[erlang-questions] Re: limited error_logger dump (vm crashes because of generating dump)

Ryan Zezeski <>
Wed Sep 22 02:34:43 CEST 2010

On Mon, Sep 20, 2010 at 9:35 AM, Max Lapshin <> wrote:

> It is a workaround, but it isn't applicable for me. I'd better remove
> all logging from erlang, than use structure, which is several times
> slower.
I've run into the same problem as you.  In my case, I have a web application
that has _very large_ files (> 700MB) posted to it.  My application reads
the entire request body as a binary and passes it around.  While developing
there would be causes where I would have a bad case match or some other
error .  This would cause SASL to attempt to print out the entire binary,
which, obviously, did not go so well.  It would cause a massive spike in
memory and typically I would have to kill the VM.

Eventually, I ironed out all the bugs (well, the ones I know about anyways
:) ), and it's since been a distant memory.  However, I remember thinking "I
wish I had a way to tell SASL to abbreviate the output."

The format_status callback looks useful for certain situations...I really
need to sit down and read through the canonical docs.

More information about the erlang-questions mailing list