Re: Resources on Using Single Vnode in Cassandra

Thanks a lot, Jordan, Jeff, Abe, guo and Jon! This is really helpful. On Wed, Oct 9, 2024 at 8:59 AM Jon Haddad wrote: > I've worked with a few hundred teams now, including the major ones that > used single token (Apple, Netflix, Spotify), and pretty much all the rest > used some form of vnodes.

Re: Tombstone Generation in Cassandra 4.1.3 Despite No Update/Delete Operations

The easiest option here, though still unpleasant, is sstabledump to json and look at the tombstoneUsually when this happens it’s because something unexpected is happening - actually writing nulls or doing deleted or weird short TTLs without realizing it Dump the sstable and look, it’ll be faster th

Re: Tombstone Generation in Cassandra 4.1.3 Despite No Update/Delete Operations

Overwriting non collections does not generate tombstones on compaction. — Jon Haddad Rustyrazorblade Consulting rustyrazorblade.com On Wed, Oct 9, 2024 at 9:57 AM James Shaw wrote: > Hi, Naman: > How does the client side load large amounts of data ? Most > likely it has a retry polic

Re: Tombstone Generation in Cassandra 4.1.3 Despite No Update/Delete Operations

Hi, Naman: How does the client side load large amounts of data ? Most likely it has a retry policy, either by tool's default or by setting. If it has a retry policy, will have duplicated records, after compaction, will have tombstones. Another possibility, client set ttl, it will overri

Re: Resources on Using Single Vnode in Cassandra

I've worked with a few hundred teams now, including the major ones that used single token (Apple, Netflix, Spotify), and pretty much all the rest used some form of vnodes. Jeff did a good job of summarizing the tradeoffs and I don't have anything to add. I would never, ever, recommend > 4 tokens.

Survey closes tomorrow!

The annual Cassandra survey closes tomorrow. Thanks to everyone who has filled out the survey and if that isn't you, don't wait! *https://forms.gle/wFbt7SAzjFJupqef8 *