lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexandre Rafalovitch <>
Subject Re: Miserable Experience Using Solr. Again.
Date Wed, 14 Sep 2016 00:46:42 GMT
On 14 September 2016 at 06:42, Aaron Greenspan
<> wrote:
> This is a potential solution, but not one I choose to pursue. For one thing, I am not
an idiot. I’ve managed Linux systems for about 18 years now and I’ve been programming
for 20. I have learned that I am rarely the best at anything, so sure, I fully admit that
there will always be others with much better skills than my own. But I’m an intelligent
person with experience in software trying to leave constructive feedback, and being told that
my feedback essentially reflects on my own stupidity kind of misses the point. I’m providing
feedback because things need fixing.

And that's the part I am confused about. Managing LInux systems is a
real pain in the ... Config files locations differ between
distributions. Upgrades are confusing, error messages are truly
critic. Debugging by dmesg and truss/strace is dark arts. Reading the
logs is nearly an art.

But with that experience, Zookeeper port is an lsof away. Or a ps away
if you want to read it from the parameters. Or a netstat away. Binding
anything to a local subnet is something of a standard firewall
operation. Couple of other things are output as part of "bin/solr
start --help" as well as part of log messages of running examples. I
understand the frustration. I truly do. And my own - committer - focus
is on improving beginner's experience (no, not looking for funding).
But the problems you list are definitely should be minor, not major
pain points to a Linux system administrator.

Solr is not like a WordPress. WordPress is designed for external users
and so is optimized for ease of used at expense of everything else.
Not correctness, not internal security (passwords are plain text,
plugins have access to everything), not ease of "good" development.
And WordPress is a _small_ product. It is a wrong comparison to the
point that apples and oranges are in the same category of fruit.

Solr is like a MySQL at least. And, frankly, changing a root password
in MySQL is also quite a pain. Or BEA weblogic (which I could...)

As to JIRA, the question was of a very high granularity on a use case
that is complicated and is not a bug/feature distinction. It also has
been discussed multiple times on the User list.


Anyway, I see one JIRA in this so far (Admin UI reclosing the log
message). If nobody else opens it, I will and have a go at it in the
next couple of days.

P.s. Aaron, perhaps you missed it with the digest mode, but I JUST
asked for feedback on an example reading group idea. I would have
expected you to jump on an opportunity as that would mean a direct
access to somebody contributing their contributor (mine!) time to
improve your understanding of Solr at whatever level of knowledge you
currently have. And - if that's not obvious - to see what kind of
things people find difficult to feed it into the next version of Solr.
Several people already showed interest, but you are not among them.
Hopefully, you will see that email and join us on your next read
through the digest. For easy reference, the survey link again is:

Newsletter and resources for Solr beginners and intermediates:

View raw message