Does SolrCloud's notion of a "collection", which appears to use cores,
override normal multi-core usage for building an offline index and
quickly swapping it into production? Some of the features in SolrCloud
look useful, if it's still possible to exert manual control over cores
and shards. Are there plans to merge the cloud branch into trunk, and
if so, will this happen before 1.5 comes out?
I get the impression that version 1.4 is being quickly left in the dust
by the 1.5 development effort. As far as I can tell nobody is trying to
even put a ballpark on when 1.5 will be frozen or released. Is it
recommended at this point to use 1.5 from trunk over 1.4? Is there a
1.4.1 coming soon that fixes the known bugs?
From what I've read so far, it seems that it should be possible to
hardlink the contents of the index directory to another core, add/update
records, optionally re-optimize it, warm it, and then swap it into
production. Is this accurate?
A minor thing - is there any way to make Thunderbird 3 stop turning
sequences of a caret followed by a number into nice-looking exponents?
For weeks I've been wondering what everyone meant by putting exponents
in their config descriptions, then the other day I got a look at the
actual text of a message and the light went on.
Thanks,
Shawn
- A bunch of questions Shawn Heisey
-
