer should happen relatively quickly.


More information about the Ale mailing list