Taxonomy shows 1.3 TB where as available space is 120 GB

[raviskiran]

Taxonomy shows 1.3 TB where as available space is 120 GB. Also the system has gone slow. Should I have to re-index?

[jllort]

How much documents do you have, and which is your configuration ( DBMS )

Put some screenshot to understanding better the problem and some disk usage screenshot too.

Which OpenKM version are you using?

On which way do you detect system is slower ... and give us some idea about how many concurrent users are logged and hardware resources ( ram etc... )

[raviskiran]

I am using mysql DB.

4GB ram

JAVA_OPTS="-Xms256m -Xmx2048m -XX:PermSize=128m -XX:MaxPermSize=256m -Djava.awt.headless=true"

40 concurrent users

Attached my statistics

[raviskiran]

Code: Select all

20:19:20,201 INFO  [BundleCache] num=1049 mem=8190k max=8192k avg=7995 hits=1263893 miss=696107
20:19:22,664 INFO  [LRUNodeIdCache] num=485/10240 hits=3783 miss=1966217
20:19:24,279 INFO  [BundleCache] num=1052 mem=8188k max=8192k avg=7971 hits=1271279 miss=698721
20:19:27,116 INFO  [LRUNodeIdCache] num=485/10240 hits=3783 miss=1976217
20:19:27,992 INFO  [BundleCache] num=1086 mem=8188k max=8192k avg=7720 hits=1278613 miss=701387

This is the last few lines of the log.

[pavila]

Which OpenKM version are you using?

Do you have many documents with the same content in the repository?

[raviskiran]

I am using 5.1.8...what do u mean by same content? can you give me an example?
Also the system is unable to get getkeywordmap cloud and is throwing an error.

[jllort]

Which is the size in hard disk of your repository folder ?
I suggest you upgrade to latest openkm version 5.1.9

[pavila]

I am using 5.1.8...what do u mean by same content? can you give me an example?

I mean the same document in several folders, even with distinct names but the same binary content.

[raviskiran]

I dont think so..but will that affect?

[raviskiran]

Actually I just checked there were quite a few duplicate files? what do we need to do

[pavila]

OpenKM 5.1.x is based on Jackrabbit and this content repository uses a technique called "Document Deduplication" this means when you upload the same document twice in disctint places in the repository, only one document will be stored in the DataStore (by default is Disk Datastore). So, two document of 20KB will use only 20KB in disk.

But the statistics ara generated browsing the repository in depth, so it will count every document stored in OpenKM. So these both document sizes will be added and you will informed about 40KB of repository size.

[pavila]

To avoid this "problem" we have disabled the "deduplication" in OpenKM 6.0