cassandra-pr mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <>
Subject [GitHub] [cassandra] yifan-c commented on issue #428: CASSANDRA-15505
Date Thu, 23 Jan 2020 06:02:00 GMT
yifan-c commented on issue #428: CASSANDRA-15505 
   1. The composite `IMatcher` seems complicated. It allows end-users to add multiple filters
that match the same condition, but the evaluation stops at the first one that returns `true`,
i.e. drop the message. For instance, I can have `filter1: (1, 2, m) -> add_delay; return
false`, `filter2:(1, 2, m) -> true`, and `filter3: (1, 2, m) -> add_m_to_list; return
true` under `cluster.filters()`. When matching, the evaluation stops at filter2, and ignores
filter3. so the code that `add_m_to_list` is not performed. 
   2. Another thought regarding the goal here. Essentially, it is matching the condition and
do interception upon matching. The interception runs **optional** custom code and at the end
makes the decision to either return **DROP** or **PASS**. _From this perspective, drop is
just a special form of intercept._  
       a) If allowing multiple filters to match the same condition, there will be multiple
decisions made from the corresponding filter. Ideally, there could be a resolver to come up
with a final decision. But for simplicity, we can have rule to say **DROP** over **PASS**.
       b) If only allowing single filter, the situation is simple.

This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:

With regards,
Apache Git Services

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message