Hi all.
In production we want to run nodetool repair but each time we do it we get the
too many open files error.
We've increased the number of available FD for Cassandra till 8192 but still
we get the same error after few seconds.
Should I increase it more?
WARN [Thread-7] 2011-07-19 12:34:00,348 CustomTThreadPoolServer.java (line
131) Transport error occurred during acceptance of message.
org.apache.thrift.transport.TTransportException: java.net.SocketException: Too
many open files
at org.apache.thrift.transport.TServerSocket.acceptImpl(TServerSocket.
java:124)
at org.apache.cassandra.thrift.TCustomServerSocket.acceptImpl
(TCustomServerSocket.java:68)
at org.apache.cassandra.thrift.TCustomServerSocket.acceptImpl
(TCustomServerSocket.java:39)
at org.apache.thrift.transport.TServerTransport.accept
(TServerTransport.java:31)
at org.apache.cassandra.thrift.CustomTThreadPoolServer.serve
(CustomTThreadPoolServer.java:121)
at org.apache.cassandra.thrift.CassandraDaemon$ThriftServer.run
(CassandraDaemon.java:155)
Caused by: java.net.SocketException: Too many open files
at java.net.PlainSocketImpl.socketAccept(Native Method)
at java.net.PlainSocketImpl.accept(PlainSocketImpl.java:408)
at java.net.ServerSocket.implAccept(ServerSocket.java:462)
at java.net.ServerSocket.accept(ServerSocket.java:430)
at org.apache.thrift.transport.TServerSocket.acceptImpl(TServerSocket.
java:119)
... 5 more
nodetool repair keyspacename -h host
Cassandra 0.7.5, 1 cluster, 5 nodes. Each node give the same output.
One more question: when repair start throwing this kind of exceptions (very
fast) we stop the process of repair ... is it dangerous for data?
Best Regards
Carlo
