trying to find articles regarding how Casandra balances the load when add a server/node? that is, after adding a node, how cassandra moves certain partitions from existing nodes to new node, and how quick it could be done?
When you add a node to your existing cluster cassandra automatically will assign token ranges to your new node and stream relevant data to it. While this happens nodetool status will show the node as JOINING.
After streaming completes the node is part of your and will handle requests as any other node also does reducing the load on them. But your data size on the old nodes won't shrink - you need to issue nodetool cleanup to get rid of the now obsolete data.
For how quick this can be done, that depends on your clusters load and data size that must be streamed - but streaming is limited on network bandwidth of course.
Related
I added a new node to Cassandra cluster by making it seed node and than started rebuilding it (nodetool rebuild ) command. Although node joined the cluster quickly, the rebuild process that started streaming from all nodes in selected caused the whole dc nodes to slow down. The impact on application is severe. I'll have to stop rebuild process in order to keep normal operation ON!.
Here, I'm seeking advice if you guys can share ways/tricks so to minimize the impact of (node rebuild ) operation on rest of dc nodes and application.
I'll much appreciate your suggestions - thanks for reading my message and your help in advance.
While adding a new node you shouldn't make it a seed node. The seed node is used to bootstrap other nodes and join them in cluster. Making the new node as a seed node will not allow to join the new node in the cluster. Follow the steps provided in the Cassandra docs provided in the link below.
https://docs.datastax.com/en/archived/cassandra/2.0/cassandra/operations/ops_add_node_to_cluster_t.html
This is the best way to add a new node in the cluster.
Note: Make sure the new node is not listed in the -seeds list. Do not make all nodes seed nodes. Please read Internode communications (gossip).
As I understand, you added a node as a seed node just so it will not bootstrap and join the cluster instantly. While is approach is right for it joins the cluster quickly, the downside is, it will not bootstrap and hence will not copy all the data from other nodes that it is responsible for. When you run a rebuild on that node, data is blindly copied (without doing any validation) from other nodes, which can choke up existing nodes throughput and your network pipeline. This approach is very safe and is recommended when using adding a new DC but not when adding nodes to an existing DC.
When you are adding a node, the simplest way is to add a node using the procedure described here. https://docs.datastax.com/en/archived/cassandra/2.0/cassandra/operations/ops_add_node_to_cluster_t.html
When the node bootstrap, it will copy data needed from other nodes but will not start taking client connections until it fully bootstraps and validates the data. So, add one node at a time and let it bootstrap so all the necessary data is copied and validated. After you are done adding the number of nodes you desire, run a cleanup on all the previously joined nodes to cleanup all the keys for which the old nodes are not responsible.
Our cluster is unbalanced and most of the data is in one node. Now the node which as most of the data is dead because of out of space.
How to bring the node UP with ready only mode and rebalance the cluster?
We are using vnodes and DSE 4.0.3
There is no explicit read only mode for Cassandra. As such you're likely to need to temporarily add some disk space to the node to get it online and then rebalance the cluster.
If that's not an option then removing snapshots can sometimes give you enough space to get going. Running nodetool cleanup can also help if it's not previously been run.
If you're using vnodes - then a common problem is if you've converted an old style token node to vnodes. The node will just grab an even range of tokens that maps to it's original range. If the other nodes in the cluster have randomly generated tokens it'll lead to a huge imbalance between them. Decommissioning the node and then re-adding it should resolve the problem.
The output of nodetool ring will show us if that's happened. Actually the chances are a decom and re-add will be the solution in any case.
Use nodetool drain to stop receiving writes on a certain node.
I have a 4 node cluster and I have upgraded all the nodes from an older version to Cassandra 1.2.8. Total data present in the cluster is of size 8 GB. Now I need to enable vNodes on all the 4 nodes of cluster without any downtime. How can I do that?
As Nikhil said, you need to increase num_tokens and restart each node. This can be done one at once with no down time.
However, increasing num_tokens doesn't cause any data to redistribute so you're not really using vnodes. You have to redistribute it manually via shuffle (explained in the link Lyuben posted, which often leads to problems), by decommissioning each node and bootstrapping back (which will temporarily leave your cluster extremely unbalanced with one node owning all the data), or by duplicating your hardware temporarily just like creating a new data center. The latter is the only reliable method I know of but it does require extra hardware.
