For COPY TO you can try increasing the page timeout or decreasing the page
size:
PAGETIMEOUT=10 - the page timeout in seconds for fetching results
PAGESIZE='1000' - the page size for fetching results
You can pass these options to the COPY command by adding "WITH
PAGETIMEOUT=1000;",
Hi,
already that copy to might not be the best way to do this. I’ll write a
small spark job.
Thanks
2016-05-10 10:36 GMT+02:00 Carlos Rolo :
> Hello,
>
> That is a lot of data to do an "COPY TO.
>
> If you want a fast way to export, and you're fine with Java, you can use
> Cassandra SSTableRead
Hello,
That is a lot of data to do an "COPY TO.
If you want a fast way to export, and you're fine with Java, you can use
Cassandra SSTableReader classes to read the sstables directly. Spark also
works.
Regards,
Carlos Juzarte Rolo
Cassandra Consultant / Datastax Certified Architect / Cassandra
sry, sent early..
more errors:
/export.cql:9:Error for (4549395184516451179, 4560441269902768904):
NoHostAvailable - ('Unable to complete the operation against any
hosts', {: ConnectionException('Host has
been marked down or removed',)}) (will try again later attempt 1 of 5)
/export.cql:9:Error f
Hi,
i try to export data of a table (~15GB) using the cqlsh copy to. It fails
with „no host available“. If I try it with a smaller table everything works
fine.
The statistics of the big table:
SSTable count: 81
Space used (live): 14102945336
Space
