Elasticsearch 5.0.0-alpha1 --> 5.0.0-alpha5

Hello,

I have a cluster in 5.0.0-alpha1. Are indexes created with this version compatible 5.0.0-alpha5? can I update without loosing my data?

Thank you!

Maybe but we don't guarantee it. I know for a fact that the wire protocol has changed so you'll need to shut down the whole cluster then upgrade then start it up again. The on disk structure might be the same depending on the features you've used but it probably isn't.

1 Like

I did a yum update...

016-09-02 17:39:37,351][INFO ][node                     ] [elk] initializing ...
[2016-09-02 17:39:37,418][INFO ][env                      ] [elk] using [1] data paths, mounts [[/ (rootfs)]], net usable_space [1.4tb], net total_space [1.9tb], spins? [unknown], types [rootfs]
[2016-09-02 17:39:37,418][INFO ][env                      ] [elk] heap size [19.9gb], compressed ordinary object pointers [true]
[2016-09-02 17:39:38,144][INFO ][node                     ] [elk] version[5.0.0-alpha5], pid[11148], build[d327dd4/2016-08-04T08:59:39.568Z], OS[Linux/3.10.0-229.14.1.el7.x86_64/amd64], JVM[Oracle Corporation/Java HotSpot(TM) 64-Bit Server VM/1.8.0_65/25.65-b01]
[2016-09-02 17:39:38,899][INFO ][io.netty.util.internal.PlatformDependent] Your platform does not provide complete low-level API for accessing direct buffers reliably. Unless explicitly requested, heap buffer will always be preferred to avoid potential system unstability.
[2016-09-02 17:39:38,903][INFO ][plugins                  ] [elk] loaded module [aggs-matrix-stats]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [ingest-common]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [lang-expression]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [lang-groovy]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [lang-mustache]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [lang-painless]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [percolator]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [reindex]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [transport-netty3]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] loaded module [transport-netty4]
[2016-09-02 17:39:38,904][INFO ][plugins                  ] [elk] no plugins loaded

and it hangs at no plugins loaded....

It is hanging after no plugins loaded. If you nuke the old data directory does it start? If so then that is ok - not great, but, again, we don't offer on disk compatibility between alpha versions. Same for betas. Only released versions have on disk compatibility and we're quite careful to keep it.

it is starting if I change the cluster name, so yes..