[OpenAFS] Re: OpenAFS freeze problems

Andrew Deason adeason@sinenomine.net
Tue, 28 Feb 2012 10:36:26 -0600

On Mon, 27 Feb 2012 20:00:29 -0600
John Tang Boyland <boyland@pabst.cs.uwm.edu> wrote:

> ] About every few hours or so, AFS "freezes" on a write:
> ] the attempt to write blocks for about 30 seconds or so.

Others answered most of this, but briefly...

> As suspected, there is no problem with the number of threads; the
> rxdebug command shows 0 threads used out of 11 while a freeze is
> happening.

rxdebug doesn't really give you a "X threads used out of Y" statistic
directly. If it said "0 calls waiting for a thread" and "11 threads are
idle", that doesn't mean 0 threads are occupied, it means that 0 threads
are stalled because there are no free threads. If that first number is
anything but 0, you are (perhaps just temporarily) over capacity. And
the second number is the number of threads not doing anything, not the
total number of threads.

> Some people suggested I blacklist clients that (apparently)
> don't respond to callback breaking.  But that won't work because
> (1) it could be that the campus wireless is blocking access
>     (not sure here)
> None of these are the fault of the client.

This first one may not be the fault of the client, but if it's the fault
of the network it still doesn't seem unreasonable to block that network
if it cannot be fixed.

Andrew Deason