Questions & Answers

[rbitdd]

Hello community,

a customer reported this ERROR during the generation

java.io.IOException: Database error - com.sleepycat.je.DatabaseException: (JE 3.3.87) Secondary persist#AccessControlDb#de.espirit.firstspirit.acl.File#_lastUpdate is corrupt: the primary record contains a key that is not present in the secondary

Do you have any ideas to fix this problem?

Thx in advance!

[hoebbel · Accepted solution]

Hello rbitdd,

the acl database is corrupt [maybe by parallel writing in the database, a full hard drive...].

You can find the database here:

<FirstSpiritROOT>/data/schedule/<sID>/acl

<sID> = ID of the scheduler Entry, which can be found within the scheduler entry itself, if you edit it.

Just delete the whole directory. Afterwards the deployment will work again.

Best regards,

  Holger Höbbel

[rbitdd]

Hello Mr. Höbbel,

I deleted the acl-directory. Is this enough? Or should I delete the sID-directory?

If it is enough, I can tell you tomorrow, whether it was the solution.

Best regards.

[hoebbel]

Hello rbitdd,

it is enough to delete the acl directory.

Hint: If you do not need the acl database, you can deactivate it within the generate action of the schedule entry.

The acl database is mainly used to

• set the file date of the generated nodes to the date, at which the appropriate node was first generate with the actual content [checked by the use of the hash value of the file] within the actual path.
• do transfer permission information to the live system [if the crc transfer is used and the acl database is used within the live system, e.g. by the multi access control filter of the security module]

Best regards,

  Holger Höbbel

[rbitdd]

The deployment reported no errors! Thank you!