Hi,
Another issue we are facing in task-manager af...
# troubleshootinga
anil suddapally
08/08/2024, 12:02 AMHi,
Another issue we are facing in task-manager after it ran fine for 2days.
java.lang.IllegalStateException: The library registration references a different set of library BLOBs than previous registrations for this job:
old:[p-0cb2d6fdcb9622d8074299b1425b69e0bdadec3d-3cfd70061f61678c43220938a4b215ba]
new:[p-0cb2d6fdcb9622d8074299b1425b69e0bdadec3d-14bd91dfef2178babffb73a4eee6d1bd]
at org.apache.flink.runtime.execution.librarycache.BlobLibraryCacheManager$ResolvedClassLoader.verifyClassLoader(BlobLibraryCacheManager.java:419) ~[flink-dist-1.17.2.jar:1.17.2]
at org.apache.flink.runtime.execution.librarycache.BlobLibraryCacheManager$ResolvedClassLoader.access$500(BlobLibraryCacheManager.java:359) ~[flink-dist-1.17.2.jar:1.17.2]
at org.apache.flink.runtime.execution.librarycache.BlobLibraryCacheManager$LibraryCacheEntry.getOrResolveClassLoader(BlobLibraryCacheManager.java:235) ~[flink-dist-1.17.2.jar:1.17.2]
at org.apache.flink.runtime.execution.librarycache.BlobLibraryCacheManager$LibraryCacheEntry.access$1100(BlobLibraryCacheManager.java:202) ~[flink-dist-1.17.2.jar:1.17.2]
d
D. Draco O'Brien
08/08/2024, 7:09 AMYou can try clearing cache of job managers
D. Draco O'Brien
08/08/2024, 7:10 AMalso check your CI/CD pipeline to make sure that you are not accidentally uploading different versions of same UDF or library jar files to the cluster over time.
D. Draco O'Brien
08/08/2024, 7:11 AMif its standalone restarting task managers might suffice to clear caches for K8 deployment you might need to scale up/scale down pods similar with YARN deployments