Tuesday, 31 December 2019
About Cell Software Processes
Oracle Exadata System Software includes the following software processes:
Cell Server (CELLSRV) services iDB requests for disk I/O and advanced Oracle Exadata Storage Server services, such as predicate processing offload. CELLSRV is implemented as a multithreaded process and should be expected to use the largest portion of processor cycles on a storage cell.
Management Server (MS) provides standalone storage cell management and configuration. MS runs on an Oracle WebLogic Server.
Restart Server (RS) monitors the CELLSRV and MS processes and restarts them, if necessary.
Tuesday, 8 October 2019
Configuring Exadata Cell Storage Servers from scratch
In this article we configure Exadata Cell Storage Servers from scratch, Storage interconnect, CELL Disks, Grid Disks, Flashcache, Flashlog, ASM Storage
Our Cell Server host interconnects as of now are these :
[root@cell ~]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:0C:29:94:9C:E2
inet addr:10.10.1.11 Bcast:10.10.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:79481 errors:0 dropped:0 overruns:0 frame:0
TX packets:47935 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:101776703 (97.0 MiB) TX bytes:9452946 (9.0 MiB)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:7642 errors:0 dropped:0 overruns:0 frame:0
TX packets:7642 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:509969 (498.0 KiB) TX bytes:509969 (498.0 KiB)
We connect to ASM first to reconfigure Cell Storage
Connect to ASM Instance
SQL> drop diskgroup data;
drop diskgroup data
*
ERROR at line 1:
ORA-15260: permission denied on ASM disk group
Connect as SYSASM
SQL> conn / as sysasm
Connected.
SQL> drop diskgroup data;
Diskgroup dropped.
[root@cell ~]# cellcli
CellCLI: Release 12.1.1.1.1 - Production on Tue Oct 08 20:55:40 IST 2019
Copyright (c) 2007, 2013, Oracle. All rights reserved.
Cell Efficiency Ratio: 941
Ensure that cell services are running, RS, CELLSRV and MS
CellCLI> alter cell startup services all
Starting the RS, CELLSRV, and MS services...
CELL-01510: Cannot start a new Restart Server (RS) at port number: 9903. An instance of RS might be already running.
Getting the state of RS services... running
Starting CELLSRV services...
The STARTUP of CELLSRV services was successful.
Starting MS services...
The STARTUP of MS services was successful.
Ensure from the linux cell OS that cellsrv process is running healthy and its parent CELLRSOMT
[root@cell ~]# ps -ef |grep cellrsomt
root 4536 3679 0 20:58 ? 00:00:00 /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellrsomt -rs_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellinit.ora -ms_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellrsms.state -cellsrv_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellrsos.state -debug 0
root 5023 4256 0 21:06 pts/0 00:00:00 grep cellrsomt
[root@cell ~]# ps -ef |grep "/cellsrv "
root 4538 4536 3 20:58 ? 00:00:12 /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellsrv 100 5000 9 5042
root 4989 4256 0 21:05 pts/0 00:00:00 grep /cellsrv
CellCLI> list griddisk
data_CD_DATA01_cell active
data_CD_DATA02_cell active
CellCLI> drop GRIDDISK ALL HARDDISK PREFIX=data
GridDisk data_CD_DATA01_cell successfully dropped
GridDisk data_CD_DATA02_cell successfully dropped
CellCLI> drop flashcache
Flash cache cell_FLASHCACHE successfully dropped
CellCLI> drop flashlog
Flash log cell_FLASHLOG successfully dropped
CellCLI> drop cell
Cell cell successfully dropped
Stopping CELLSRV services...
The SHUTDOWN of CELLSRV services was successful.
CellCLI> CREATE CELL interconnect1=eth0
Cell cell successfully created
Starting CELLSRV services...
The STARTUP of CELLSRV services was successful.
Flash cell disks, FlashCache, and FlashLog will be created...
CellDisk FD_00_cell successfully created
Flash log cell_FLASHLOG successfully created
Flash cache cell_FLASHCACHE successfully created
## In this case we will use eth0 since this is a test environment, In your machine it will be Infiband interconnect IBBOND0, and IBBOND1 or depends on your machine configuration and environment
## Ideally you should have two interconnects, but since this is a test environment we are using eth0 for test only
## Ideally we use this :- CREATE CELL interconnect1=interface_name, interconnect2=interface_name
## But in our test environment we use below
CellCLI> CREATE CELL interconnect1=eth0
Cell cell successfully created
Starting CELLSRV services...
The STARTUP of CELLSRV services was successful.
Flash cell disks, FlashCache, and FlashLog will be created...
CellDisk FD_00_cell successfully created
Flash log cell_FLASHLOG successfully created
Flash cache cell_FLASHCACHE successfully created
CellCLI> CREATE CELLDISK ALL
CellDisk CD_DATA01_cell successfully created
CellDisk CD_DATA02_cell successfully created
CellCLI> list celldisk
CD_DATA01_cell normal
CD_DATA02_cell normal
FD_00_cell normal
CellCLI> CREATE GRIDDISK ALL HARDDISK PREFIX=data, size=20G
CELL-02646: CREATE GRIDDISK failed due to insufficient free space on cell disks: CD_DATA01_cell, CD_DATA02_cell.
## In this case we will use 10G since we are limited with space in the test environment
CellCLI> CREATE GRIDDISK ALL HARDDISK PREFIX=data, size=10G
GridDisk data_CD_DATA01_cell successfully created
GridDisk data_CD_DATA02_cell successfully created
CellCLI> list griddisk
data_CD_DATA01_cell active
data_CD_DATA02_cell active
Connect to ASM Instance
SQL> CREATE DISKGROUP data external REDUNDANCY disk 'o/10.10.1.11/data_CD_DATA02_cell'
ATTRIBUTE 'AU_SIZE' = '4M',
'compatible.rdbms'='11.2.0.4',
'compatible.asm'='12.1.0.2',
'cell.smart_scan_capable'='TRUE'
/
Diskgroup created.
