Would be useful if task id and Data- folder names were linked
somehow. For example display the folder name in task info. Now it's a
bit complicated to automate the process and make sure you delete only
outdated folders.
On 14/02/2022 21.30, Andy Seaborne wrote:
Yes, it's a good idea.
On 18/02/2022 11:05, Mikael Pesonen wrote:
Yes I think we ran out of disk space also last time and it might corrupt
something.
How does one dump the database without fuseki?
tdb2.tdbdump
Thanks!
On 17/02/2022 23.07, Andy Seaborne wrote:
Mikael,
You've had this problem before.
To
Yes I think we ran out of disk space also last time and it might corrupt
something.
How does one dump the database without fuseki?
Thanks!
On 17/02/2022 23.07, Andy Seaborne wrote:
Mikael,
You've had this problem before.
To revert your database:
1/ Stop the Fuseki server
2/ Delete
Mikael,
You've had this problem before.
To revert your database:
1/ Stop the Fuseki server
2/ Delete Data-0002
3/ You can bnow restart
Now the undelying problem:
Do you have a backup?
Try to dump the database without Fuskei running.
If that works (I doubt it), run compact outside Fuseki.
I get this exception with compact after couple of hours:
15:38:41 WARN Compact :: [10] Exception in compact
org.apache.jena.tdb2.TDBException: NodeTableTRDF/Read
at
org.apache.jena.tdb2.store.nodetable.NodeTableTRDF.readNodeFromTable(NodeTableTRDF.java:87)
Yes, it's a good idea.
TDB2 in Fuseki has the "compact" operation to do this without stopping
the server. It creates a new "Data-/" directory and you can delete
lower numbered databases.
TDB1 - needs the server stopping for a rebuild. If you can stop updates,
stop updates, 9server now
Hi,
we have now 13M triples and space usage of Jena data folder is 88G which
seems high. This is not including text index.
Should we cleanup/compress/rebuild etc the database regularly in order
to keep disk usage lower, or is this normal disk usage?
BR
