I think most of you know this already, but for the record, a new xrootd
has been cut and is available for download.
http://www.slac.stanford.edu/exp/lsst/qserv/download/current/xrootd-qsPatch4.tar.gz
(I've also put up the qserv branch for xrootd up on github:
https://github.com/wangd/xrootd You can get the qsPatch4 tarball from
there too.)
Some of the qserv specializations may make it into xrootd, but nothing's
for certain yet.
Due to API changes in xrootd, you must pick a version of qserv that has
been ported. The code porting happened in commit 52b459437579 , and the
master branch now includes this commit. There are two tags worth mentioning.
u/danielw/lst2013_2 : "new parser" + xrootd 3.2.8 port
u/danielw/lst2013_3 : previous + objectId + near-neighbor
FYI, I would like to tag 0.6.0 as the version we use for generating
benchmark numbers in the large scale test. We won't know what/when to
tag until we have benchmark numbers that we are satisfied with.
-Daniel
On 07/18/2013 05:19 PM, Daniel L. Wang wrote:
> Hi Douglas,
>
> FYI, Andy identified a part of the xrootd code that might be causing the
> problem. That code was fixed in the mainline xrootd. I will prepare a
> new xrootd for you and let you know when I have a new tarball.
>
> If we can still reproduce the problem, then Andy will take another look.
>
> -Daniel
>
> ########################################################################
> Use REPLY-ALL to reply to list
>
> To unsubscribe from the QSERV-L list, click the following link:
> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the QSERV-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1
|