Recommended Resources
Abstract
Everything you should know about DB2 pureScale.
Content
- Introduction
- Installation & Maintenance
Prerequisite check
Preparing to install the DB2 pureScale Feature for DB2 Enterprise Server Edition
db2prereqcheck reports wrong ksh version
Fail to create DB2 Purescale instance on SUSE or Redhat Linux if multipath device is enabled
Installation Methods
Upgrading DB2 pureScale environments to DB2 Version 10.1 roadmap
Upgrading DB2 pureScale environments to DB2 Version 10.5 roadmap
Upgrading DB2 pureScale environments to DB2 Version 11.1 roadmap
- Upgrading pS
Offline Upgrade
Installing offline fix pack updates to a DB2 pureScale instance (manual method)
Rolling Online Upgrade
Online Fixpack Update: InstallFixPack can fail at restarting GPFS upon GPFS update on AIX
Changes to procedure for Performing rolling updates in an automated high availability disaster recovery (HADR) environment - Backup and Restore
Topology-changing backup and restore usage scenario in db2 purescale v10.5
Restoring mksysb backup image in DB2 pureScale environment
ProtectionMode of some resources on IBM.Application gets back to 1 (Critical) after restoring the database on pureScale system
- Upgrading pS
- Core pS components
DB2 Members
What is DB2 Member?- Connections from DB2 clients go only to 'first' logical member on each server in purescale setup
Why do surviving members in a pureScale instance hang until GPFS recovery completion when a member is doing restart light due to its host outage even they don't request any of the retained locks ?
Why do surviving members hang when one of the member is recovering restart light mode due to it's host failure in a pureScale instance?
Why does DB2 member in a pureScale instance automatically restart when attaching debugger to the member process or when issuing gencore command against it ?
Start member failed after enter/exit maintenance in purescale
What is the performance impact of online adding member on the running purescale cluster?
adding member follows dropping member will fail with sqlcode -1015 error in purescale
DB2 pureScale AIX Coresident DB2 Member and Cluster Caching Facily: CPU binding
One senario for Purescale's member state is WAITING_FOR_FAILBACK can't start on home host. Cluster caching facility (CF)
What is DB2 CF?Purescale CF cannot be started on Linux AMD64 platform if CF_MEM_SZ is defined to larger than 256 GB
Set primary CF using command "db2cluster -cm -set -option pprimary -value ***" not working correctly
DB2 CF memory segments show "dest" when listing them under "ipcs -m"
Purescale CF cannot be stated due to the Mellanox adapter registration memory limit
CF trace in DB2 pureScale environment
DB2 pureScale AIX Coresident DB2 Member and Cluster Caching Facily: CPU binding
The actual system memory usage by the CF is higher than what is reported from MON_GET_CF - DB2 cluster services
TSA
What is TSA?
TSA Administrator's and User's guide- How to verify that the DB2 pureScale cluster domain has stopped
Upgrading DB2 LUW Purescale or non-purescale to Fixpack 8 may stop TSA working
db2instance -list and lssam could not work on one of the pureScale node RSCT
Reliable Scalable Cluster Technology Documentation
RSCT Concepts
Frequently asked questions about RSCT peer domain problemsAIX 7.1 TL3 SP4, AIX 6.1 TL9 SP4 + pureScale V10.5 : RSCT 3.2.0.x issues
How to change the DB2 cluster service tiebreaker ?
Node who was IBM.RecoveryRM master node cannot rejoin purescale cluster after rebooting host
RSCT WILL NOT NOTIFY DB2 THAT THE PORT IS DOWN WHEN WE MOVE THAT PORT TO ANOTHER DIFFERENT VLAN GPFS (Shared disk storage)
General Parallel File System Documentation
GPFS CommandsDB2 LUW pureScale ONLINE Maintenenace: Applying a GPFS efix and/or PTF
Creating pureScale instance fails due to GPFS cluster settings for the disk is not cleaned up
Using Hitachi Storage Systems in DB2 pureScale Environments
Online upgrade of GPFS level with a security-related efix
Installing a GPFS efix in a DB2 pureScale cluster while keeping the instance online
DB2 with pureScale feature installation fails because of (GPFS) Portability Layer (GPL) module
Purescale Automatic storage management
Should GPFS cluster configuration servers be up before GPFS daemon starts in pureScale environment?
