Deep understanding of Oracle RAC 12c notes

Deep understanding of Oracle RAC 12cJump to: Navigation, search

Directory 1 Overview 2 Cluster parts management and fault diagnosis 3 Operational Practices 4 new Features 5 Storage and ASM 6 problems in application design 7 managing and tuning a complex RAC environment 8 Backup and Recovery 9 Network Practice Ten Optimizing the RAC database One Lock and deadlock A Parallel Queries - cluster components and database upgrades - RAC One Node

overview

1. P5 RAC uses cache Fusion technology to manage data caching for all nodes and to maintain consistency ... Why does
   1. not read from the high-speed IO network, but from the other node's cache?
   2. p15 GCS GES GRD
   3. There are some background processes: LMS lmon LMD LCK DIAG
2. cluster architecture
   1. full Share
      1. Primary-from: SQL server cluster, Oracle Fail Safe, RAC one Node
      2. primary-primary: RAC
   2. no sharing
3. 12CR1 Flex cluster
   1. p28 loosely coupled leaf-node architecture with a few hub nodes tightly coupled
4. multiple physical network cards are combined as one logical network card: Linux bonding/ms NIC Teaming
5. 11.2.0.2 Haip
6. RAC Component: ASM + clusterware
   1. crsctl
   2. OCR (cluster registry)
   3. p19 ONS (Notification Service) FAN (Quick app notification) FCF (fast connection failover
7. brain fissure: stonith algorithm, healthy node let the problem node restart?
8. Transparent application failover (TAF)
9. virtual IP
10. apply continuity (AC)
11. rolling upgrade?
12. Cost of ownership: The standard version of the Oracle database contains RAC, but the total number of CPU slots <=4
    1. is cheaper compared to large SMP servers because Linux and low-cost hardware are used?
13. RAC can withstand up to N-1 failures, with n-1/n headroom for each hardware plan

Cluster parts management and fault diagnosis

1. Ohas:gpnp GNS GIPC MDNS osysmond ologgerd
2. CRS (upper Stack): CRS CSS CSS proxy CSS monitor CTSS EVM ONS oraagent orarootagent
   1. A bunch of management command-line programs: slightly

Operational Practices

1. Lreg: Performing a listener registration
2. Failover type: NONE SESSION SELECT (remember cursor position) TRANSACTION
3. 11.2 SCAN (front-end redirection?) ）
4. 12c GDS
5. Temporary table space
   1. SS queue, DFS lock handle with CI queue, 11g+ release extents cache in bulk (100 each)
6. Modification of large amounts of data
   1. Reduce global cache transfers and allocate all PX services within the same node as possible
7. Performance Metrics collection: Oswatcher, AWR, Statspack
8. Parameter file Management
9. Password file Management
10. SVRCTL: Managing Databases and instances

new Features

1. Flex Cluster and Flex ASM
   1. Slightly
2. Multi-tenancy: There are multiple PDB in a CDB
   1. Cdb$root
   2. Pdb$seed
   3. Pdb
3. Other: IPv6 GDS ACFS/ADVM online resource Property modification
   1. Policy-based database management
   2. ASM Disk Group
   3. ...

Storage and ASM

1. Storage protocol: SCSI FC IP NAS
2. Multipath: A set of SCSI devices pointing to the same storage device (LUN)
3. Asmlib?
   1. P169 strongly recommends using an external RAI configuration like RAID 1+0 to ensure ASM redundancy
4. Storing OCR and voting disk on ASM
5. ACFS

