Yes, that works. But: you should ensure that you use the keys argument correctly, passing any keys in as arguments. If the script uses multiple keys, you should use { / } to ensure that all the keys are in the same slot.
Marc
--
You received this message because you are subscribed to the Google Groups "Redis DB" group.
To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+u...@googlegroups.com.
To post to this group, send email to redi...@googlegroups.com.
Visit this group at http://groups.google.com/group/redis-db.
For more options, visit https://groups.google.com/d/optout.
By "slot", I mean: in cluster, each key is hashed and modulo applied to produce an integer in a given range (0-16K). This is the "slot". The hash algorithm is deterministic, so the same key will always produce the same slot. Each server owns a pool of slots. In a simple 4 server case, they might each own 4k slots. Because the slots allocation can change (manually, not currently automatically - although that would be an awesome v3.something feature), your apps can't really rely on which slots are adjacent. But to help with your scenario, you can use "hash tags". Meaning: if you know that (for example) you want to talk multi-key about data about a customer, the rather than "/cust/123/balance" and "/cust/123/recent" you can use "{/cust/123}/balance" and "{/cust/123}/recent". The addition of braces means the the hash algorithm for both is now based on just "/cust/123". This means they will always hash to the same slot, so **now** you can use multi-key operations including scripts