Re: Create View of Existing HBase table

2017-06-19 Thread Randy Hu
You are very welcome. Glad it helped. For people using Phoenix with existing HBase tables for a nice SQL interface, they would very likely experience the same issue again. Wondering if the Phoenix team can change the strategy to use maximum timestamp for select on views. This would not cause probl

Re: Best strategy for UPSERT SELECT in large table

2017-06-19 Thread Jonathan Leech
I think you could add additional pk columns, but not change or remove existing ones. > On Jun 19, 2017, at 11:58 AM, Michael Young wrote: > > Regarding your idea to use the snapshot/restore method (with a new name). Is > it possible to add a PK column with that approach? For example, if I wa

Re: Cant run map-reduce index builder because my view/idx is lower case

2017-06-19 Thread Batyrshin Alexander
Hello again, Could you, please, help me to run map-reduce for indexing view with lower-case name? Here is my test try on Phoenix-4.8.2: CREATE TABLE "table" ( c1 varchar, c2 varchar, c3 varchar CONSTRAINT pk PRIMARY KEY (c1,c2,c3) ) CREATE VIEW "table_view" AS SELECT * FROM "ta

Re: Best strategy for UPSERT SELECT in large table

2017-06-19 Thread Michael Young
Regarding your idea to use the snapshot/restore method (with a new name). Is it possible to add a PK column with that approach? For example, if I wanted to change a PK column type from VARCHAR to FLOAT, is this possible? On Sun, Jun 18, 2017 at 10:50 AM, Jonathan Leech wrote: > Also, if you'r

Re: Create View of Existing HBase table

2017-06-19 Thread M. Aaron Bossert
Thanks for that pointer! When I got to work this morning, I redid my ingest with all timestamps in milliseconds and the view populate nicely. Thanks again! Sent from my iPhone > On Jun 18, 2017, at 01:12, Randy Hu wrote: > > That's likely the problem. The value in HBase/Java is in millisecon