[akka-user] Supervision: Who just died

2015-11-11 Thread Arne Claassen
The one thing I keep struggling with is the opaqueness of failures in Akka, i.e. that all the conditions at failure time are not available to either the actor or the supervisor. What I just came across was a failure during creation. The supervisor saw it and logged it and I found where it died

Re: [akka-user] Capturing the message that killed an actor

2015-10-17 Thread Arne Claassen
e for?) > > -- > Cheers, > √ > On 15 Oct 2015 22:44, "Arne Claassen" <sde...@gmail.com > > wrote: > >> So I came up with this workaround: >> >> object SafeReceive { >> def apply(receive: Receive)(recover: Any => PartialFunction[Throwa

Re: [akka-user] Capturing the message that killed an actor

2015-10-15 Thread Arne Claassen
carnation of the actor will pick up where the dead actor left > off (minus the actor that caused the failure). > > On Thursday, October 15, 2015 at 12:31:39 PM UTC-6, Arne Claassen wrote: >> >> The scenario in which this came up was one of a bug causing the death. >> I.e. it wa

Re: [akka-user] Capturing the message that killed an actor

2015-10-15 Thread Arne Claassen
ctober 15, 2015 at 1:25:13 PM UTC-7, Arne Claassen wrote: > > Actually according to those docs, the supervisor stopping does cause the > message to be dropped. The message is taken from the mailbox, causes an > exception and then the supervisor stops the actor, never returning the >

[akka-user] Capturing the message that killed an actor

2015-10-15 Thread Arne Claassen
I'm trying to respond to the sender of a message that killed my actor. I know I get the message if in preRestart if I chose to restart, but if my supervisorStrategy decided to stop (because of max retries or some other reason), there doesn't seem to be any way to get at the message. The actor

Re: [akka-user] Capturing the message that killed an actor

2015-10-15 Thread Arne Claassen
I.e. create a temporary actor to deal with the dangerous message and then > it can send a message to sender in postStop? > > On Thu, Oct 15, 2015 at 7:38 PM, Arne Claassen <sde...@gmail.com > > wrote: > >> I'm trying to respond to the sender of a message that killed

Re: [akka-user] Best strategy for zero-downtime cluster upgrades with changed messages

2015-08-07 Thread Arne Claassen
, so accidental changes of contracts were more obvious than with Actor messaging. thanks, arne On Thursday, August 6, 2015 at 5:42:31 AM UTC-7, Brice Figureau wrote: On Wed, 2015-08-05 at 11:04 -0700, Arne Claassen wrote: I'm trying to figure out what the best strategy for making changes