danny.leinad wrote:

I've changed the summary of bug, also on pl.wiki users report this problem.

PS Both wikis are on s2.

Same for itwiki, eswiki, ptwiki, ruwiki, svwiki and zhwiki. There seems to be a problem with search-pool3.svc.pmtpa.wmnet (10.2.1.13)?

danny.leinad wrote:

(In reply to comment #1)

PS Both wikis are on s2.

re-checked according to http://noc.wikimedia.org/conf/highlight.php?file=lucene.php the problem affects:

} elseif ( in_array( $wgDBname, array( 'itwiki', 'eswiki', 'ptwiki', 'plwiki', 'nlwiki', 'ruwiki', 'svwiki', 'zhwiki' ) ) ) {

1. Pool 3 LVS $wgLuceneHost = '10.2.1.13';

And it's possibly fixed itself:
[17:06] <nagios-wm> RECOVERY - LVS Lucene on search-pool3.svc.pmtpa.wmnet is OK: TCP OK - 0.000 second response time on port 8123

[17:11] <apergos> !log restarted searchd on search7
[17:12] <morebots> Logged the message, Master
[17:12] <p858snake|l> or apergos could be here :p
[17:12] <apergos> look at the ganglia graphs
[17:12] <apergos> http://ganglia.wikimedia.org/?c=Search&h=search7.pmtpa.wmnet&m=cpu_report&r=day&s=descending&hc=3&mc=3
[17:12] <jeremyb> moin apergos
[17:12] <apergos> morning. barely.
[17:13] <apergos> that fixed it.
[17:13] <jeremyb> which graph?
[17:13] <apergos> search7
[17:13] <jeremyb> load?
[17:13] <apergos> for daily
[17:13] <apergos> all of them

romaine.wiki wrote:

Again same problems on nl-wiki.

• Bug 30967 has been marked as a duplicate of this bug. ***

Is this issue still current? Per Ryan_Lane there are no longer errors visible in nagios. Apparently some notion that Mark and Tim have worked on the issue, but no idea of the status.

Tim restarted searchd on 7 the day when it rebroke.

reclosing.