I understand - I was actually trying to use the containerized Cassandra, so upgrading in place doesn't make sense.
I guess I'll wait until that's better supported. On Wed, Jul 1, 2015 at 12:14 PM, Jonathan Haddad <j...@jonhaddad.com> wrote: > Adding new nodes of a different version is asking for trouble, as others > have said. I don't know if this particular version bump would expect any > issues, but why risk it? If you want to upgrade, do it to the nodes in > place. Don't bootstrap new nodes, don't run repair, don't remove nodes. > My 2 cents. > > On Wed, Jul 1, 2015 at 11:59 AM David Aronchick <aronch...@gmail.com> > wrote: > >> I mean add an additional two nodes to my cluster and pointing them at the >> other nodes in the cluster, to handle data migration. >> >> On Wed, Jul 1, 2015 at 11:40 AM, Jonathan Haddad <j...@jonhaddad.com> >> wrote: >> >>> When you say add 2 nodes, do you mean bootstrap, or upgrade in place? >>> >>> On Wed, Jul 1, 2015 at 11:37 AM David Aronchick <aronch...@gmail.com> >>> wrote: >>> >>>> This helps - so let me understand: >>>> >>>> Starting point: >>>> - 4 nodes running 2.1.4 >>>> - System is healthy >>>> >>>> Decide to upgrade: >>>> - Add 2 nodes running 2.1.5 >>>> - Run nodetool upgradestable >>>> - Wait until system is healthy >>>> - Stop 2 nodes running 2.1.4 >>>> - Run nodetool upgradestable >>>> - Add 2 nodes running 2.1.5 >>>> - Run nodetool upgradestable >>>> - Stop 2 nodes running 2.1.4 >>>> - Run nodetool upgradestable >>>> >>>> Finished? >>>> >>>> On Wed, Jul 1, 2015 at 1:49 AM, Alain RODRIGUEZ <arodr...@gmail.com> >>>> wrote: >>>> >>>>> Basically, when you add nodes, add them on the correct version to >>>>> avoid schema / network issues in your streams. Also, try to update all the >>>>> node using rolling restarts in a reduced time frame after stopping >>>>> repairs, >>>>> with all the nodes up, etc. >>>>> >>>>> You must have a healthy cluster before performing an upgrade. >>>>> >>>>> Remember running a nodetool upgradesstable after any upgrade (if not >>>>> needed it will end fast, so I would run it always as a best practice, just >>>>> in case). >>>>> >>>>> C*heers, >>>>> >>>>> Alain >>>>> >>>>> 2015-07-01 2:16 GMT+02:00 David Aronchick <aronch...@gmail.com>: >>>>> >>>>>> That is a GREAT lead! So it looks like I can't add a few nodes to the >>>>>> cluster of the new version, have it settle down, and then upgrade the >>>>>> rest? >>>>>> >>>>>> On Tue, Jun 30, 2015 at 11:58 AM, Alain RODRIGUEZ <arodr...@gmail.com >>>>>> > wrote: >>>>>> >>>>>>> "Would it matter that I'm mixing cassandra versions? " >>>>>>> >>>>>>> From: >>>>>>> http://docs.datastax.com/en/upgrade/doc/upgrade/datastax_enterprise/upgrdLim.html >>>>>>> >>>>>>> "General upgrade limitations¶ >>>>>>> >>>>>>> Do not run nodetool repair. >>>>>>> Do not enable new features. >>>>>>> Do not issue these types of queries during a rolling restart: DDL, >>>>>>> TRUNCATE >>>>>>> *During upgrades, the nodes on different versions show a schema >>>>>>> disagreement*." >>>>>>> >>>>>>> I think this is a good lead. >>>>>>> >>>>>>> C*heers, >>>>>>> >>>>>>> Alain >>>>>>> >>>>>>> 2015-06-30 20:22 GMT+02:00 David Aronchick <aronch...@gmail.com>: >>>>>>> >>>>>>>> I appreciate the thoughts! My issue is that it seems to work >>>>>>>> perfectly, until the node goes away. Would it matter that I'm mixing >>>>>>>> cassandra versions? (2.1.4 and 2.1.5)? >>>>>>>> >>>>>>>> On Tue, Jun 30, 2015 at 5:23 AM, Alain RODRIGUEZ < >>>>>>>> arodr...@gmail.com> wrote: >>>>>>>> >>>>>>>>> Hi David ? >>>>>>>>> >>>>>>>>> What does a "nodetool describecluster" output look like ? >>>>>>>>> >>>>>>>>> My guess is you might be having a schema version >>>>>>>>> desynchronisation. If you see a node with different schema version you >>>>>>>>> might want to try a "nodetool resetlocal*schema* - Reset >>>>>>>>> node's local *schema* and resync" >>>>>>>>> >>>>>>>>> You asked for any thoughts, this is a thought, not sure if it will >>>>>>>>> help, I hope so. >>>>>>>>> >>>>>>>>> C*heers, >>>>>>>>> >>>>>>>>> Alain >>>>>>>>> >>>>>>>>> 2015-06-30 1:44 GMT+02:00 Robert Coli <rc...@eventbrite.com>: >>>>>>>>> >>>>>>>>>> On Mon, Jun 29, 2015 at 2:43 PM, David Aronchick < >>>>>>>>>> aronch...@gmail.com> wrote: >>>>>>>>>> >>>>>>>>>>> Ping--- any thoughts here? >>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> I don't have any thoughts on your specific issue at this time, >>>>>>>>>> but FWIW #cassandra on freenode is sometimes a better forum for >>>>>>>>>> interactive >>>>>>>>>> debugging of operational edge cases. >>>>>>>>>> >>>>>>>>>> =Rob >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>> >>>> >>
