Thanks for the logs. Is the job restore actually failing? If yes, there should be an exception for the exact cause of the failure.
Otherwise, the AvroSerializer warnings in the taskmanager logs is actually expected behaviour when restoring from savepoint versions before 1.7.x, and shouldn't cause job failures (unless something unexpected is happening). Shortly put, to describe the cause of that warning: Previously in 1.6.x, the AvroSerializer was Java-serialized into savepoints. In 1.7.x, when restoring from previous version savepoints, that serializer will still be attempted to be read using Java serialization (which explains the InvalidClassException in the WARN log). However, starting from 1.7 we no longer rely on serializers being written directly into savepoints, so whether or not reading that serializer was successful should not matter and the restore should proceed normally. Please do let me know if the job is actually failing, then we should investigate further. If it is failing, there should be an exception in the JM logs identifying the cause of job failure. CC'ing Igal, as he worked on the AvroSerializer for 1.7.x and might have more info. Cheers, Gordon On Wed, Jan 23, 2019 at 7:42 PM pwestermann <no.westerm...@genesys.com> wrote: > There is not much in the log as this immediately happens when I start the > job. I attached one of the taskmanager logs. The first error message I see > is /Could not read a requested serializer. Replaced with a > UnloadableDummyTypeSerializer./ and the exception is > > > taskmanager.log > < > http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/file/t1547/taskmanager.log> > > > > > > -- > Sent from: > http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/ >
