Maintaining your Cassandra database
You should perform regular maintenance on your Feature Server Cassandra database, to be certain that the database and its backups are all in sync. When any of these events occur, run the corresponding maintenance software routine:
- A week has passed.
- Following a Feature Server restart caused by an upgrade.
- After adding new nodes to the ring, or removing them.
Weekly on all Feature Server nodes
You can perform this procedure:
- as a cron job on Linux. (See https://en.wikipedia.org/wiki/Cron.)
- as a scheduled task on Windows.
STEVE ASKS: ARE MACHINE NAMES AND ADDRESSES PHONY?? THEY MUST NOT BE REAL.
cd /genesys/fs/work/jetty0.0.0.08080fs.
war_
fsany/
webapp/WEBINF/
lib
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 repairAfter a Feature Server restart
A restart can be triggered by adding or removing Feature Server nodes, in a new or existing data center.
cd /genesys/fs/work/jetty0.0.0.08080fs.
war_
fsany/
webapp/WEBINF/
lib
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 repairAdditional Node Maintenance
If you remove a node from the ring, you must also remove the corresponding tokens. Use these software routines:
Run nodetool ring to obtain tokens
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 ring
10.51.29.117 usw1 RAC2 Up Normal 1.29 MB 100.00% 26003787676682001822918611294472056316
10.51.28.221 usw1 RAC1 Up Normal 1.17 MB 66.65% 167086018864645871692761019448293152722
10.52.89.189 aps2 RAC1 Down Normal 1.15 MB 25.90% 41007983964572150951275225962045789866
10.52.88.190 aps2 RAC2 Down Normal 1.16 MB 7.45% 53685600614278234503162023330018045221STEVE ASKS: AM I PLUGGING INFORMATION FROM THE ROUTINE ABOVE INTO THE ROUTINE BELOW?
Run the nodetool remove token routine to remove nodes from ring:
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 removetoken53685600614278234503162023330018045221
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 removetoken41007983964572150951275225962045789866Run the nodetool repair routine
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 repairRun on the ??? nodetool ring to validate updated topology
java cp
libthrift0.7.0.
jar:cassandrathrift1.1.6.
jar:commonscli1.1.
jar:cassandraall1.1.6.
jar org.apache.cassandra.tools.NodeCmd h
localhost p
9192 ring sipfs
10.51.29.117 usw1 RAC2 Up Normal 1.31 MB 100.00% 26003787676682001822918611294472056316
10.51.28.221 usw1 RAC1 Up Normal 1.3 MB 100.00% 167086018864645871692761019448293152722
Comments or questions about this documentation? Contact us for support!
