Yeah, I think a jira issue or two was filed for it, but I didn't see
anything user facing go in. You can do it for queries by asking the
overseer to publish a DOWN state though. It won't drop indexing leadership
until you close the core, but it will prevent the temporary slow/hotspot
you get if you
Yeah ive mentioned it a number of times, but it’s absolutely something we
should have. Give up leadership, dont accept new replicas, dont accept new
requests. Maybe remove live_node!
- Houstob
On Thu, Sep 28, 2023 at 6:45 PM David Smiley
wrote:
> Somewhat related to this, we don't yet have a wa
I did some searching, but didn’t find the specific reference I was hoping to
share about how to encourage good practices on a large code set…. Maybe someone
else can chime in.
Basically, what you do is run additional checks on files that have been
changed, so that as we modify a java file, if i
Somewhat related to this, we don't yet have a way to signal to the cluster
that a node will soon be shut down, but has not shut down yet. With such
information, we'd prefer to not send queries there, and perhaps could even
begin shard/overseer leadership changes. This was mentioned somewhere in
c
When I see a class without any javadocs, especially a public top level
class, it's really sad and disappointing. My colleagues have noted the
same. I'd like to see a build mechanism to up our level of standards
here. Of course we have tons of existing such classes so a solution would
have to som
Hi,
It would be unfortunate to get this close and miss out on the Lucene
release. I would want to give it one more day, if things do not work out, I
will go ahead with the release.
best,
alex
On Thu, Sep 28, 2023 at 2:25 PM Jan Høydahl wrote:
> I see there is energy and traction into finaliz
I see there is energy and traction into finalizing the Lucene 9.8 integration,
so that's fine as long as we believe the risk to be low-ish.
Jan
> 28. sep. 2023 kl. 21:53 skrev Anshum Gupta :
>
> I don't like holding up releases, but it's nice to not have too many
> releases from an operational
That did require some changes around live node handeling, which is why a
different approach as you suggest would also be reasonable. You still do
want to solve for the original motivation of DOWN - stopping search traffic
to the node before things start closing.
Yeah, I took the DOWN state out all together in shutdown as its problematic
and effectively sugar for the user view of the cluster state - as far as
the system goes, if the ephemeral live node is gone, that node is down,
regardless of the replica state. There is some value in being able to
remove a
It's a shame to see that we seem to maybe need to use the underlying Apache
HttpClient in tests thanks to the presence of SSL randomization. Hossman,
do you have ideas on how we could not have such a dependency?
~ David Smiley
Apache Lucene/Solr Search Developer
http://www.linkedin.com/in/davidws
By default, Solr returns all metrics, of which there are a ton of them!
That's great for looking at metrics if you are exploring them for the first
time. But as you'll see in the ref guide, there are many filtering options
which I would consider to be mandatory if you are going to use metrics "for
I don't like holding up releases, but it's nice to not have too many
releases from an operational standpoint. If fixing Solr for the Lucene
upgrade will take time, I think it's worth the trade-off and moving forward
with the older version of Lucene.
On Thu, Sep 28, 2023 at 6:05 AM Jan Høydahl wro
You are right, sorry. We are also running a fork of Solr and in our
fork we have made some optimizations to avoid processing DOWNNODE
messages for nodes that only host PRS collections. Those optimizations
have not made it upstream at this point. I can take a look at
upstreaming those changes or som
Hello,
I have no strong views on Lucene 9.7 or Lucene 9.8 for Solr 9.4 but if it's not
too late to ask, I'd love to see the (small)
https://github.com/apache/solr/pull/1950 change for
https://issues.apache.org/jira/browse/SOLR-16994 considered for being in the
9.4 release.
Thanks,
Christine
I'd like to raise an issue that I'm having with Solr 9.3 that might make a
good case for getting another release (9.3.1 or 9.4) out soon.
I've had a 9.3 cluster of 180 nodes up and running for about a week. After
the weekend, I noticed that metrics from the cluster were failing to ship.
Upon inves
Can I propose we do NOT hold the 9.4 release to add the (too fresh) Lucene 9.8?
We can do a rapid 9.5 with any benefits reaped from the new Lucene version.
Jan
> 28. sep. 2023 kl. 00:36 skrev Alex Deparvu :
>
> Hi,
>
> Some updates to share on the upgrade.
> I have identified a cause for most
16 matches
Mail list logo
