I am leaning some basic concept of cluster computing and I have some questions to ask.
According to this article:
If a cluster splits into two (or more) groups of nodes that can no longer communicate with each other (aka.partitions), quorum is used to prevent resources from starting on more nodes than desired, which would risk data corruption.
A cluster has quorum when more than half of all known nodes are online in the same partition, or for the mathematically inclined, whenever the following equation is true:
total_nodes < 2 * active_nodes
For example, if a 5-node cluster split into 3- and 2-node paritions, the 3-node partition would have quorum and could continue serving resources. If a 6-node cluster split into two 3-node partitions, neither partition would have quorum; pacemaker’s default behavior in such cases is to stop all resources, in order to prevent data corruption.
Two-node clusters are a special case.
By the above definition, a two-node cluster would only have quorum when both nodes are running. This would make the creation of a two-node cluster pointless
Questions:
From above,I came out with some confuse, why we can not stop all cluster resources like “6-node cluster”?What`s the special lies in the two node cluster?
You are correct that a two node cluster can only have quorum when they are in communication. Thus if the cluster was to split, using the default behavior, the resources would stop.
The solution is to not use the default behavior. Simply set Pacemaker to no-quorum-policy=ignore. This will instruct Pacemaker to continue to run resources even when quorum is lost.
...But wait, now what happens if the cluster communication is broke but both nodes are still operational. Will they not consider their peers dead and both become the active nodes? Now I have two primaries, and potentially diverging data, or conflicts on my network, right? This issue is addressed via STONITH. Properly configured STONITH will ensure that only one node is ever active at a given time and essentially prevent split-brains from even occurring.
An excellent article further explaining STONITH and it's importance was written by LMB back in 2010 here: http://advogato.org/person/lmb/diary/105.html
Related
We have a 13 nodes nifi cluster with around 50k processors. The size of the flow.xml.gz is around 300MB. To bring up the 13 nodes Nifi cluster, it usually takes 8-10 hours. Recently we split the cluster into two parts, 5nodes cluster and 8 nodes cluster with the same 300MB flow.xml.gz in both. Since then we are not able to get the Nifi up in both the clusters. Also we are not seeing any valid logs related to this issue. Is it okay to have the same flow.xml.gz . What are the best practices we could be missing here when splitting the Nifi Cluster.
You ask a number of questions that all boil down to "How to improve performance of our NiFi cluster with a very large flow.xml.gz".
Without a lot more details on your cluster and the flows in it, I can't give a definite or guaranteed-to-work answer, but I can point out some of the steps.
Splitting the cluster is no good without splitting the flow.
Yes, you will reduce cluster communications overhead somewhat, but you probably have a number of input processors that are set to "Primary Node only". If you load the same flow.xml.gz on two clusters, both will have a primary node executing these, leading to contention issues.
More importantly, since every node still loads all of the flow.xml.gz (probably 4 Gb unzipped), you don't have any other performance benefits and verifying the 50k processors in the flow at startup still takes ages.
How to split the cluster
Splitting the cluster in the way you did probably left references to nodes that are now in the other cluster, for example in the local state directory. For NiFi clustering, that may cause problems electing a new cluster coordinator and primary node, because a quorum can't be reached.
It would be cleaner to disconnect, offload and delete those nodes first from the cluster GUI so that these references are deleted. Those nodes can then be configured as a fresh cluster with an empty flow. Even if you use the old flow again later, test it out with an empty flow to make it a lot quicker.
Since you already split the cluster, I would try to start one node of the 8 member cluster and see if you can access the cluster menu to delete the split-off nodes (disconnecting and offloading probably doesn't work anymore). Then for the other 7 members of the cluster, delete the flow.xml.gz and start them. They should copy over the flow from the running node. You should adjust the number of candidates expected in nifi.properties (nifi.cluster.flow.election.max.candidates) so that is not larger than the number of nodes to slightly speed up this process.
If successful, you then have the 300 MB flow running on the 8 member cluster and an empty flow on the new 5 member cluster.
Connect the new cluster to your development pipeline (NiFi registry, templates or otherwise). Then you can stop process groups on the 8 member cluster, import them on the new and after verifying that the flows are running on the new cluster, delete the process group from the old, slowly shrinking it.
If you have no pipeline or it's too much work to recreate all the controllers and parameter contexts, you could take a copy of the flow.xml.gz to one new node, start only that node and delete all the stuff you don't need. Only after that should you start the others (with their empty flow.xml.gz) again.
For more expert advice, you should also try the Apache NiFi Users email list. If you supply enough relevant details in your question, someone there may know what is going wrong with your cluster.
I have two clusters of NetApp (main and dr), in each I have two nodes.
If one of the nodes in either cluster goes down, the other node kicks in and act as one node cluster.
Now my question is, what happens when a whole cluster falls down due to problems of power supply?
I've heard about "Metro Cluster" but I want to ask if there is another option to do so.
It depends on what RPO you need. Metrocluster does synchronous replication of every write and thus provides zero RPO (data loss)
On the other hand you could use Snapmirror which basically takes periodic snapshots and stores them on the other cluster. As you can imagine you should expect some data loss.
