Unable to read artifacts from Nexus - Constant OrientDB WARN log messages

Over the past few days we have been struggling with an issue where we are unable to read or push anything towards our Nexus instance and the only way to recover is to restore it from a backup.

At the time tons of the below message appear (like 2-3 every milli second).

2022-06-23 07:49:30,635+0000 WARN [qtp1258398841-613] ci.user com.orientechnologies.orient.core.storage.impl.local.paginated.OPaginatedCluster - $ANSI{green {db=component}} Page in file asset_14.pcl with index 154 was placed in wrong free list, this error will be fixed automatically.

The issue re-appears randomly after a restore on the same asset file and the same indexes are reported.

Around a month ago we have upgraded our instance to 3.37.3, where the issue initially occurred. Yesterday we have upgraded to 3.39.0, hoping that it would sort this out.

Looking at OrientDB’s github repository, there was a similar issue reported and there seems to be a solution in version 3, however Nexus is still using v2 of the database.

So far we have checked the filesystem for consistency checks, ran checks on the orientdb, rebuilt the indexes. All seems to be in order, no issues were noted.

Do you have idea what could be going wrong here? Any feedback would be appreciated.

Thank you.

Adrian

You might try the suggestion of the orient db folks - Orient - Page File index placed in wrong list Warning · Issue #7971 · orientechnologies/orientdb · GitHub