Shulou(Shulou.com)05/31 Report--

Today, I would like to talk to you about how to understand the process of Oracle CRS/GI. Many people may not know much about it. In order to make you understand better, the editor has summarized the following for you. I hope you can get something according to this article.

The Oracle cluster in 10g and 11.1 is called CRS (Oracle Cluster Ready Service), and the cluster in 11.2 Oracle is called GI (Grid Infrastructure). For CRS/GI, some of their core processes have similar functions, but in 11.2, many new Deamon processes have been added.

10.2 CRS:

$ps-ef | grep crs/bin

Root 4373 3605 0 Feb25? 00:02:49 / u01/app/crs/bin/crsd.bin reboot

Oracle 4380 4379 0 Feb25? 00:00:03 / u01/app/crs/bin/evmd.bin

Oracle 4925 4888 0 Feb25? 00:00:00 / u01/app/crs/bin/oclsomon.bin

Root 4928 4483 0 Feb25? 00:00:00 / u01/app/crs/bin/oprocd.bin run-t 1000-m 10000-

Hsi 5, 10, 50, 75, 90-f.

Oracle 5065 4512 0 Feb25? 00:00:54 / u01/app/crs/bin/ocssd.bin

Oracle 6825 4380 0 Feb25? 00:00:00 / u01/app/crs/bin/evmlogger.bin-o

11.2 GI:

$ps-ef | grep grid/bin

Root 6656 1 0 Feb25? 00:09:20 / u01/app/11.2.0/grid/bin/ohasd.bin reboot

Grid 7960 1 0 Feb25? 00:16:54 / u01/app/11.2.0/grid/bin/oraagent.bin

Grid 7972 1 0 Feb25? 00:00:05 / u01/app/11.2.0/grid/bin/mdnsd.bin

Grid 7982 1 0 Feb25? 00:03:45 / u01/app/11.2.0/grid/bin/gpnpd.bin

Grid 7993 1 0 Feb25? 00:14:19 / u01/app/11.2.0/grid/bin/gipcd.bin

Root 7995 11 Feb25? 00:41:42 / u01/app/11.2.0/grid/bin/orarootagent.bin

Root 8010 10 Feb25? 00:28:03 / u01/app/11.2.0/grid/bin/osysmond.bin

Root 8026 1 0 Feb25? 00:01:31 / u01/app/11.2.0/grid/bin/cssdmonitor

Root 8040 1 0 Feb25? 00:02:22 / u01/app/11.2.0/grid/bin/cssdagent

Grid 8057 11 Feb25? 00:48:04 / u01/app/11.2.0/grid/bin/ocssd.bin

Root 8136 1 0 Feb25? 00:02:49 / u01/app/11.2.0/grid/bin/octssd.bin reboot

Grid 8158 1 0 Feb25? 00:02:24 / u01/app/11.2.0/grid/bin/evmd.bin

Root 8278 1 0 Feb25? 00:04:18 / u01/app/11.2.0/grid/bin/crsd.bin reboot

Grid 8357 8158 0 Feb25? 00:00:00 / u01/app/11.2.0/grid/bin/evmlogger.bin-o

/ u01/app/11.2.0/grid/evm/log/evmlogger.info-l / u01/app/11.2.0/grid/evm/log/evmlogger.log

Grid 8406 1 0 Feb25? 00:11:22 / u01/app/11.2.0/grid/bin/oraagent.bin

Root 8414 11 Feb25? 00:36:22 / u01/app/11.2.0/grid/bin/orarootagent.bin

Let's explain some of these common clustering processes:

What each version has in common:

1. Ocssd.bin: this is a very core process, and if it terminates abnormally, it will cause the cluster or host of this node to restart. This process is mainly used to check whether the voting disk can be accessed normally and whether the communication between private networks of nodes is normal. The LMON process on the database instance is also registered with CSSD, so that CSSD can learn about the health of the database instance through LMON.

If an automatic restart of the host occurs on the node, you need to check the log of ocssd, located at: / log//cssd.

You can refer to the following documentation to learn about node restart:

How to diagnose node restart problems

How does 11gR2 diagnose Node restart problems

2. Crsd.bin: this process is mainly used to manage resources in the cluster. It is used to start and stop checking some resources, such as database instance, ASM, snooping, disk group, VIP, etc. At 11.2, these specific operations are performed by the corresponding agent. In addition, the maintenance of OCR is also done by CRSD.

When you find that some resources are terminated abnormally, you first need to check the log of crsd: / log//crsd.

3. Evmd.bin: event monitoring (event monitor) process, which publishes cluster events, such as instance start, stop, and so on.

4. Ons process: Oracle Notification Service daemon, which is used to receive cluster events sent by evmd, and then send these events to application subscribers or local listeners, so that FAN (Fast Application Notification) can be implemented, and applications can receive these events and process them.