I configured a geo cluster using pacemaker and DRBD.
The cluster has 3 different nodes, each node is in a different geographic location.
The locations are pretty close to one another and the communication between them is fast enough for our requirements (around 80MB/s).
I have one master node, one slave node and the third node is an arbitrator.
I use AWS route 53 failover DNS record to do a failover between the nodes in the different sites.
A failover will happen from the master to the slave only if the slave has a quorum, thus ensuring it has communication to the outside world.
I have read that using booth is advised to perform failover between clusters/nodes in different locations - but having a quorum between different geographic locations seems to work very well.
I want to emphasize that I don't have a cluster of clusters - it is a single cluster, with each node in a different geo-location.
My question is - do I need booth in my case? If so - why? Am I missing something?
Booth helps in overlay cluster consisting of clusters running at different sites.
You have one single cluster and hence you should be okay with just Quorum.
I need to provide many elasticSearch instances for different clients but hosted in my infrastructre.
For the moment it is only some small instances.
I am wondering if it is not better to build a big ElastSearch Cluster with 3-5 servers to handle all instances and then each client gets a different index in this cluster and each instance is distributed over servers.
Or maybe another idea?
And another question is about quorum, what is the quorum for ES please?
thanks,
You don’t have to assign each client to different index, Elasticsearch cluster will automatically share loading among all nodes which share shards.
If you are not sure how many nodes are needed, start from a small cluster then keep monitoring the health status of cluster. Add more nodes to the cluster if server loading is high; remove nodes if server loading is low.
When the cluster continuously grow, you may need to assign a dedicated role to each node. In this way, you will have more control over the cluster, easier to diagnose the problem and plan resources. For example, adding more master nodes to stabilize the cluster, adding more data nodes to increase searching and indexing performance, adding more coordinate nodes to handle client requests.
A quorum is defined as majority of eligible master nodes in cluster as follows:
(master_eligible_nodes / 2) + 1
I have a RabbitMQ cluster with two nodes in production and the cluster is breaking with these error messages:
=ERROR REPORT==== 23-Dec-2011::04:21:34 ===
** Node rabbit#rabbitmq02 not responding **
** Removing (timedout) connection **
=INFO REPORT==== 23-Dec-2011::04:21:35 ===
node rabbit#rabbitmq02 lost 'rabbit'
=ERROR REPORT==== 23-Dec-2011::04:21:49 ===
Mnesia(rabbit#rabbitmq01): ** ERROR ** mnesia_event got {inconsistent_database, running_partitioned_network, rabbit#rabbitmq02}
I tried to simulate the problem by killing the connection between the two nodes using "tcpkill". The cluster has disconnected, and surprisingly the two nodes are not trying to reconnect!
When the cluster breaks, HAProxy load balancer still marks both nodes as active and send requests to both of them, although they are not in a cluster.
My questions:
If the nodes are configured to work as a cluster, when I get a network failure, why aren't they trying to reconnect afterwards?
How can I identify broken cluster and shutdown one of the nodes? I have consistency problems when working with the two nodes separately.
RabbitMQ Clusters do not work well on unreliable networks (part of RabbitMQ documentation). So when the network failure happens (in a two node cluster) each node thinks that it is the master and the only node in the cluster. Two master nodes don't automatically reconnect, because their states are not automatically synchronized (even in case of a RabbitMQ slave - the actual message synchronization does not happen - the slave just "catches up" as messages get consumed from the queue and more messages get added).
To detect whether you have a broken cluster, run the command:
rabbitmqctl cluster_status
on each of the nodes that form part of the cluster. If the cluster is broken then you'll only see one node. Something like:
Cluster status of node rabbit#rabbitmq1 ...
[{nodes,[{disc,[rabbit#rabbitmq1]}]},{running_nodes,[rabbit#rabbitmq1]}]
...done.
In such cases, you'll need to run the following set of commands on one of the nodes that formed part of the original cluster (so that it joins the other master node (say rabbitmq1) in the cluster as a slave):
rabbitmqctl stop_app
rabbitmqctl reset
rabbitmqctl join_cluster rabbit#rabbitmq1
rabbitmqctl start_app
Finally check the cluster status again .. this time you should see both the nodes.
Note: If you have the RabbitMQ nodes in an HA configuration using a Virtual IP (and the clients are connecting to RabbitMQ using this virtual IP), then the node that should be made the master should be the one that has the Virtual IP.
From RabbitMQ doc: Clustering and Network Partitions
RabbitMQ also three ways to deal with network partitions automatically: pause-minority mode, pause-if-all-down mode and autoheal mode. The default behaviour is referred to as ignore mode.
In pause-minority mode RabbitMQ will automatically pause cluster nodes which determine themselves to be in a minority (i.e. fewer or equal than half the total number of nodes) after seeing other nodes go down. It therefore chooses partition tolerance over availability from the CAP theorem. This ensures that in the event of a network partition, at most the nodes in a single partition will continue to run. The minority nodes will pause as soon as a partition starts, and will start again when the partition ends. This configuration prevents split-brain and is therefore able to automatically recover from network partitions without inconsistencies.
