the relationship between metadata size and heap size

[Ian Young]

> On 14 Sep 2015, at 17:19, Cantor, Scott <cantor.2 at osu.edu> wrote:
> 
>> I was quite surprised to find that ScheduledThreadPoolExecutor is *defined* to do this by default, for example.
> 
> It's that "catch silently" thing I'm having a hard time with. If that's what their code does, they need a good slap.

I'd have to look this up again to be sure, but my recollection is that they catch (Throwable) in the job, and pass any exception to a hook you can define but which is not present by default. So, by default, exceptions in jobs are just silently swallowed. Moreover, this causes any scheduled periodic repeat to be abandoned.

Yes, I think that design deserves a good slap but it's hardly unique within the Java API in this regard.

> I'm not saying "container catches it" is a good thing, just saying *we* better not *ever* silently swallow errors and if we are, we need a slap too.

+1

    -- Ian




-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5250 bytes
Desc: not available
URL: <http://shibboleth.net/pipermail/users/attachments/20150914/51b9d047/attachment.p7s>