LIMITATIONS OF CODE THAT CAN BE WRITTEN IN HBASE COPROCESSORS

[chaitanya eswar <ch...@gmail.com>]

Hi,

I am trying to  do this,
When a put is done in one table  i have to store those puts using a static
variable in coprocessor and whenever client issues flush i have to update
the other table.
When i try to do something new coprocessor isn't getting deployed and
sometimes even the master isn't getting started.
I wish to know whether there are any limitations to functionality of
coprocessors

Re: LIMITATIONS OF CODE THAT CAN BE WRITTEN IN HBASE COPROCESSORS

[Michael Segel <mi...@hotmail.com>]

I'm not sure of your question... 

Is it that when you wish to push out a new version of your coprocessor, that you have issues because the older version is being executed instead? 
If that's the case, did you try to run a rolling restart on the region servers? 

On Feb 27, 2013, at 11:55 PM, chaitanya eswar <ch...@gmail.com> wrote:

> Hi,
> 
> I am trying to  do this,
> When a put is done in one table  i have to store those puts using a static
> variable in coprocessor and whenever client issues flush i have to update
> the other table.
> When i try to do something new coprocessor isn't getting deployed and
> sometimes even the master isn't getting started.
> I wish to know whether there are any limitations to functionality of
> coprocessors

Re: LIMITATIONS OF CODE THAT CAN BE WRITTEN IN HBASE COPROCESSORS

[Andrew Purtell <ap...@apache.org>]

> > When a put is done in one table  i have to store those puts using a
static
> > variable in coprocessor

Let's start here. Static variable?


On Thu, Feb 28, 2013 at 9:32 AM, Ted Yu <yu...@gmail.com> wrote:

> bq. whenever client issues flush i have to update the other table.
>
> Are you updating second table in coprocessor ? That is anti-pattern
> according to Andrew Purtell.
>
> If you can show us some log (region server, master), that would help us
> troubleshoot.
>
> BTW there is no need to capitalize every letter in the email subject.
>
> Cheers
>
> On Wed, Feb 27, 2013 at 9:55 PM, chaitanya eswar <
> chaitanyaeswar.chaitanya@gmail.com> wrote:
>
> > Hi,
> >
> > I am trying to  do this,
> > When a put is done in one table  i have to store those puts using a
> static
> > variable in coprocessor and whenever client issues flush i have to update
> > the other table.
> > When i try to do something new coprocessor isn't getting deployed and
> > sometimes even the master isn't getting started.
> > I wish to know whether there are any limitations to functionality of
> > coprocessors
> >
>



-- 
Best regards,

   - Andy

Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)

Re: LIMITATIONS OF CODE THAT CAN BE WRITTEN IN HBASE COPROCESSORS

[Ted Yu <yu...@gmail.com>]

bq. whenever client issues flush i have to update the other table.

Are you updating second table in coprocessor ? That is anti-pattern
according to Andrew Purtell.

If you can show us some log (region server, master), that would help us
troubleshoot.

BTW there is no need to capitalize every letter in the email subject.

Cheers

On Wed, Feb 27, 2013 at 9:55 PM, chaitanya eswar <
chaitanyaeswar.chaitanya@gmail.com> wrote:

> Hi,
>
> I am trying to  do this,
> When a put is done in one table  i have to store those puts using a static
> variable in coprocessor and whenever client issues flush i have to update
> the other table.
> When i try to do something new coprocessor isn't getting deployed and
> sometimes even the master isn't getting started.
> I wish to know whether there are any limitations to functionality of
> coprocessors
>