I set up pgpool (master-slave mode) + postgresql streaming replication. Then I check the correctness of failover_command. If I turn off the network master node, then pgpool waits specified time, if the master node during this time is not responding, then the failover_command. And if I do restart postgresql master via the init.d script that executes pgpool failover_command immediately. As a result, the master node switches to the slave node and the slave node to the master node. To avoid this, it is necessary to switch off pgpool, and then do a restart the master node and include pgpool. How can this problem be solved?
P.S. Sorry for my English :)
Related
I am having a strange issue and am hoping you guys might be able to help!
Problem: I have a 2 node SQL Server 2019 Availability Group Cluster utilising a FSW. Both nodes are using the same DBEngine Service account. and it's been working fine for quite some time.
Today I restarted the passive node DBEngine account . When the node came back up, it was no longer synchronising with node 1. The state of the replica was disconnected, and I could see lots of login failures on Node 1 (active node) SQL Logs.
I found that the DBEngine service account had locked. I had it unlocked, but it soon locked again.
Has anyone got any ideas? Any input would be greatly received!
Steps I tried:
created a new service account to rule out the account being used elsewhere, started both nodes under the new account.... account locked out when node 2 started
unlocked the account, stopped node 2. restarted node 1. Account fine... waited.. account still fine. Started node 2 service... account locked out.
recreated mirroring endpoints on both nodes and reapplied connect permissions to the dbengine service account. - this didn't fix it.
restarted both Servers.
removed the node 2 replica from the availability group, removed all databases (from node 2) and dropped the mirroring endpoint on node 2. restarted node 2 service. - at this point both nodes were happily running under the same service account.
tried re-adding node 2 as a replica using the wizard. It added it, backed up the database, restored to node 2, and got to the very last step where it connects it, and the password locked out again!
The account gets locked if someone is using wrong password.
You can check task scheduler if any task using service account.
If application uses same service account. It could be due to caching of old credentials.
I need to build a cluster with one master node and three worker nodes using TDengine. I followed all the step from the official website (https://www.taosdata.com/en/documentation/cluster), but I still suffered by the offlines from the "show dnodes" command in the taos shell. I think it somehow connected but still miss something. I can use the taos shell to see the cluster status in all worker nodes but just cannot start other slave dnodes. What I did was
clean up all the previous data
use the "create dnode xxx" command in taos shell
modify the FirstEP to the master node for all the taos.cfg in worker nodes
add the internal ip and hostname to each nodes' /etc/hosts
start all the taosd services in all nodes.
you should start all taosd services in all nodes first, then use "create dnode xxx" command in taos shell
I am new to fail-over clustering, I have a two node Windows fail-over cluster (windows server 2016) with SQL Server 2016 Installed in cluster mode which configured successfully and everything is working fine. I needed to configure an Always-On Availability to a DR site, but ended up in error while adding the third node.
I have uninstalled Antivirus,checked for duplicate name in the AD of which there are non and it didn't fix my problem.
Cluster service on node xxxxxxxx did not reach the running state. The error code is 0x5b4. For more information check the cluster log and the system event log from node xxxxxx. This operation returned because the timeout period expired.
Operation failed, attempting cleanup.
The server 'xxxxxx.xxx.net' could not be added to the cluster. An error occurred while adding node 'xxxxxx.xxxx.net' to cluster 'xxxxxx'.
This operation returned because the timeout period expired
From the event viewer in failover cluster manager it says xxxxx has been evicted from the cluster.
You should first run following PowerShell command
Clear-ClusterNode -Name nodeName -Force
After running this command try to add node back to the cluster
Have You tried to add third server from Failover Cluster Manager (FCM) i.e. have you spinned the Role/Server from FCM or from Hyper V manager. Please try to spin Vm from FCM
Trying with VoltDB cluster, created a cluster of 2 nodes with k=1
Cluster initialization was successful, both the nodes are up.
Now, how do i connect to this cluster, i could not find any documentation to setup single IP for cluster.
Will the client connect to particular node IP or cluster IP ?
I am using VoltDB community edition.
In general, you can connect to one node, or to multiple nodes. For simple usage, one node is fine. For a client application where you want lower latency and higher throughput, you should connect to all of the nodes in the cluster. See Connecting to the VoltDB Database for the java client, and in particular section 6.1.2 on using the auto-connecting client which enables you to connect to only one node and the client will automatically connect to all of the other nodes.
For command-line access, see the sqlcmd reference:
--servers=server-id[,...]
Specifies the network address of one or more nodes in the database cluster. By default, sqlcmd attempts to connect to a database on localhost.
Disclosure: I work at VoltDB.
If you wish to connect to a single node try
jdbc:voltdb://192.168.1.5:<port>
as the connection URL or if you wish to connect to cluster try
jdbc:voltdb://192.168.1.5:<port>,192.168.1.6:<port>,<any additional nodes you might have in your cluster>
as the connection url.
I want PostgreSQL Synchronous Streaming Database Replication Status = sync.
I deployed PostgreSQL cluster with 3 node and write sync type - Synchronous. But when i check type SELECT * FROM pg_stat_replication;
- i get first node - sync_state=sync,and other async, what is ? Why its two different type ?
With synchronous streaming replication in PostgreSQL, the commit on the primary is delayed until one of the standby servers has received the corresponding WAL information (the exact meaning of this is configurable with synchronous_commit).
The standby server who first confirms the reception of the WAL information is the one with sync_state 'sync', the other will be 'async'.