## Your ASM storage with Disk group is now ready for the database
Sunday, 6 October 2019
dbnodeupdate.sh fails for YUM update on Exadata Compute Node
If you are patching Exadata, you might encounter the issue dbnodeupdate.sh fails for YUM update on the compute node, this is because of obsolete RPM packages, check out the details here from Raj Kumar https://lnkd.in/f8UVss8 hashtag#exadata hashtag#patching
Shutdown and Restarting Cell services in Exadata
Shutting down and restarting cell services in Exadata
[oracle@dbnode ~]$ ssh root@10.10.1.11
root@10.10.1.11's password:
Last login: Sun Oct 6 22:59:47 2019 from 10.10.1.21
[root@cell ~]# cellcli
CellCLI: Release 12.1.1.1.1 - Production on Sun Oct 06 23:06:15 IST 2019
Copyright (c) 2007, 2013, Oracle. All rights reserved.
Cell Efficiency Ratio: 2
CellCLI> alter cell shutdown services all
Stopping the RS, CELLSRV, and MS services...
The SHUTDOWN of services was successful.
CellCLI> alter cell startup services all
Starting the RS, CELLSRV, and MS services...
Getting the state of RS services... running
Starting CELLSRV services...
The STARTUP of CELLSRV services was successful.
Starting MS services...
The STARTUP of MS services was successful.
Thursday, 13 December 2018
CELL-01548: Error in Exadata Cell Storage Server
The reason for error CELL-01548 is that diskgroup DATA1 is still mounted, hence you get the message CELL-01548: Unable to shut down CELLSRV because disk group DATA1 may be forced to dismount due to reduced redundancy
[root@cell ~]# cellcli
CellCLI: Release 12.1.1.1.1 - Production on Thu Dec 13 16:36:20 IST 2018
Copyright (c) 2007, 2013, Oracle. All rights reserved.
Cell Efficiency Ratio: 1
CellCLI> alter cell shutdown services all
Stopping the RS, CELLSRV, and MS services...
The SHUTDOWN of ALL services was not successful.
CELL-01548: Unable to shut down CELLSRV because disk group DATA1 may be forced to dismount due to reduced redundancy.
Getting the state of CELLSRV services... running
Getting the state of MS services... running
Getting the state of RS services... running
After i dismounted the diskgroup with command SQL> Alter diskgroup data1 dismount;, i was able to shutdown the cell services
[root@cell ~]# cellcli
CellCLI: Release 12.1.1.1.1 - Production on Thu Dec 13 16:41:22 IST 2018
Copyright (c) 2007, 2013, Oracle. All rights reserved.
Cell Efficiency Ratio: 1
CellCLI> alter cell shutdown services all
Stopping the RS, CELLSRV, and MS services...
The SHUTDOWN of services was successful.
CellCLI>
CELLSRV in Exadata
Being an Exadata guy, many times i had been asked what CELLSRV is and what it does and how it is invoked ? Cell Server (CELLSRV) services iDB requests for disk I/O and advanced Oracle Exadata Storage Server services, such as predicate processing offload. CELLSRV is implemented as a multithreaded process and should be expected to use the largest portion of processor cycles on a storage cell. Cellsrv is launched by cellrsomt by executable /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellrsomt. Lets see how it works, in below output process id 3673 is parent of cellsrv (3682) and hence proved 3673 is cellrsomt which actually launched cellsrv
[root@cell ~]# ps -ef |grep "/cellsrv "
root 3682 3673 2 13:25 ? 00:02:39 /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellsrv 100 5000 9 5042
root 6983 6437 0 14:54 pts/0 00:00:00 grep /cellsrv
[root@cell ~]#
[root@cell ~]# ps -ef |grep 3673
root 3673 3665 0 13:25 ? 00:00:06 /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellrsomt -rs_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellinit.ora -ms_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellrsms.state -cellsrv_conf /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/deploy/config/cellrsos.state -debug 0
root 3682 3673 2 13:25 ? 00:02:40 /opt/oracle/cell12.1.1.1.1_LINUX.X64_140712/cellsrv/bin/cellsrv 100 5000 9 5042
root 6986 6437 0 14:54 pts/0 00:00:00 grep 3673
[root@cell ~]#
I/O Resource Management Profiles in Exadata
I/O Resource Management (IORM) interdatabase plans support profiles to ease management, and configuration of interdatabase plans for hundreds of databases.
Profiles introduce a way to allocate I/O resources for a database. This is done using the database initialization parameter
db_performance_profile. Database administrators can classify different databases as, GOLD, SILVER, BRONZE, by setting the db_performance_profile parameter. As with Oracle Database Resource Manager plans, the db_performance_profile information is automatically pushed to all the storage servers (cells). The following SQL command displays how the profile parameter can be set for a database:SQL> ALTER SYSTEM SET db_performance_profile=gold SCOPE=spfile;
Profiles are specified as directives for the interdatabase plan, and are configured using the CellCLIutility. A profile directive consists of an identifier (name), and a set of attributes. To differentiate between a database directive and a profile directive, a qualifier attribute called
type is used. The type attribute can be set to either database or profile. The following is an example of the type attribute syntax:CellCLI> ALTER IORMPLAN DBPLAN=((name=gold, share=10, limit=100, type=profile), \
(name=silver, share=5, limit=60, type=profile), (name=bronze, share=1, limit=20, \
type=profile))
Subscribe to:
Posts (Atom)