-Re: Hbase 2077 status?
Himanshu Vashishtha 2011-05-19, 23:13
Hey, I am also facing the similar issue but in a different context (related
to coprocessor), but that's a totally different use case (still need to look
in to it).
But in this case, will reducing the scanner cache size helps (at least
temporarily)? In a case when scanner is busy collecting/computing rows just
to meet up the cache limit expectations.
On Thu, May 19, 2011 at 5:09 PM, Vidhyashankar Venkataraman <
[EMAIL PROTECTED]> wrote:
> >> maybe you could bump the timeouts high enough so that you don't
> >> hit the issue at all?
> Don't you think setting a high timeout might be a little ad hoc? This might
> just work except that it could lead to a really long delay during cases when
> there should be a timeout. Also we have non-homogeneous data, the timeout
> setting may have to be set differently for different access sizes.
> I can go through the patch and ping you guys back.
> On 5/19/11 3:42 PM, "Jean-Daniel Cryans" <[EMAIL PROTECTED]> wrote:
> The latest patch would need some more work, I did more than what's
> really required.
> If you are really taking more than a minute to do a single next()
> call, maybe you could bump the timeouts high enough so that you don't
> hit the issue at all? The default is pretty arbitrary.
> On Thu, May 19, 2011 at 3:37 PM, Vidhyashankar Venkataraman
> <[EMAIL PROTECTED]> wrote:
> > I had spoken a while back about this problem (clients timing out when
> scanners do not return with a row yet: search for "A possible bug in the
> scanner. "
> > I am trying to fix the problem in the next few days: our system is a
> little crippled without the fix (We use filters in scans and the bug crops
> up during filters that return sparse sets.).
> > JD, can you let me know the status of the recentmost patch attached
> > Thank you
> > Vidhya