These are chat archives for RBMHTechnology/eventuate

15th
Sep 2015
Ian Clegg
@ianclegg
Sep 15 2015 13:57
@krasserm Have you got any plans to implement a Last Writer Wins Register?
Ian Clegg
@ianclegg
Sep 15 2015 14:06
@krasserm Actually the implementation is pretty trivial, but i'm new to the code, would you accept a pull and review it?
Martin Krasser
@krasserm
Sep 15 2015 14:06
@ianclegg It's not high prio for me at the moment but I think that's easy to implement. Instead of tracking multiple values like in MV-Register, apply an user-defined function that orders concurrent updates for determining the "last" writer (whatever that means for your application: higher wall-clock timestamp and in case of equal timestamps, the writer with the higher emitter id, or whatever ...). Would love to see a pull request :smiley:. Are you interested?
Uh, this was a concurrent update :smile:
Ian Clegg
@ianclegg
Sep 15 2015 14:07
@krasserm yer, thats what I was think - i'll submit a PR - cheers
@krasserm I need an OR Map too - thats document online in 'A comprehensive study of Convergent and Commutative
Replicated Data Types'
@krasserm I might need some more help with that
Martin Krasser
@krasserm
Sep 15 2015 14:09
@ianclegg great, thanks! and don't forget to sign the CLA
This message was deleted
@ianclegg I'm happy to help
Martin Krasser
@krasserm
Sep 15 2015 14:15
@ianclegg with OR-Map do you mean spec 21 in the paper?
Ian Clegg
@ianclegg
Sep 15 2015 14:27
yes Spec 21
Martin Krasser
@krasserm
Sep 15 2015 14:31
That would be a great addition!
Ian Clegg
@ianclegg
Sep 15 2015 16:01
@krasserm I opened a PR, but the build failed on BasicReplicationSpecCassandraMultiJvmNode1. A barrier timed out in the test with a cassandra exception: All host(s) tried for query failed (tried: /127.0.0.1:9142 (com.datastax.driver.core.TransportException: [/127.0.0.1:9142] Cannot connect)
the commit di'dnt touch the code in the test path and the tests passed locally?
Martin Krasser
@krasserm
Sep 15 2015 16:08
@ianclegg that's an issue with cassandra-unit that comes up in approx. 1 of 10 test runs and is not related to your PR. I didn't investigate further so far because I also cannot produce it locally ...
Ian Clegg
@ianclegg
Sep 15 2015 16:13
@krasserm much appreciated - no worries, i'll take a deeper look at the intermittent error tomorrow