IndexUpdater info

discuss · September 23, 2018, 10:03am

Fess last commit (yesterday), Java 10 and elasticsearch 6.4.1.

It happens the performance of a smb crawling are progressively degrading, after 130000 docs becoming very slow. In this case I see similar logs on fess-crawler.log and the crawling stop with low CPU usage and low IO throughput, so this is not a hardware resource problem. I have tuned garbage collector but with litte success.

2018-09-23 11:56:31,190 [IndexUpdater] INFO  Sent 10 docs (Doc:{process 80ms, send 1309ms, size 4MB}, Mem:{used 1GB, heap 8GB, max 8GB})
2018-09-23 11:56:31,194 [IndexUpdater] INFO  Processing no docs (Doc:{access 4ms, cleanup 35ms}, Mem:{used 1GB, heap 8GB, max 8GB})

What this messages mean?

Thank you

discuss · September 24, 2018, 7:43am

(from marevol (Shinsuke Sugaya) · GitHub)
You need to set proper settings(# of shards, heap size,…) for your environment.

2018-09-23 11:56:31,190 [IndexUpdater] INFO Sent 10 docs (Doc:{process 80ms, send 1309ms, size 4MB}, Mem:{used 1GB, heap 8GB, max 8GB})

Fess sent 10 docs to fess.YYYYMMDD index in elasticsearch.
For details, see IndexUpdater.
If you need more support, please contact Commercial Support.