In the conf/cassandra.yaml you will need to comment out the initial_token parameter, and enable the num_tokens parameter (by default 256 I believe). Do this for each node. Then you will have to restart the cassandra service on each node. And wait for the data to get redistributed throughout the cluster. 8 GB should not take too much time (provided your nodes are all in the same cluster), and read requests will still be functional, though you might see degraded performance until the redistribution of data is complete.
EDIT: Here is a potential strategy to migrate your data:
Decommission two nodes of the cluster. The token-space should get distributed 50-50 between the other two nodes.
On the two decommissioned nodes, remove the existing data, and restart the Cassandra daemon with a different cluster name and with the num_token parameters enabled.
Migrate the 8 GB of data from the old cluster to the new cluster. You could write a quick script in python to achieve this. Since the volume of data is small enough, this should not take too much time.
Once the data is migrated in the new cluster, decommission the two old nodes from the old cluster. Remove the data and restart Cassandra on them, with the new cluster name and the num_tokens parameter. They will bootstrap and data will be streamed from the two existing nodes in the new cluster. Preferably, only bootstrap one node at a time.
With these steps, you should never face a situation where your service is completely down. You will be running with reduced capacity for some time, but again since 8GB is not a large volume of data you might be able to achieve this quickly enough.
TL;DR;
No you need to restart servers once the config has been edited
The problem is that enabling vnodes means a lot of the data is redistributed randomly (the docs say in a vein similar to the classic ‘nodetool move’
Is it possible to add a new node to an existing cluster in cassandra 1.2 without running nodetool cleanup on each individual node once data has been added?
It probably isn't but I need to ask because I'm trying to create an application where each user's machine is a server allowing for endless scaling.
Any advice would be appreciated.
Yes, it is possible. But you should be aware of the side-effects of not doing so.
nodetool cleanup purges keys that are no longer allocated to that node. According to the Apache docs, these keys count against the allocated data for that node, which can cause the auto bootstrap process for the next node to not properly balance the ring. So depending on how you are bringing new user machines into the ring, this may or may not be a problem.
Also keep in mind that nodetool cleanup only needs to be run on nodes that lost keyspace to the new node - i.e. adjacent nodes, not all nodes, in the cluster.
So there is a fair amount of documentation on how to scale up a Cassandra, but is there a good resource on how to "unscale" Cassandra and remove nodes from the cluster? Is it as simple as turning off a node, letting the cluster sync up again, and repeating?
The reason is for a site that expects high spikes of traffic, climbing from the daily few thousand hits to hundreds of thousands over a few days. The site will be "ramped up" before hand, starting up multiple instances of the web server, Cassandra, etc. After the torrent of requests subsides, the goal is to turn off the instances that are not longer used, rather than pay for servers that are just sitting around.
If you just shut the nodes down and rebalance cluster, you risk losing some data, that exist only on removed nodes and hasn't replicated yet.
Safe cluster shrink can be easily done with nodetool. At first, run:
nodetool drain
... on the node removed, to stop accepting writes and flush memtables, then:
nodetool decommission
To move node's data to other nodes, and then shut the node down, and run on some other node:
nodetool removetoken
... to remove the node from the cluster completely. The detailed documentation might be found here: http://wiki.apache.org/cassandra/NodeTool
From my experience, I'd recommend to remove nodes one-by-one, not in batches. It takes more time, but much more safe in case of network outages or hardware failures.
When you remove nodes you may have to re-balance the cluster, moving some nodes to a new token. In a planed downscale, you need to:
1 - minimize the number of moves.
2 - if you have to move a node, minimize the amount of transfered data.
There's an article about cluster balancing that may be helpful:
Balancing Your Cassandra Cluster
Also, the begining of this video is about add node and remove node operations and best strategies to minimize the cluster impact in each of these operations.
Hopefully, these 2 references will give you enough information to plan your downscale.
First, on the node, which will be removed, flush memory (memtable) to SSTables on disk:
-nodetool flush
Second, run command to leave a cluster:
-nodetool decommission
This command will assign ranges that the node was responsible for to other nodes and replicates the data appropriately.
To monitor a process you can use command:
- nodetool netstats
Found an article on how to remove nodes from Cassandra. It was helpful for me scaling down cassandra.All actions are described step-by-step there.