-
mapop has quit
-
mapop joined the channel
-
fllr joined the channel
-
kimchy has quit
-
kimchy joined the channel
-
incry6t joined the channel
-
fllr has quit
-
dblessing joined the channel
-
aberdine has quit
-
mapop has quit
-
Midbits has quit
-
Midbits joined the channel
-
Jippi joined the channel
-
lynnroth has quit
-
lynnroth joined the channel
-
nikitosiusis has quit
-
jlawson joined the channel
-
ar15nut joined the channel
-
danshultz joined the channel
-
ggoZ joined the channel
-
cwebber joined the channel
-
danshultz has quit
-
radiocats joined the channel
-
nmische_ joined the channel
-
piavlo|2 has quit
-
piavlo|2 joined the channel
-
juicer2 joined the channel
-
jerius joined the channel
-
techminer has left the channel
-
aberdine joined the channel
-
logstashbot
-
mapop joined the channel
-
Tinuviel joined the channel
-
tessellare joined the channel
-
Tinuviel
Hi, I have 7servers with ES indexes, I wish to reduce it to six, Is there any easy way to do that?
-
Jippi has quit
-
vali
Tinuviel, is your data distributed over all the nodes ?
-
Tinuviel
Yes, but no shards for indexes
-
jimmyy has quit
-
so what I need to do is to move out three logstash indexes.
-
Jippi joined the channel
-
supersheep joined the channel
-
What I'm trying to do is to move out from server 7 all data write and let there only logstash.
-
stackedsax joined the channel
-
mveitas joined the channel
-
vali
-
logstashbot
-
Tinuviel
I tried to convert es on server number 7 to nodata node, but it's not smart enough to start moving indexes automatially
-
Thanks vali, may be helpfull
-
davuxx joined the channel
-
vali
your welcome
-
Tinuviel
and second thing, logstash will always create it's own elasticsearch node?
-
davuxx has quit
-
davuxx joined the channel
-
I created es node well configured and I wish logstash to use it, but it always create it's own one.
-
vali
Tinuviel, from what I know there is a node for each logstash instance pushing to an ES cluster
-
the LS node is pushing to a ES data node
-
Tinuviel
So logstash will use this node fro all communication with ES? read and writes right?
-
vali
yes
-
davuxx has quit
-
fllr joined the channel
-
Jippi has quit
-
Tinuviel
That's explein a lot
-
I have on logstash server also ES DATA node
-
davuxx joined the channel
-
but now I see I need to configure also logsatsh NODATA node
-
give it more ram, more workers maybe
-
vali
why ?
-
Tinuviel
When I'm sending query for big set of data
-
I'm getting Java heap space error
-
vali
you can use -w I thinkg to specify the numbers of workers for the started LS agent
-
or use more indexers/shippers
-
Tinuviel
and this is happening on logstash server
-
vali
Tinuviel, you need more RAM for that error
-
Tinuviel
what is in fact using this ram? logstash ES node or just data nodes?
-
mveitas has quit
-
vali
If you run out of memory you should also have some dump files in that folder
-
with the pid that died
-
logstash node is not a data node
-
Tinuviel
yes I know
-
vali
ES is separated from LS
-
so if your LS agent is dying you need more ram for that
-
not for ES
-
ksclarke joined the channel
-
Tinuviel
but when logsatsh is starting it's own ES node, and use it to query ES data node isn't it (ES node created by logstash) require more ram?
-
fllr has quit
-
kick-buttowski
hey
-
jessemdavis joined the channel
-
what's the ES version required by logstash 1.3 these days ?
-
sysmonk
vali: he's speaking about embeded ES
-
vali
:)
-
I saw it now
-
thank you
-
Tinuviel
I see that I need more ram for ES, I'm just nor sure whould it be ram for local ES Data node or for no data node.
-
kick: 0.90.9
-
-
logstashbot
-
vali
Tinuviel, are you talking about embedded ES ? :)
-
lostcuaz joined the channel
-
vanberge has quit
-
Tinuviel
vali - no :)
-
I was using it long time ago
-
and I thought I know the differenc
-
but now I'm confused
-
vanberge joined the channel
-
i thougth that I need to setup separetly in ES nodata node, then just point logstash output to it and thats all.
-
but logstash is still creating it's own node
-
vali
Tinuviel, you set the RAM limits for ES in elasticsearch.in.sh
-
start with an ok value for your server and go up from there if needed
-
Tinuviel
ok, and how it will be split against local ES data node and logstash nodata node?
-
when I start elasticsearch I will only start data node
-
vali
that I don't know but you can ask on #elasticsearch
-
Tinuviel
when I start logstash, then I will also start nodata node from logsatsh, almost without any configuration.
-
ok I will.
-
Do you familiar with Kibana? Especially queries?
-
*Are you
-
tobias- joined the channel
-
vali
Tinuviel, I'm not sure why are you asking so much about starting no data nodes
-
danshultz joined the channel
-
tobias-
Quick question; how do I get fqdn in host instead of shortname?
-
vali
those are just nodes (the actual agent hostname) that join an ES cluster
-
tobias-, you have to have it in your logs
-
tobias-
vali: it's not from syslog, it's from a file input. Am I missing something?
-
_maes_ has quit
-
vali
tobias-, if it's in your files that just save it/use it in a variable
-
fllr joined the channel
-
mikber has quit
-
tobias-
it's not from the log-file, it's in the event as "host", the logfile-rows gets in @message
-
vali
great, then just grok the message and save it in some variable
-
fllr has quit
-
fllr joined the channel
-
rhoml has quit
-
rhoml joined the channel
-
fllr has quit
-
Jippi joined the channel
-
devOpsEv joined the channel
-
devOpsEv
hello all
-
can anyone confirm is LOGSTASH-1406 is still an issue in 1.3.2?
-
logstashbot
-
devOpsEv
I know it's unresolved but Logstash has had some major revisions since
-
I don't want to update though unless it's fixed
-
Jippi has quit
-
Jippi_ joined the channel
-
radiocats has quit
-
astol joined the channel
-
matanya has quit
-
anonymuse joined the channel
-
radiocats joined the channel
-
astol
hi all, any help with json filter? getting _jsonparsefailure, trying to parse this: "json" => "{\"gid\":\"69163cccd64aafb91eba8437a2ad844e\"}"
-
"json" => "{\"gid\": \"69163cccd64aafb91eba8437a2ad844e\"}" to be precise