Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: Apache Wiki <wikidiffs@apache.org>
To: Apache Wiki <wikidiffs@apache.org>
Date: Mon, 14 Jun 2010 13:22:58 -0000
Message-ID: <20100614132258.16921.10570@eos.apache.org>
Subject: =?utf-8?q?=5BCassandra_Wiki=5D_Update_of_=22FAQ=22_by_JonathanEllis?=

Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Cassandra Wiki" for=
 change notification.

The "FAQ" page has been changed by JonathanEllis.
The comment on this change is: mutations against a single key are atomic.
http://wiki.apache.org/cassandra/FAQ?action=3Ddiff&rev1=3D73&rev2=3D74

--------------------------------------------------

  =3D Frequently asked questions =3D
+  *
-  * [[#cant_listen_on_ip_any|Why can't I make Cassandra listen on 0.0.0.0 =
(all my addresses)?]]
+  [[#cant_listen_on_ip_any|Why can't I make Cassandra listen on 0.0.0.0 (a=
ll my addresses)?]]
+ =

+  *
-  * [[#ports|What ports does Cassandra use?]]
+  [[#ports|What ports does Cassandra use?]]
+ =

+  *
-  * [[#slows_down_after_lotso_inserts|Why does Cassandra slow down after d=
oing a lot of inserts?]]
+  [[#slows_down_after_lotso_inserts|Why does Cassandra slow down after doi=
ng a lot of inserts?]]
+ =

+  *
-  * [[#existing_data_when_adding_new_nodes|What happens to existing data i=
n my cluster when I add new nodes?]]
+  [[#existing_data_when_adding_new_nodes|What happens to existing data in =
my cluster when I add new nodes?]]
+ =

+  *
-  * [[#modify_cf_config|Can I add/remove/rename Column Families on a worki=
ng cluster?]]
+  [[#modify_cf_config|Can I add/remove/rename Column Families on a working=
 cluster?]]
+ =

+  *
-  * [[#node_clients_connect_to|Does it matter which node a Thrift client c=
onnects to?]]
+  [[#node_clients_connect_to|Does it matter which node a Thrift client con=
nects to?]]
+ =

+  *
-  * [[#what_kind_of_hardware_should_i_use|What kind of hardware should I r=
un Cassandra on?]]
+  [[#what_kind_of_hardware_should_i_use|What kind of hardware should I run=
 Cassandra on?]]
+ =

+  *
-  * [[#architecture|What are SSTables and Memtables?]]
+  [[#architecture|What are SSTables and Memtables?]]
+ =

+  *
-  * [[#working_with_timeuuid_in_java|Why is it so hard to work with TimeUU=
IDType in Java?]]
+  [[#working_with_timeuuid_in_java|Why is it so hard to work with TimeUUID=
Type in Java?]]
+ =

+  *
-  * [[#i_deleted_what_gives|I delete data from Cassandra, but disk usage s=
tays the same. What gives?]]
+  [[#i_deleted_what_gives|I delete data from Cassandra, but disk usage sta=
ys the same. What gives?]]
+ =

+  *
-  * [[#reads_slower_writes|Why are reads slower than writes?]]
+  [[#reads_slower_writes|Why are reads slower than writes?]]
+ =

+  *
-  * [[#cloned|Why does nodeprobe ring only show one entry, even though my =
nodes logged that they see each other joining the ring?]]
+  [[#cloned|Why does nodeprobe ring only show one entry, even though my no=
des logged that they see each other joining the ring?]]
+ =

+  *
-  * [[#range_ghosts|Why do deleted keys show up during range scans?]]
+  [[#range_ghosts|Why do deleted keys show up during range scans?]]
+ =

+  *
-  * [[#change_replication|Can I change the ReplicationFactor on a live clu=
ster?]]
+  [[#change_replication|Can I change the ReplicationFactor on a live clust=
er?]]
+ =

+  *
-  * [[#large_file_and_blob_storage|Can I store large files or BLOBs in Cas=
sandra?]]
+  [[#large_file_and_blob_storage|Can I store large files or BLOBs in Cassa=
ndra?]]
+ =

+  *
-  * [[#jmx_localhost_refused|Nodetool says "Connection refused to host: 12=
7.0.1.1", for any remote host. What gives?]]
+  [[#jmx_localhost_refused|Nodetool says "Connection refused to host: 127.=
0.1.1", for any remote host. What gives?]]
+ =

+  *
-  * [[#iter_world|How can I iterate over all the rows in a ColumnFamily?]]
+  [[#iter_world|How can I iterate over all the rows in a ColumnFamily?]]
+ =

+  *
-  * [[#no_keyspaces|Why were none of the keyspaces described in storage-co=
nf.xml loaded?]]
+  [[#no_keyspaces|Why were none of the keyspaces described in storage-conf=
.xml loaded?]]
+ =

+  *
-  * [[#gui|Is there a GUI admin tool for Cassandra?]]
+  [[#gui|Is there a GUI admin tool for Cassandra?]]
+ =

