You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cassandra.apache.org by Varun Barala <va...@gmail.com> on 2016/10/06 04:17:53 UTC
Counter global logical clock
Hi all Developers,
Recently I was restoring some sstables by changing their timestamp.
But in case of counter It's different there are few things I'd like to ask:-
* What is the maximum value of global logical clock? *Long.MAX_VALUE* ?
** what will happen If updates hit this MAX_VALUE value ?*
I read rule of merging of counter columns:-
- global + global = keep the shard with the highest logical clock
- global + local = keep the global one
- global + remote = keep the global one
- local + local = sum counts (and logical clocks)
- local + remote = keep the local one
- remote + remote = keep the shard with the highest logical clock
So in my case ::
** Can I change to 'set as global and set logical clock to maximum to make
sure it always win when merge' ? * How can we modify a counter column so it
will always win when merge?*
I'm using *ka* sstable version.
Thanking You!!
Regards,
Varun Barala
Fwd: Counter global logical clock
Posted by Varun Barala <va...@gmail.com>.
Hi all Users,
Recently I was restoring some sstables by changing their timestamp.
But in case of counter It's different there are few things I'd like to ask:-
* What is the maximum value of global logical clock? *Long.MAX_VALUE* ?
** what will happen If updates hit this MAX_VALUE value ?*
I read rule of merging of counter columns:-
- global + global = keep the shard with the highest logical clock
- global + local = keep the global one
- global + remote = keep the global one
- local + local = sum counts (and logical clocks)
- local + remote = keep the local one
- remote + remote = keep the shard with the highest logical clock
So in my case ::
** Can I change to 'set as global and set logical clock to maximum to make
sure it always win when merge' ? * How can we modify a counter column so it
will always win when merge?*
I'm using *ka* sstable version.
Thanking You!!
Regards,
Varun Barala