In pause-if-all-down mode, RabbitMQ will automatically pause cluster nodes which cannot reach any of the listed nodes. In other words, all the listed nodes must be down for RabbitMQ to pause a cluster node. This is close to the pause-minority mode, however, it allows an administrator to decide which nodes to prefer, instead of relying on the context. For instance, if the cluster is made of two nodes in rack A and two nodes in rack B, and the link between racks is lost, pause-minority mode will pause all nodes. In pause-if-all-down mode, if the administrator listed the two nodes in rack A, only nodes in rack B will pause. Note that it is possible the listed nodes get split across both sides of a partition: in this situation, no node will pause. That is why there is an additional ignore/autoheal argument to indicate how to recover from the partition.
In autoheal mode RabbitMQ will automatically decide on a winning partition if a partition is deemed to have occurred, and will restart all nodes that are not in the winning partition. Unlike pause_minority mode it therefore takes effect when a partition ends, rather than when one starts.
The winning partition is the one which has the most clients connected (or if this produces a draw, the one with the most nodes; and if that still produces a draw then one of the partitions is chosen in an unspecified way).
You can enable either mode by setting the configuration parameter cluster_partition_handling for the rabbit application in the configuration file to:
autoheal
pause_minority
pause_if_all_down
If using the pause_if_all_down mode, additional parameters are required:
nodes: nodes which should be unavailable to pause
recover: recover action, can be ignore or autoheal
...
Which Mode to Pick?
It's important to understand that allowing RabbitMQ to deal with network partitions automatically comes with trade offs.
As stated in the introduction, to connect RabbitMQ clusters over generally unreliable links, prefer Federation or the Shovel.
With that said, here are some guidelines to help the operator determine which mode may or may not be appropriate:
ignore: use when network reliability is the highest practically possible and node availability is of topmost importance. For example, all cluster nodes can be in the same a rack or equivalent, connected with a switch, and that switch is also the route to the outside world.
pause_minority: appropriate when clustering across racks or availability zones in a single region, and the probability of losing a majority of nodes (zones) at once is considered to be very low. This mode trades off some availability for the ability to automatically recover if/when the lost node(s) come back.
autoheal: appropriate when are more concerned with continuity of service than with data consistency across nodes.
One other way to recover from this kind of failure is to work with Mnesia which is the database that RabbitMQ uses as the persistence mechanism and for the synchronization of the RabbitMQ instances (and the master / slave status) are controlled by this. For all the details, refer to the following URL: http://www.erlang.org/doc/apps/mnesia/Mnesia_chap7.html
Adding the relevant section here:
There are several occasions when Mnesia may detect that the network
has been partitioned due to a communication failure.
One is when Mnesia already is up and running and the Erlang nodes gain
contact again. Then Mnesia will try to contact Mnesia on the other
node to see if it also thinks that the network has been partitioned
for a while. If Mnesia on both nodes has logged mnesia_down entries
from each other, Mnesia generates a system event, called
{inconsistent_database, running_partitioned_network, Node} which is
sent to Mnesia's event handler and other possible subscribers. The
default event handler reports an error to the error logger.
Another occasion when Mnesia may detect that the network has been
partitioned due to a communication failure, is at start-up. If Mnesia
detects that both the local node and another node received mnesia_down
from each other it generates a {inconsistent_database,
starting_partitioned_network, Node} system event and acts as described
above.
If the application detects that there has been a communication failure
which may have caused an inconsistent database, it may use the
function mnesia:set_master_nodes(Tab, Nodes) to pinpoint from which
nodes each table may be loaded.
At start-up Mnesia's normal table load algorithm will be bypassed and
the table will be loaded from one of the master nodes defined for the
table, regardless of potential mnesia_down entries in the log. The
Nodes may only contain nodes where the table has a replica and if it
is empty, the master node recovery mechanism for the particular table
will be reset and the normal load mechanism will be used when next
restarting.
The function mnesia:set_master_nodes(Nodes) sets master nodes for all
tables. For each table it will determine its replica nodes and invoke
mnesia:set_master_nodes(Tab, TabNodes) with those replica nodes that
are included in the Nodes list (i.e. TabNodes is the intersection of
Nodes and the replica nodes of the table). If the intersection is
empty the master node recovery mechanism for the particular table will
be reset and the normal load mechanism will be used at next restart.
The functions mnesia:system_info(master_node_tables) and
mnesia:table_info(Tab, master_nodes) may be used to obtain information
about the potential master nodes.
Determining which data to keep after communication failure is outside
the scope of Mnesia. One approach would be to determine which "island"
contains a majority of the nodes. Using the {majority,true} option for
critical tables can be a way of ensuring that nodes that are not part
of a "majority island" are not able to update those tables. Note that
this constitutes a reduction in service on the minority nodes. This
would be a tradeoff in favour of higher consistency guarantees.
The function mnesia:force_load_table(Tab) may be used to force load
the table regardless of which table load mechanism is activated.
This is a more lengthy and involved way of recovering from such failures .. but will give better granularity and control over data that should be available in the final master node (this can reduce the amount of data loss that might happen when "merging" RabbitMQ masters).