Unable to find transaction 1 in any of my logical logs

[srinathgudimetl]

I am setting up Neo4j cluster with 2 core nodes and few Read Replicas. For now, I have not added any Read Replicas. Using Neo4j Enterprise edition 3.5.0 and OS on all nodes is Ubuntu 18.x. No data has been loaded yet into the database.

Core 1 : 10.161.240.82
Core 2 : 10.161.218.172

I have configured the neo4j.conf file as described in this link: https://neo4j.com/docs/operations-manual/current/clustering/setup-new-cluster/#causal-clustering-new...

Config file:

dbms.connectors.default_listen_address=0.0.0.0
dbms.connectors.default_advertised_address=10.161.240.82
dbms.mode=CORE

causal_clustering.minimum_core_cluster_size_at_formation=2
causal_clustering.minimum_core_cluster_size_at_runtime=2
causal_clustering.discovery_type=LIST
causal_clustering.initial_discovery_members=10.161.240.82:5000,10.161.218.172:5000
causal_clustering.discovery_listen_address=10.161.240.82:5000
causal_clustering.transaction_listen_address=10.161.240.82:6000
causal_clustering.raft_listen_address=10.161.240.82:7000

After starting the Neo4j service on both nodes, I see the below error for the 1st core:

2018-12-04 20:37:02.994+0000 INFO  Discovering other core members in initial members set: [10.161.218.172:5000, 10.161.240.82:5000]
2018-12-04 20:38:52.196+0000 INFO  Bound to cluster with id 860f1833-fc16-4af9-b078-958450249d97
2018-12-04 20:38:52.231+0000 INFO  Discovered core member at 10.161.218.172:5000
2018-12-04 20:38:56.377+0000 ERROR Failed to start Neo4j: Starting Neo4j failed: Component 'org.neo4j.server.database.LifecycleManagingDatabase@6abef7bd' was successfully initialized, but failed to start. Please see the attached cause exception "Unable to find transaction 1 in any of my logical logs: Couldn't find any log containing 1". Starting Neo4j failed: Component 'org.neo4j.server.database.LifecycleManagingDatabase@4beaf6bd' was successfully initialized, but failed to start. Please see the attached cause exception "Unable to find transaction 1 in any of my logical logs: Couldn't find any log containing 1".

What does this error mean and how to resolve it?

[aguyhasnoname]

Hi srinathgudimetla,

I'm running into a similar issue and wondering, even though no one replied to this question, did you manage to solve it on your own? Any insights into what's going on would be greatly appreciated!

Tommy

[srinathgudimetl]

I (kind of) fixed this issue by starting the cores in single mode before running them in cluster mode. If the issue doesn't fix, you can try unbinding the graph db or delete the graph.db (if you don't have any data)

[aguyhasnoname]

Many thanks for the reply. I'm running them all via a kubernetes manifest so not sure how easy it'd be to apply one of those solutions.

Thanks though!

[sourinkarmakar]

I too got this error while uploading a database in neo4j causal cluster with 3 core servers and 6 read replicas.
But I got a fix for this...

1. Delete the graph.db folder inside <neo_4j_path>/data/database and then try neo4j-admin import in the core server (I chose Leader).
2. Then started the server 'neo4j start'.
3. Stop all the other servers, delete the graph db in other servers and start the servers back.

For me it worked. Just try out.

[dana_canzano]

see https://neo4j.com/docs/operations-manual/current/clustering/setup-new-cluster/#unbind-instance-from-...

if you modify/move/remove/copy in a new graph.db then you must run bin/neo4j-admin unbind on the stopped instance before restart.

[sourinkarmakar]

Thank you for putting the point @dana.canzano
Sorry I forgot to mention.
I have done ./neo4j unbind in all the servers after stopping the servers and then I did the above steps.