John, can you explain what the usecase for such a new API is? I don't
see a user of the API in your code. Is there a query you can optimize
with this or what is the reasoning behind this change? I personally
think it's quite invasive to add this information and there must be a
good reason to add th
OK, I’ll do it against master. We’d love to see it in an 8.x release, though.
This would protect against the major Solr outages we’ve had in the past
few years.
A question for someone who knows the Java metrics stuff better than I do:
Is it OK to use com.sun.management.OperatingSystemMXBean inste
Jason, I know you've seen SOLR-15051 (Shared storage -- BlobDirectory), but
perhaps not a part of the linked document:
https://docs.google.com/document/d/1kjQPK80sLiZJyRjek_Edhokfc5q9S3ISvFRM2_YeL8M/edit#
"ZK Data: Listings" that discusses an aspect of the design that is very
similar to a part of
Do the shard metadata files list all of the segments that make up the
backup, or only the segments that were uploaded in this incremental update?
On Fri, Jan 8, 2021 at 3:19 PM Jason Gerlowski
wrote:
> Sure thing. I put together a writeup on the file layout and formats
> here:
> https://cwiki.a
I don't remember, to be honest. I think javadocs, html linters - they
all contribute. We could remove some of them (via -x) on github if
they're too heavy.
On Fri, Jan 8, 2021 at 9:08 PM Michael Sokolov wrote:
>
> Whoops! Well I guess we'll see. I wonder what other checks we are running now
> th
Sure thing. I put together a writeup on the file layout and formats
here:
https://cwiki.apache.org/confluence/display/SOLR/Incremental+Backup+File+Format
The details get a little verbose, so I made it a subpage that the
SIP-proper calls out to.
Let me know what you think when you get a chance to
The problem with a single repository is that it will/may conflict at times.
Also, I still don't see the problem with having the extra repo as long as
we aren't releasing anything.
The problem with (b) is that you can't create a PR from a random repository
to a repo it isn't a fork of. I also don't
I ran 3000+ iterations with that seed (on JDK11) and it did not
reproduce. Then another 2000+ with JDK-16 runtime, and no fails there
either.
On Fri, Jan 8, 2021 at 10:32 AM David Smiley wrote:
>
> Hm; this is spooky
> ~ David Smiley
> Apache Lucene/Solr Search Developer
> http://www.linkedin.com
Whoops! Well I guess we'll see. I wonder what other checks we are running
now that's were not part of precommit, aside from spotless?
On Fri, Jan 8, 2021, 2:28 PM Dawid Weiss wrote:
> The switch to full check (minus tests) makes the "precommit" really
> heavy... it's around 14 minutes on my not-
The switch to full check (minus tests) makes the "precommit" really
heavy... it's around 14 minutes on my not-so-fancy machine. I'm
curious if it's going to pass on github or timeout somehow.
On Fri, Jan 8, 2021 at 8:26 PM Houston Putman wrote:
>
> Yeah, that should work Dawid. I'll create a PR t
Yeah, that should work Dawid. I'll create a PR that tests it out.
- Houston
On Fri, Jan 8, 2021 at 2:02 PM Dawid Weiss wrote:
> Can those jobs just run a sequence of two commands -
>
> ./gradlew localSettings
> ./gradlew check -x test
>
> This would solve the problem as proper JVM settings (tun
Can those jobs just run a sequence of two commands -
./gradlew localSettings
./gradlew check -x test
This would solve the problem as proper JVM settings (tuned for the
machine/ image it's running on) would be written and used on
subsequent run.
Dawid
On Fri, Jan 8, 2021 at 7:44 PM Uwe Schindler
The problembcomes from the fact that Gradle sets a Xmx for itself in the
settings file. Jenkins installs a settings file before.
As every GitHub run is a clean checkout in new working dir, the settings never
persist.
IMHO, we should change the command line and pass JVM options to set heap size
Hello!
A few quick answers:
In theory, an FST node can have inCount > 1 (when it is in the "suffix"
part of the FST), but it sounds like maybe you found an exciting bug if
indeed FST Builder fails to ever increment it. E.g. try building some FSTs
here: http://examples.mikemccandless.com/fst.py t
Glad to see you contributing Walter!
Unless you know it only applies to 8x, you should branch against master.
~ David Smiley
Apache Lucene/Solr Search Developer
http://www.linkedin.com/in/davidwsmiley
On Thu, Jan 7, 2021 at 4:26 PM Walter Underwood
wrote:
> Starting work on this change. Shoul
Perhaps the OOMs are because .github/workflows/gradle-precommit.yml
yesterday switched from doing "gradlew check -x test" to "gradlew precomit"
? CC Michael Sokolov
~ David Smiley
Apache Lucene/Solr Search Developer
http://www.linkedin.com/in/davidwsmiley
On Fri, Jan 8, 2021 at 1:06 PM Timothy
Same for my PR too ... OOMs about 14 minutes in ...
On Fri, Jan 8, 2021 at 9:45 AM Houston Putman
wrote:
> Weirdly enough, Github PR precommit actions have started to OOM. Not sure
> if it's a github thing or something that changed on our end...
>
> On Fri, Jan 8, 2021 at 11:37 AM Joel Bernstein
Weirdly enough, Github PR precommit actions have started to OOM. Not sure
if it's a github thing or something that changed on our end...
On Fri, Jan 8, 2021 at 11:37 AM Joel Bernstein wrote:
> It turned out to be this while I merged branches:
>
> warning: inexact rename detection was skipped due
It turned out to be this while I merged branches:
warning: inexact rename detection was skipped due to too many files.
warning: you may want to set your merge.renamelimit variable to at least
1639 and retry the command.
Joel Bernstein
http://joelsolr.blogspot.com/
On Fri, Jan 8, 2021 at 11:16
Thanks Eric, I'll do a fresh clone, something must be out of wack with my
local repo.
Joel Bernstein
http://joelsolr.blogspot.com/
On Fri, Jan 8, 2021 at 10:55 AM Eric Pugh
wrote:
> It ran for me just fine. I *think* you may not be up to date, as
> dataimporthandler/ is no longer in master!
It ran for me just fine. I *think* you may not be up to date, as
dataimporthandler/ is no longer in master!
> On Jan 8, 2021, at 10:08 AM, Joel Bernstein wrote:
>
> I'm getting failing gradle precommits in master:
>
> > Task :solr:contrib:validateSourcePatterns FAILED
> tabs instead spaces:
Hm; this is spooky
~ David Smiley
Apache Lucene/Solr Search Developer
http://www.linkedin.com/in/davidwsmiley
On Fri, Jan 8, 2021 at 5:18 AM Policeman Jenkins Server
wrote:
> Build: https://jenkins.thetaphi.de/job/Lucene-Solr-master-Linux/29209/
> Java: 64bit/jdk-16-ea+30 -XX:+UseCompressedOops
I'm getting failing gradle precommits in master:
*> Task :solr:contrib:validateSourcePatterns* FAILED
tabs instead spaces:
/Users/joelbernstein/committer/lucene-solr/solr/contrib/dataimporthandler/build/test-results/test/TEST-org.apache.solr.handler.dataimport.TestDocBuilder.xml
tabs instead spa
Not necessarily. Most people contribute to Apache Lucene/Solr using
external repositories (forks) and raise pull requests against Apache owned
repositories. There's no SGA needed on such occasions.
I see two paths forward from here.
a) Lets setup a single repository for all packages/plugins, say
24 matches
Mail list logo
