4 votes

Cassandra nodes disk space usage above 90%

Warning: If you have any legitimate snapshots of keyspaces, this will clear those out as well (so you'll want to back those up). Go on each of the nodes and call in a terminal: nodetool ...
patrickceg's user avatar
4 votes

Cannot connect to my Cassandra

By default Thrift is not up, you have to start it with nodetool enablethrift
Perry Cox's user avatar
4 votes
Accepted

Error When Starting Cassandra for First Time

This is a known problem, tracked as CASSANDRA-14173 - until new release you need to downgrade Java to at least build 152
Alex Ott's user avatar
  • 316
3 votes
Accepted

Cassandra multinode installation,Cannot open file /var/log/cassandra/gc.log due to Permission denied

Make sure, your /etc/hosts file format is valid, such as root@my-cass1-1:~# cat /etc/hosts | egrep -v "^#|::" 127.0.0.1 localhost 192.168.xxx.13 my-cass1-1.int.xxxxxxxxx.tld my-cass1-1 And no ...
Gabor Vincze's user avatar
3 votes

Can't remove tombstones cells in Cassandra

Tombstone were correctly garbage collected but gc was respecting the gc_grace_seconds value set for the column family (default 10 days). After altered the table with the following cqlsh command: ...
BangTheBank's user avatar
3 votes
Accepted

Cassandra tmpdir change location

Ah, looks like both options have to be present in the cassandra-env.sh file: JVM_OPTS="$JVM_OPTS -Djna.tmpdir=/home/cassandra/tmp" JVM_OPTS="$JVM_OPTS -Djava.io.tmpdir=/home/cassandra/tmp"
Vladimir's user avatar
  • 331
3 votes
Accepted

Cassandra nodes disk space usage above 90%

Think he said deleted data from his one and only keyspace, and thus now got tombstones, try to see here beware about doing manual/major compaction and possible needed free head room on disk. If you've ...
Steffen Winther Sørensen's user avatar
3 votes
Accepted

Is it better to generate "UUID" and "TIMESTAMP" within the NodeJS application or using the database built-in functions?

Re MySQL or MariaDB: Mostly the performance difference is negligible. In general, fetching a row is much more costly than any 'standard' function being used in it. If you do need the value in both ...
Rick James's user avatar
  • 2,548
2 votes
Accepted

cassandra data migration from regular drive to SSD on production server

While Cassandra is running, rsync from HD to SSD. Make sure you use the -H and --delete flags to address hardlinks and deleted sstables/commitlogs. Don't forget commitlogs if they're on the same ...
Dave's user avatar
  • 133
2 votes
Accepted

cassandra nodetool drain running for days

Problem is related with a bug in versions prior to 2.1 (https://issues.apache.org/jira/browse/CASSANDRA-5911). Commit logs are not removed after a flush.
ftrujillo's user avatar
  • 149
2 votes

cassandra unable to create thrift socket

I was having the same issue. Before switching to sync, try increasing rpc_max_threads and use hsha as the rpc_server_type
Brandon Kearby's user avatar
2 votes
Accepted

cassandra unable to create thrift socket

It seems that there is a conflict with rpc_server_type: hsha when switched to sync i ve got first a null point exception and then retried few second later it started thrift.
Raffaello's user avatar
  • 163
2 votes
Accepted

PHP 7.0 extcassandra on Centos 7

Indeed you found an issue (which should be better reported on github) libuv have been recently updated (incompatible version) in EPEL. The new build (1.2.2-4) available in the repository fixes this ...
Remi Collet's user avatar
  • 2,121
2 votes
Accepted

cassandra port is closed

The database is listening on 127.0.0.1:9160. That means it is currently configured to listen for loopback IP address only. In other words, it only listens for connections originating from the same ...
telcoM's user avatar
  • 4,588
2 votes
Accepted

Cassandra initial token in cassandra.yaml

It seems that once the cluster has been initialized, the initial_token value gets written somewhere inside the Cassandra structure. Even after restarting the entire new cluster without the ...
Vladimir's user avatar
  • 331
2 votes

What is the equivalent command of "minikube start --memory 5120 --cpus=4" for "Kind"?