Security Bulletin: IBM® DB2® LUW on AIX and Linux Affected by Multiple Vulnerabilities in GPFS.
Security Bulletin: IBM® DB2® LUW on AIX and Linux Affected by Multiple Vulnerabilities in GPFS (CVE-2016-2984, CVE-2016-2985).
- How to verify that the DB2 pureScale cluster domain has stopped
- Cluster Interconnect
RDMA Protocol (uDapl)
validateUdaplPing reports connection failure even though ping command is successful
Infiniband
InfiniBand how-to topics
ROCE
Setting up a RoCE network in a GDPC environment (AIX) - Other Key components
Global Bufferpool (GBP)
Buffer pools in a DB2 pureScale environment
Global Lock Manager (GLM)
Lock monitoring in a DB2 pureScale environment overview
- Connections from DB2 clients go only to 'first' logical member on each server in purescale setup
- DB2 HADR in pureScale
HADR on pureScale
Set up and manage disaster recovery for pureScale using HADR- High availability disaster recovery (HADR) monitoring in a DB2 pureScale environment
HADR takeover operations in a DB2 pureScale environment
Changes to procedure for Performing rolling updates in an automated high availability disaster recovery (HADR) environment
Scenario: Deploying HADR in a DB2 pureScale environment
- High availability disaster recovery (HADR) monitoring in a DB2 pureScale environment
- Client features
Automatic client reroute
Connections from DB2 clients go only to 'first' logical member on each server in purescale setup
Client affinities
DB2 pureScale client affinity configuration on Tomcat application server
JDBC connections are not rerouted to other DB2 pureScale member with ERRORCODE=-4499, SQLSTATE=08001 error in client affinity configuration on Weblogic
Workload balancing
Customize workload balancing with member subsets
High availability and load balancing for host database connectivity
- Performance
Small sequence cache size in PureScale can cause degraded performance.
How to safely collect data for hang issue in purescale?
Client connection to pureScale server is slow
Why is collecting db2pd -stack all so intrusive and affecting performance in pureScale ?
The transactions are getting much slower on pureScale members with DMA Error messages
Why do applications writing LOB columns hang when a member is doing restart light due to its host outage even they don't request any of the retained locks ?
What is the performance impact of online adding member on the running purescale cluster?
Why purescale MCR takes longer than non-purescale crash recovery if there are a large number of dirty pages to be recovered?
Why do surviving members hang when one of the member is recovering restart light mode due to it's host failure in a pureScale instance?
Why do surviving members in a pureScale instance hang until GPFS recovery completion when a member is doing restart light due to its host outage even they don't request any of the retained locks ?
- Troubleshooting
How can I identify whether or not a DB2 instance/environment is pureScale?
db2start failed with sql1726 in purescale in VM
"db2pd -db-serverlist" does not return server list information
Create database on purescale with CF_TRANSPORT_METHOD=TCP may encounter a reproduceable crash issue
Create database command in a Purescale environment might fail with SQL10003C error.
Uncommitted Read behaviour in purescale
Network Time Protocol (NTP) configuration error during DB2 10.1 instance creation with pureScale feature on SLES11 SP2
How to recovery RU record when it is missing in the output of "db2hareg -dump" ?
DB2CLUSTER -CLEAR -ALERT FAILS TO CLEAR EXISTING ALERT FILES ON THE INSTANCE SHARED FILE SYSTEM
The db2diag.log is created in the current working directory when 'db2cluster -cfs -stop -all' issued
The purpose of the entries in purescale HA registry file
How to solve the db2start fails with SQL1517N in Purescale environment ?
How to fix issue of "UNKNOWN" instance state reported from "db2instance -list" ?
Why there is no core file generated in purescale for trap/panic issue and how to enable it?
Phantom reads symptom can be seen on the concurrent delete, insert, commit transactions on DB2 pureScale system
db2licm -l shows v9.7 as version for DB2 Purescale
How to change a fenced userid in Purescale ?
Was this topic helpful?
Document Information
Modified date:
21 May 2023
UID
swg22001319