problems in application design

1. ' Right Index growth competition ' (because of the self-increment key?) ）：
   1. Hash index partition
   2. Hash table partitioning (logical reads increase massively if you do not have the partition trimmed data)
   3. Reverse index (scope query not supported)
   4. Use multiple sequences of different ranges (applied control)
2. A lot of Truncate/drop
   1. Object-level checkpoint events can be raised, and global waits can cause serious problems
   2. Global temp table (GIT)?
3. Sequence Cache
   1. Order,cache 20
4. Free Block list and automatic segment table space Management (ASSM)
5. Too many commits
   1. LGWR delay causes LMS delay, further causes the other node to cache the event more long waits, ' GC buffer busy wait '
6. A transaction that has not been committed for a long time (resulting in a lot of rollbacks?) ）
7. Local Access *
8. Small table Update
9. Index Design
   1. Bitmap indexes are good for read-only tables; ' Compressed index '
10. Inefficient execution plan
11. Too many parallel scans
12. Full table Scan: direct path?
13. Relevance between apps *
14. Implementation of application changes
    1. Library Cache Lock/pin

managing and tuning a complex RAC environment

1. Share vs non-shared home directory
2. Server pool
3. 11gR2 Instance Lock
   1. sql> ALTER SYSTEM SET cpu_count=2 scope=both sid= ' instance_name ';
4. Small-scale vs large-scale
5. Split Brain: Node expulsion, fencing, STONITH
   1. Building redundancy at the network layer
   2. Allocate enough cpu/memory for application workloads and set quotas
   3. Implementation of DBRM and Iorm
   4. Avoid configuring too many lms_processes
   5. AUTO DOP
6. Node Eviction
7. Extended distance Clusters-summary, generalization, and best practices
8. RAC installation configuration in different OS considerations/Tips
   1. Linux:hugepages, ASMM, GI
   2. Windows:
      1. Have public NICs take precedence over private NICs?
      2. Using asynchronous IO

Backup and Recovery

1. p240 using rapid Recovery Zone (FRA)
2. Multi-Channel
   1. Rman> Configure Channel 1 device type disk connect ' Sys/[email protected]_1 ';
3. P250 block Change Tracking (BCT)

Network Practice

1. p265 RDS protocol requires specific hardware (InfiniBand) and kernel drivers
   1. Similar to UDP, there is no ACK mechanism in RDS, and further, the RDS design is 0 replicated; completely skipped IP (? ）
2. The presence of the p270 VIP avoids the need to wait for TCP timeouts
3. Cluster inline/private network
   1. Each NIC needs to be configured with a separate subnet segment (11.2+) =〉 Otherwise, if you unplug the network cable on the first NIC, ARP will not update the ARP cache correctly (why?). ）
4. Jumbo Frame: mtu=9000
5. LB and fail-over
   1. 12c HAVIP
6. Kernel parameters
   1. Net.ipv4.ipfrag_low/high_thresh, Net.ipv4.ipfrag_time
7. GC Lost block problem
8. Gns

Optimizing the RAC database

1. grd*
2. RAC Wait Event *
   1. GC Current Block 2-way/3-way
   2. GC CR Block 2-way/3-way
   3. GC CR Grant 2-WAY/GC current Grant 2-way
   4. GC CR Block BUSY/GC current block Busy
   5. GC CR Block CONGESTED/GC current block congested
3. Placeholder Wait Events
4. Send-side analysis
5. GCS Log Flush Sync
6. GC Buffer Busy Acquire/release
   1. Unique index
   2. Table Block
7. Overview of the DRM process *
8. AWR reports and ADDM

Lock and deadlock

1. Global resource type:
   1. BL (Buffer Lock), GCS Management
   2. Non-BL:TX, TM, sq,ges management
2. Lock mode *
3. Troubleshooting Methods for lock contention *
4. Queue contention (temporary)
   1. TX
   2. Tm
   3. HW
5. DFS Lock Handle
   1. SV Resources
   2. CI Resources
   3. p373 Queue type: Ss/iv/ci
6. Library Cache Locks/pins
7. Queue statistics Information
8. V$wait_chains
9. Hanganalyze
10. Dead lock
11. LMD Trace File

Parallel Queries

1. p416 partition-wise Join
   1. The traditional PX uses the Producer/consumer mode,...

cluster components and database upgradesRAC One Node

Deep understanding of Oracle RAC 12c notes