+  *
-  * [[#a_long_is_exactly_8_bytes|Insert operation throws InvalidRequestExc=
eption with message "A long is exactly 8 bytes"]]
+  [[#a_long_is_exactly_8_bytes|Insert operation throws InvalidRequestExcep=
tion with message "A long is exactly 8 bytes"]]
+ =

+  *
-  * [[#clustername_mismatch|Cassandra says "ClusterName mismatch: oldClust=
erName !=3D newClusterName" and refuses to start]]
+  [[#clustername_mismatch|Cassandra says "ClusterName mismatch: oldCluster=
Name !=3D newClusterName" and refuses to start]]
+ =

+  *
-  * [[#batch_mutate_atomic|Are batch_mutate operations atomic?]]
+  [[#batch_mutate_atomic|Are batch_mutate operations atomic?]]
+ =

  =

  <<Anchor(cant_listen_on_ip_any)>>
  =

@@ -80, +124 @@

  =

   1. You can maintain a list of contact nodes (all or a subset of the node=
s in the cluster), and configure your clients to choose among them.
   1. Use round-robin DNS and create a record that points to a set of conta=
ct nodes (recommended).
+  1.
-  1. Use the `get_string_property("token map")` RPC to obtain an update-to=
-date list of the nodes in the cluster and cycle through them.
+  Use the `get_string_property("token map")` RPC to obtain an update-to-da=
te list of the nodes in the cluster and cycle through them.
+ =

   1. Deploy a load-balancer, proxy, etc.
  =

  <<Anchor(what_kind_of_hardware_should_i_use)>>
@@ -203, +249 @@

  =3D=3D Can I change the ReplicationFactor on a live cluster? =3D=3D
  Yes, but it will require restarting and running repair manually to change=
 the replica count of existing data.
  =

+  *
-  * Alter the ReplicationFactor for the desired keyspace(s) in the storage=
 configuration on each node in the cluster.
+  Alter the ReplicationFactor for the desired keyspace(s) in the storage c=
onfiguration on each node in the cluster.
+ =

   * Restart cassandra on each node in the cluster
  =

  If you're reducing the ReplicationFactor:
@@ -221, +269 @@

  =

   * The main limitation on a column and super column size is that all the =
data for a single key and column must fit (on disk) on a single machine(nod=
e) in the cluster.  Because keys alone are used to determine the nodes resp=
onsible for replicating their data, the amount of data associated with a si=
ngle key has this upper bound. This is an inherent limitation of the distri=
bution model.
  =

+  *
-  * When large columns are created and retrieved, that columns data is loa=
ded into RAM which  can get resource intensive quickly.  Consider, loading =
 200 rows with columns  that store 10Mb image files each into RAM.  That sm=
all result set would consume about 2Gb of RAM.  Clearly as more and more la=
rge columns are loaded,  RAM would start to get consumed quickly.  This can=
 be worked around, but will take some upfront planning and testing to get a=
 workable solution for most applications.  You can find more information re=
garding this behavior here: [[MemtableThresholds|memtables]], and a possibl=
e solution in 0.7 here: [[https://issues.apache.org/jira/browse/CASSANDRA-1=
6|CASSANDRA-16]].
+  When large columns are created and retrieved, that columns data is loade=
d into RAM which  can get resource intensive quickly.  Consider, loading  2=
00 rows with columns  that store 10Mb image files each into RAM.  That smal=
l result set would consume about 2Gb of RAM.  Clearly as more and more larg=
e columns are loaded,  RAM would start to get consumed quickly.  This can b=
e worked around, but will take some upfront planning and testing to get a w=
orkable solution for most applications.  You can find more information rega=
rding this behavior here: [[MemtableThresholds|memtables]], and a possible =
solution in 0.7 here: [[https://issues.apache.org/jira/browse/CASSANDRA-16|=
CASSANDRA-16]].
  =

+ =

+  *
-  * Please refer to the notes in the Cassandra limitations section for mor=
e information: [[CassandraLimitations|Cassandra Limitations]]
+  Please refer to the notes in the Cassandra limitations section for more =
information: [[CassandraLimitations|Cassandra Limitations]]
+ =

  =

  <<Anchor(jmx_localhost_refused)>>
  =

@@ -289, +341 @@

  <<Anchor(batch_mutate_atomic)>>
  =

  =3D=3D Are batch_mutate operations atomic? =3D=3D
- No.  [[API#batch_mutate|batch_mutate]] is a way to group many operations =
into a single call in order to save on the cost of network round-trips.  If=
 `batch_mutate` fails in the middle of its list of mutations, no rollback o=
ccurs and the mutations that have already been applied stay applied. The cl=
ient should typically retry the mutation.
+ As a special case, mutations against a single key are atomic, but more ge=
nerally no.  [[API#batch_mutate|batch_mutate]] allows grouping operations o=
n many keys into a single call in order to save on the cost of network roun=
d-trips.  If `batch_mutate` fails in the middle of its list of mutations, n=
o rollback occurs and the mutations that have already been applied stay app=
lied. The client should typically retry the `batch_mutate` operation.
 =20