cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jens Rantil (JIRA)" <>
Subject [jira] [Created] (CASSANDRA-11562) "Could not retrieve endpoint ranges" for sstableloader
Date Wed, 13 Apr 2016 15:27:25 GMT
Jens Rantil created CASSANDRA-11562:

             Summary: "Could not retrieve endpoint ranges" for sstableloader
                 Key: CASSANDRA-11562
             Project: Cassandra
          Issue Type: Bug
          Components: Tools
         Environment: $ uname -a
Linux bigdb-100 3.2.0-99-virtual #139-Ubuntu SMP Mon Feb 1 23:52:21 UTC 2016 x86_64 x86_64
x86_64 GNU/Linux

I am using Datastax Enterprise 4.7.5-1 which is based on 2.1.11.
            Reporter: Jens Rantil

I am setting up a second datacenter and have a very slow and shaky VPN connection to my old
datacenter. To speed up import process I am trying to seed the new datacenter with a backup
(that has been transferred encrypted out of bands from the VPN). When this is done I will
issue a final clusterwide repair.

However...sstableloader crashes with the following:

sstableloader -v --nodes XXX --username MYUSERNAME --password MYPASSWORD --ignore YYY,ZZZ
Could not retrieve endpoint ranges:
java.lang.RuntimeException: Could not retrieve endpoint ranges:
Caused by: java.lang.IllegalArgumentException
        at java.nio.Buffer.limit(
        at org.apache.cassandra.utils.ByteBufferUtil.readBytes(
        at org.apache.cassandra.serializers.CollectionSerializer.readValue(
        at org.apache.cassandra.serializers.MapSerializer.deserializeForNativeProtocol(
        at org.apache.cassandra.serializers.MapSerializer.deserializeForNativeProtocol(
        at org.apache.cassandra.serializers.CollectionSerializer.deserialize(
        at org.apache.cassandra.db.marshal.AbstractType.compose(
        at org.apache.cassandra.cql3.UntypedResultSet$Row.getMap(
        at org.apache.cassandra.config.CFMetaData.fromSchemaNoTriggers(
        at org.apache.cassandra.config.CFMetaData.fromThriftCqlRow(
        ... 2 more
(where YYY,ZZZ are nodes in the old DC)

The files in ./backupdir/MYKEYSPACE/MYTABLE/ are an exact copy of a snapshot from the older
datacenter that has been taken with the exact same version of Datastax Enterprise/Cassandra.
The backup was taken 2-3 days ago.

Question: ./backupdir/MYKEYSPACE/MYTABLE/ contains the non-"*.db" file  "manifest.json". Is
that an issue?

My workaround for my quest will probably be to copy the snapshot directories out to the nodes
of the new datacenter and do a DC-local repair+cleanup.

Let me know if I can assist in debugging this further.

 * This _might_ be a duplicate of
claims this could happen when dropping a column, but don't think I've dropped any column for
this column ever.

This message was sent by Atlassian JIRA

View raw message