[erlang-questions] Problems with Mnesia and failure situations

[Teemu Antti-Poika]

On Wed, Apr 22, 2009 at 6:56 PM, Hakan Mattsson <hakan@REDACTED> wrote:
> On Wed, 22 Apr 2009, Teemu Antti-Poika wrote:
>
>> Apparently mnesia timing out on node A when I unplug/shutdown node B
>> is a necessary evil?
>
> What is timing out? Mnesia on node A should just reconfigure itself
> and recover those transactions that involved node B.

Mnesia itself survived, so my worry below is not justified (data is
safe). In my application using mnesia the mnesia transactions timed
out, eventually causing the supervisors to react. But I guess that is
to be expected.

>> Since my process controlling the mnesia bound data is timing out on
>> remote communications, it cannot react to orders to dump tables on
>> disk. Eventually it is forcefully shut down by the supervisor, causing
>> RAM-resident data be lost, I think?
>
> Why do you dump the tables manually, instead of using disc_copies?
> If you use disc_copies, Mnesia will keep the table updated on disk for you.

That is/was just a premature optimization (minimizing disk i/o). I've
changed the tables to disc_copies without observable change in
performance.

Teemu A-P