Tag Info

New answers tagged

1

Based on what you've described I would go with your second option and here is why. Creating just one array and slicing it up into several logical volumes won't be any different than just having one large logical volume when it comes to I/O since you are still sharing the same underlying drives that make up the arrays. SQL server runs best if you can ...


0

You might find that the limit is not where you think. I've had trouble with open file limits and that's set by Centos itself. Please see: How to permanently raise ulimit 'open files' and MySQL 'open_files_limit' Some key points from that link: The default ulimit on open files is 1024. MySQL sets its open_files_limit to whatever the system's ulimit is ...


1

work_mem Obviously, the sort operation spills to disk: Sort Method: external merge Disk: 36224kB More work_mem can help the query, like @Kassandry already suggested. Increase the setting until you see Memory instead of Disk in the EXPLAIN output. But it's probably a bad idea to increase the general setting based on one query. Proper setting depends on ...


1

One of the easiest and most effective ways to increase performance on this sort of query is to execute SET work_mem=40MB (because you have ~32MB of temp file for sorting, and a little extra often helps) then run your query, and see if the EXPLAIN ANALYZE plan changes from disk to an in-memory sort. Afterwards, run RESET work_mem to put the value back to the ...


1

Depending on how selective the combined predicates are, I would imagine a good index for this particular query would be: CREATE INDEX index_name ON products (above_revenue_average ASC, start_date DESC) WHERE status > 100 AND category_id <> 5; The SELECT * is potentially problematic because the index above does not contain all columns. ...


0

Tom Li- MSFT described a solution to what sounds like your problem, but his solution was for SQL Server 2008. (But just how much has the installer been improved since then?) https://social.msdn.microsoft.com/forums/sqlserver/en-US/f009e660-503a-4569-a92d-9cc2ef790ca3/install-failure-certificate-chain-trust Key points from his post include: This ...


2

You cannot set IPAll to "no" on a failover cluster instance (as mentioned before), see: How to configure SQL server to listen on different ports on different IP addresses? More than likely, though, you are wondering how to secure your failover cluster instance of SQL Server to a certain IP address, or to verify that it isn't indeed listening on all IP ...


1

This is by design: Case 4: SQL Server is installed in a clustered environment. On cluster, you cannot configure SQL Server to listen on a specific IP addresses. You must chose IPALL. Source: How to configure SQL server to listen on different ports on different IP addresses?


1

Just to update, we are going to end up moving some of the load to a different server, followed by a complete rebuild of the environment. Thank you for your time and responses.


4

DBA_TABLESPACE_USAGE_METRICS shows units in database blocks, not bytes. 4 MB is not enough for a SYSTEM tablespace. Most likely what you saw was 4M blocks, which is 32 GB with a 8K blocksize tablespace, and that is the maximum size of a datafile in a 8K smallfile tablespace. That is why your attempts had no effect. You can add a new datafile to the ...


2

I agree with the above. I would also check your queries hitting the CPU and determine if any query tuning could be implemented or at the very least some plan guides. But all in all, your configuration is just not sustainable.


7

Things I see wrong right away Way to many instances Not enough vCPUs vNUMA probably doesn't match NUMA I'm guessing there's a ton of missing indexes MAXDOP of 1 or 2 Cost threshold of 50 Optomize for adhoc workloads = on (it's probably off) Questions What do the waits look like? What are all the memory configs setup for? Fixing this is going to take ...


2

Bug 11872813 - V$PGA_TARGET_ADVICE VIEW IS EMPTY. After setting PGA_AGGREGATE_TARGET > approximately 12G the V$PGA_TARGET_ADVICE view is empty. This issue will be fixed in version 12.1 and a fix is expected to be included in patch set 11.2.0.4. Apply Patch 11872813


1

I think Ops Manager (can be deployed as local MMS) fits your requirement: … Ops Manager assists with: Deployment. Any topology, at any scale Management. Deploy new clusters. Manage, monitor, and back up existing ones Upgrades. In minutes, with no downtime Scaling. Add capacity, without taking the application offline Point-in-time, ...


0

The --port parameter controls the base port that a mongod or mongos server listens to for TCP connections. The HTTP interface is accessible at a port number that is 1000 greater than the base port, so given a --port of 28017 you can connect via port 29017 if you have enabled the HTTP interface. You can see both ports listed in the server log on startup: ...


3

MongoDB doesn't have an automatic backup feature built-in -- you have to choose an appropriate backup strategy using command-line or third party tools. The backup procedure will vary depending on your deployment type (standalone, replica set, sharded cluster), disk/dbpath configuration, and backup requirements. Unless you are using a third party tool, ...


0

You're on RHEL6 or a derivative from your kernel version 2.6.32-431.3.1.el6.x86_64. The default firewall only allows ssh incoming. You'll need to poke a hole in your firewall. Look here for firewall tips: http://wiki.centos.org/HowTos/Network/IPTables



Top 50 recent answers are included