Strange exception when storing heavy data in cassandra 2.0.0...

[Krishna Chaitanya <bn...@gmail.com>]

Hello,
         I am newbie to the  Cassandra world. I am currently using
Cassandra 2.0.0 with thrift 0.8.0 for storing netflow packets using
libQtCassandra library. Currently, I am generating about 1000 netflows/sec
and store\ing them into the database. The program is crashing with the an
exception on the lines of--- "what() :frame size reads negative value". I
re ran the program a few times with same exception after running for a few
seconds. But, after few times a new default TException occured and
Cassandra is now failing to even to startup. When I try kill and start it
up again, the exception says--" exception encountered during startup:
unfinished compactions reference missing sstables.This should never happen
since compactions are marked finished before we start removing the old
sstables".
                                     Is this a known issue because it did
not occur when we were using Cassandra-1.2.6 and previous versions with
pycassa library for accessing the database. I am just running a single
cassandra node with default settings. How can I avoid this exception and
also is there any way in which I can get back my node to running state even
if it means reinstalling cassandra???  Thank You in advance for any help.

-- 
Regards,
BNSK*.
*

Re: Strange exception when storing heavy data in cassandra 2.0.0...

[Robert Coli <rc...@eventbrite.com>]

On Fri, Nov 1, 2013 at 10:29 PM, Krishna Chaitanya
<bn...@gmail.com>wrote:

>          I am newbie to the  Cassandra world. I am currently using
> Cassandra 2.0.0 with thrift 0.8.0 for storing netflow packets using
> libQtCassandra library.  ... Is this a known issue because it did not occur
> when we were using Cassandra-1.2.6 and previous versions with pycassa
> library for accessing the database. ... How can I avoid this exception and
> also is there any way in which I can get back my node to running state even
> if it means reinstalling cassandra???  Thank You in advance for any help.
>

https://engineering.eventbrite.com/what-version-of-cassandra-should-i-run/

You could try upgrading to 2.0.2, or downgrading back to 1.2.6. The former
is likely to be easier/more possible than the latter, but may or may not
resolve your problem. The latter may require a dump/load or reload of your
cluster's data, if sstables have been upgraded to a 2.0 version.

note : I have not looked for your particular issue in the apache JIRA,
there may be value to you doing so.

=Rob