Hi all,

I'm french, excuse my poor english. ;-)


in few words this is my case and my configuration :

   - nxlog for eventslog and sending them to rsyslog in json.
   - syslog on linux and aix servers
   - rsyslog to receive and elasticsearch for indexation/save.
   
Elasticsearch :

   - 1 node
   - index.number_of_shards: 1
   - index.number_of_replicas: 0
   

With nxlog, we've sent a whole windows eventslog to elasticsearch near 
6:30pm. Memory and CPU use grow up during the logs transfert (near 
10minutes) then CPU and memory down to normal. Few minutes later (~30mn), 
CPU/RAM occupation grew up again and stay to 100% on one CPU from 7pm to 
9am this morning until i restart elasticsearch service.
This morning kibana won't respond and elasticsearch directory show indexes 
from years 2011, 2012 and 2013. I think i'ts because of old messages in the 
windows eventlog.

After restarting elasticsearch, everythings came to normal but i've lost 
all logs message between yesterday 7pm and today 9am.

<https://lh5.googleusercontent.com/-LkJcQF5fk5k/VFIaHXfBAUI/AAAAAAAALMQ/nx7vLKKEx7M/s1600/Capture.PNG>

This is what i think, can someone say me if i'm wrong or right ?


The server took in 10minutes near 500MB of logs with only 1 node and 1 
shard. The indexation start and took all the night. I stopped it too soon. 
All the logs in cue were lost when i've restart the service.

Have you some tweaks or tips to optimize my confirguration ?

Regards

Bastien

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/96297107-01ea-445a-a205-03fa03a24cdd%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to