Hi Noble, Great stuff, no problem, I really think the Solr development team is excellent and takes pride in delivering high quality software!
And we're going into production with a brand new Solr based system in a few weeks as well, so I'm really happy that this is fixed now. Bye, Jaco. 2009/1/24 Noble Paul നോബിള് नोब्ळ् <noble.p...@gmail.com> > hi Jaco, > We owe you a bing THANK YOU. > > We were planning to roll out this feature into production in the next > week or so. Our internal testing could not find this out. > > > > --Noble > > > On Fri, Jan 23, 2009 at 6:36 PM, Jaco <jdevr...@gmail.com> wrote: > > Hi, > > > > I have tested this as well, looking fine! Both issues are indeed fixed, > and > > the index directory of the slaves gets cleaned up nicely. I will apply > the > > changes to all systems I've got running and report back in this thread in > > case any issues are found. > > > > Thanks for the very fast help! I usually need much, much more patience > with > > commercial software vendors.. > > > > Cheers, > > > > Jaco. > > > > > > > 2009/1/23 Noble Paul നോബിള് नोब्ळ् <noble.p...@gmail.com> > > > >> I have opened an issue to track this > >> https://issues.apache.org/jira/browse/SOLR-978 > >> > >> On Fri, Jan 23, 2009 at 5:22 PM, Noble Paul നോബിള് नोब्ळ् > >> <noble.p...@gmail.com> wrote: > >> > I tested with the patch > >> > it has solved both the issues > >> > > >> > On Fri, Jan 23, 2009 at 5:00 PM, Shalin Shekhar Mangar > >> > <shalinman...@gmail.com> wrote: > >> >> > >> >> > >> >> On Fri, Jan 23, 2009 at 2:12 PM, Jaco <jdevr...@gmail.com> wrote: > >> >>> > >> >>> Hi, > >> >>> > >> >>> I applied the patch and did some more tests - also adding some > >> LOG.info() > >> >>> calls in delTree to see if it actually gets invoked > (LOG.info("START: > >> >>> delTree: "+dir.getName()); at the start of that method). I don't see > >> any > >> >>> entries of this showing up in the log file at all, so it looks like > >> >>> delTree > >> >>> doesn't get invoked at all. > >> >>> > >> >>> To be sure, explaining the issue to prevent misunderstanding: > >> >>> - The number of files in the index directory on the slave keeps > >> increasing > >> >>> (in my very small test core, there are now 128 files in the slave's > >> index > >> >>> directory, and only 73 files in the master's index directory) > >> >>> - The directories index.xxxxx are still there after replication, but > >> they > >> >>> are empty > >> >>> > >> >>> Are there any other things I can do check, or more info that I can > >> provide > >> >>> to help fix this? > >> >> > >> >> The problem is that when we do a commit on the slave after > replication > >> is > >> >> done. The commit does not re-open the IndexWriter. Therefore, the > >> deletion > >> >> policy does not take affect and older files are left as is. This can > >> keep on > >> >> building up. The only solution is to re-open the index writer. > >> >> > >> >> I think the attached patch can solve this problem. Can you try this > and > >> let > >> >> us know? Thank you for your patience. > >> >> > >> >> -- > >> >> Regards, > >> >> Shalin Shekhar Mangar. > >> >> > >> > > >> > > >> > > >> > -- > >> > --Noble Paul > >> > > >> > >> > >> > >> -- > >> --Noble Paul > >> > > > > > > -- > --Noble Paul >