Short answer: There is no equivalent in kind to achieve what you want. However, kind does not enforce any memory/cpu constraint by default, so unless you really want to do that, you can skip below ...
mello's user avatar
  • 46
1 vote
Accepted

cassandra replication not 100 percent on each node

That's a normal situation. Most likely you had a replication factor 3, and with 3 nodes this lead that every node had a replicas for all other nodes. When you added 2 more nodes, every node will keep ...
Alex Ott's user avatar
  • 316
1 vote
Accepted

Apache Cassandra repair snapshots are not deleted

What setting are you using for Repair parallelism in your schedule? We had an issue where we had this set to DATACENTER_AWARE and snapshots were not being cleaned up. We are now using "Parallel" ...
vaguefaded's user avatar
1 vote

Cassandra long queries (33s for dropping empty table)

More than several hundreds tables in Cassandra is a clear sign of the bad data model - if you have thousands of tables, then you need to think about how you're trying to solve your task. You need to ...
Alex Ott's user avatar
  • 316
1 vote
Accepted

Can't authenticate to cassandra: cannot achieve consistency level LOCAL_ONE

When you created cluster - did you change the replication factor for system_auth keyspace? If not, then you need to bring that node back, and change replication factor for it to 3. See detailed ...
Alex Ott's user avatar
  • 316
1 vote

I have 2 DCs, RF is 3 in both of them and consistency is set to quorum. What will happen when a read/write request comes and one of the DC is down?

if you use QUORUM, then you need to have 4 servers with replica's online. So in your case, if one DC is down, operations will fail. If possible, consider use of LOCAL_QUORUM instead.
Alex Ott's user avatar
  • 316
1 vote

How to restrict Cassandra to use fixed memory

You can control the heap size with the MAX_HEAP_SIZE setting in the cassandra-env.sh, however Apache Cassandra also has many components that utilise off-heap memory. Notably, compression offsets, ...
Kurt's user avatar
  • 11
1 vote

Should I see identical sstables on multiple cassandra nodes?

SSTable files are created when flush of memtable happen, and when the SSTables are compacted. Every node may have this happen at different times (plus other factors, like, short downtime, etc.). To ...
Alex Ott's user avatar
  • 316
1 vote
Accepted

Ubuntu - Install cassandra version 3.0.9 fails: Version '3.0.9' for 'cassandra' was not found

If using the http://www.apache.org/dist/cassandra/debian 30x main repo there is only a 3.0.12 package in it. I would recommend that for what its worth. 3.0.9 had some pretty bad bugs including data ...
Chris Lohfink's user avatar
1 vote

opscenter "services" tab greyed out

OpsCenter 5.2.4 only supported up to DSE 4.8, which was C* 2.1. Cassandra 2.2 and 3.0 wont work well with it. Also mostly everything in the services tab is for enterprise users and relies on DSE ...
Chris Lohfink's user avatar
1 vote

aws cluster overprovisioning detection framework/tools

Cassandra exposes many metrics via JMX that you can use to assess cluster load. One of the common open-source tools to monitor it is Graphite, and I've seen folks use collectd to feed JMX (and other ...
Mike Lococo's user avatar
1 vote

Cassandra 2.2 - "nodetool repair" VS "nodetool repair -dc dc1,dc2,dc3"

For complete repair of any Cassandra cluster, you need to run 'nodetool repair' on all nodes, one by one, at least once before reaching the time specified in the gc_grace_seconds setting. You can't ...
deniszh's user avatar
  • 142
1 vote

Cassandra 2.2 - Streaming errors at repair/bootstrap/decommission

You should really look to Cassandra's JIRA, it is a good source of information for strange messages or behaviour. The last strace seems to be related to this tracker for instance
Baptiste Mille-Mathias's user avatar
1 vote
Accepted

Cassandra 3.x on Kubernetes 1.2-release (5 cassandra nodes)

Looks like there are two additional port that need to be open for the containers to talk to each other 7000 and 7001 All better.
user2363318's user avatar

Only top scored, non community-wiki answers of a minimum length are eligible