5. Gsd: only required if you need to manage 9i databases on CRS or GI. At 11.2, gsd defaults to offline.

You can refer to the My Oracle Support documentation: GSD Is Used Only if 9i RAC Database is Present (Doc ID 429966.1)

Specific to 10g and 11.1:

1. Oprocd.bin:Oracle Clusterware Process Monitor Daemon is used to monitor the host hang. If the host hang is found, the host restart will be initiated. Run only if you are not using third-party clustering software, which has been used on Linux since 10.2.0.4.

The oprocd log will be at: / etc/oracle/oprocd/*.log.* or / var/opt/oracle/oprocd/*.log.*.

2. Oclsomon.bin: used to monitor whether the ocssd process is hang, and if hang is found, reboot will be initiated.

11.2 unique:

1. Cssdagent (11.2): this process is started by ohasd, which then starts, stops, checks the ocssd process, and runs as root.

Its log is located at: / log//agent/ohasd/oracssdagent_root

2. Cssdmonitor (11.2): monitor cssdagent and check the node hang (similar to oprocd), monitor whether the ocssd process is hang (similar to oclsomon), monitor vendor clusterware (similar to vmon), and run as root.

Its log is located at: / log//agent/ohasd/oracssdmonitor_root

3. Ohasd.bin: when GI starts, ohasd starts first, then it starts agent (orarootagent,oraagent, cssdagnet and cssdmonitor), and each agent starts the corresponding process.

If there is a problem with the GI startup process, you can check the ohasd log: / log//agent/ohasd

4. Mdnsd.bin: this process discovers the nodes and all network card information in the cluster through Multicast. It is important to make sure that the network cards in the cluster support multicast and that the communication between nodes is normal.

Its log is located at: / log//mdnsd

5. Gpnpd.bin: publish the bootstrap information needed to build the cluster and synchronize the gpnp profile among all nodes in the cluster.

Its log is located at: / log//gpnpd

6. Gipcd.bin: this process is responsible for managing all cluster interconnect NICs in the cluster. Private network information is obtained through gpnpd.

Its log is located at: / log//gipcd

7. Gnsd (optional): Grid Naming Service. It is equivalent to a sub-DNS, which is similar to DNS and replaces the use of / etc/hosts for host resolution.

Its log is located at: / log//gnsd

8. Octssd.bin: The Cluster Time Sync Service (CTSS) is used for clock synchronization between nodes. The clock of one node in the cluster is used as a reference node, and the other nodes synchronize the clock with this node. Note: when third-party time synchronization software (e.g. NTP) exists, CTSS runs as an 'observer' and does not modify the node time, but if CTSS does not find third-party time synchronization software, it will start to modify the node time to keep pace with the reference node.

Its log is located at: / log//ctssd.

9. Osysmond.bin: this is the main process of Oracle Cluster Health Monitor (CHM), which runs on all nodes. Sysmond will send the resource usage of each node to cluster logger service, which will receive and save the information of all nodes to CHM's database.

Its log is located at: / log//crfmond/crfmond.log

10. Ologgerd: this is another major process of Oracle Cluster Health Monitor (CHM). In a cluster, ologgerd will have a host point (master) and a standby node (standby). When ologgerd cannot be started on the current node due to a problem, it is enabled on the standby node.

Its log is located at: / log//crflogd/crflogd.log

For CHM, please refer to:

New feature of 11gR2: introduction to Oracle Cluster Health Monitor (CHM)

11. Agent.bin: at 11.2, the start, stop, and check of each resource is performed by agent. Ohasd will start these agent.

There are two types of Agent, one of which is ohasd:

Oraagent_grid: start / stop / check / clear ora.asm, ora.evmd, ora.gipcd, ora.gpnpd, ora.mdnsd and other resources.

Orarootagent_root: start / stop / check / clear ora.crsd, ora.ctssd, ora.diskmon, ora.drivers.acfs, ora.crf (11.2.0.2) and other resources.

Oracssdagent_root: start / stop / check the ocssd process.

Oracssdmonitor_root: monitors the cssdagent process.

Their logs are located at: / log//agent/ohasd

The other is crsd's:

Oraagent_grid: start / stop / check / clear asm, ora.eons, ora.LISTENER.lsnr, SCAN listeners, ora.ons, diskgroup and other resources

Oraagent_oracle: start / stop / check / clear service, database and other resources

Orarootagent_root: start / stop / check / clear GNS, VIP, SCAN VIP and network and other resources.

Scriptagent_grid: application service customized service.

Their logs are located at: / log//agent/crsd

After reading the above, do you have any further understanding of how to understand the process of Oracle CRS/GI? If you want to know more knowledge or related content, please follow the industry information channel, thank you for your support.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.