0

我在 AWS 上使用 3 个 EC2 实例节点的 cassandra 集群。

每个节点挂载一个 EBS 卷到路径（/usr/lib/cassandra/），因此 cassandra 的数据存储在 EBS 卷中。

所有节点都有一个 dns 指向自己的私有 ip。并且所有 cassandra 配置文件都使用 DNS 而不是私有 ip。

我的问题是：当一个节点被终止时，我会启动另一个 cassandra，并将 EBS 挂载到新启动的实例上，这样数据就不会丢失。但是在启动 cassandra 服务时，cassandra 服务会失败。日志如下：

我不知道为什么它会使用其他节点的ip？

 INFO 12:47:36,364 Writing Memtable-schema_keyspaces@1358300025(251/2510 serialized/live bytes, 8 ops)
 INFO 12:47:36,367 Enqueuing flush of Memtable-schema_columns@1982787565(91991/919910 serialized/live bytes, 2375 ops)
 INFO 12:47:36,367 Enqueuing flush of Memtable-schema_columnfamilies@59370809(38866/388660 serialized/live bytes, 1052 ops)
 INFO 12:47:36,385 Completed flushing /var/lib/cassandra/data/system/schema_keyspaces/system-schema_keyspaces-jb-10-Data.db (214 bytes) for commitlog position ReplayPosition(segmentId=1382719654978, position=289)
 INFO 12:47:36,391 Writing Memtable-schema_columns@1982787565(91991/919910 serialized/live bytes, 2375 ops)
 INFO 12:47:36,627 Completed flushing /var/lib/cassandra/data/system/schema_columns/system-schema_columns-jb-24-Data.db (19358 bytes) for commitlog position ReplayPosition(segmentId=1382719654978, position=289)
 INFO 12:47:36,628 Writing Memtable-schema_columnfamilies@59370809(38866/388660 serialized/live bytes, 1052 ops)
 INFO 12:47:36,660 Completed flushing /var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-24-Data.db (8750 bytes) for commitlog position ReplayPosition(segmentId=1382719654978, position=289)
 INFO 12:47:36,661 Log replay complete, 64 replayed mutations
 INFO 12:47:36,746 Compacting [SSTableReader(path='/var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-23-Data.db'), SSTableReader(path='/var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-22-Data.db'), SSTableReader(path='/var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-24-Data.db'), SSTableReader(path='/var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-21-Data.db')]
 INFO 12:47:37,584 Compacted 4 sstables to [/var/lib/cassandra/data/system/schema_columnfamilies/system-schema_columnfamilies-jb-25,].  34,990 bytes to 8,759 (~25% of original) in 809ms = 0.010325MB/s.  16 total rows, 4 unique.  Row merge counts were {1:0, 2:0, 3:0, 4:4, }
 INFO 12:47:38,411 Cassandra version: 2.0.1
 INFO 12:47:38,412 Thrift API version: 19.37.0
 INFO 12:47:38,425 CQL supported versions: 2.0.0,3.1.1 (default: 3.1.1)
 INFO 12:47:38,485 Loading persisted ring state
ERROR 12:47:38,537 Exception encountered during startup
java.lang.RuntimeException: Unknown host /172.31.9.175 with no default configured
        at org.apache.cassandra.locator.PropertyFileSnitch.getEndpointInfo(PropertyFileSnitch.java:90)
        at org.apache.cassandra.locator.PropertyFileSnitch.getDatacenter(PropertyFileSnitch.java:113)
        at org.apache.cassandra.locator.DynamicEndpointSnitch.getDatacenter(DynamicEndpointSnitch.java:127)
        at org.apache.cassandra.locator.TokenMetadata$Topology.addEndpoint(TokenMetadata.java:1049)
        at org.apache.cassandra.locator.TokenMetadata.updateNormalTokens(TokenMetadata.java:187)
        at org.apache.cassandra.locator.TokenMetadata.updateNormalTokens(TokenMetadata.java:159)
        at org.apache.cassandra.service.StorageService.initServer(StorageService.java:470)
        at org.apache.cassandra.service.StorageService.initServer(StorageService.java:428)
        at org.apache.cassandra.service.CassandraDaemon.setup(CassandraDaemon.java:343)
        at org.apache.cassandra.service.CassandraDaemon.activate(CassandraDaemon.java:442)
        at org.apache.cassandra.service.CassandraDaemon.main(CassandraDaemon.java:485)
java.lang.RuntimeException: Unknown host /172.31.9.175 with no default configured
        at org.apache.cassandra.locator.PropertyFileSnitch.getEndpointInfo(PropertyFileSnitch.java:90)
        at org.apache.cassandra.locator.PropertyFileSnitch.getDatacenter(PropertyFileSnitch.java:113)
        at org.apache.cassandra.locator.DynamicEndpointSnitch.getDatacenter(DynamicEndpointSnitch.java:127)
        at org.apache.cassandra.locator.TokenMetadata$Topology.addEndpoint(TokenMetadata.java:1049)
        at org.apache.cassandra.locator.TokenMetadata.updateNormalTokens(TokenMetadata.java:187)
        at org.apache.cassandra.locator.TokenMetadata.updateNormalTokens(TokenMetadata.java:159)
        at org.apache.cassandra.service.StorageService.initServer(StorageService.java:470)
        at org.apache.cassandra.service.StorageService.initServer(StorageService.java:428)
        at org.apache.cassandra.service.CassandraDaemon.setup(CassandraDaemon.java:343)
        at org.apache.cassandra.service.CassandraDaemon.activate(CassandraDaemon.java:442)
        at org.apache.cassandra.service.CassandraDaemon.main(CassandraDaemon.java:485)
Exception encountered during startup: Unknown host /172.31.9.175 with no default configured
4

2 回答 2

0

I may be wrong, but in your cassandra.yaml, you may want to look at the seeds variable, it probably is incorrectly referencing an old IP. Another way around this is to use some of the floating IPs in AWS. That way if you spin up a new machine you point the floating IPs to it, and the configuration wouldn't have to change the IPs.

于 2013-10-26T23:05:59.267 回答
0

检查您是否在您的 snitch 属性文件中为未知节点配置了默认值。

default =<Any of your DC name>:RAC1

重新启动更改后，希望这会有所帮助。

于 2015-12-13T11:58:37.233 回答