Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Oracle Database 12.1.0.2
Hi Erman, I am new to Oracle GRID installation. I have a few questions regarding the setup. Below is doc used for setup: https://docs.oracle.com/database/121/LADBI/oraclerestart.htm#LADBI7717 6.5.1 Installing Oracle Grid Infrastructure for a Standalone Server with a New Database Installation Should I perform steps 6.4 before proceeding steps 6.5.1? in step 4, which disks should I choose? If I have 16 disks, which should I choose? Should I choose all 16 disks if I want to enable Oracle ASM stripping? If RAID 10 is used, will the disk groups be automatically mirrored? Which AU size do you recommend me to use(1, 2, 4, 8, 16, 32 or 64 MB)? Section 6.3.5 How many disk groups should I create using asm(assume 16 disks). Should I select normal redundancy? Please guide me Thanks, Roshan |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
Step 6.4 iand 6.5 is different things.
Please read the documentation carefully.. you do 6.4 or 6.5.. It depends on your choice.. Please read the titles of 6.4 and 6.5.. Then, you can decide which way to go according to your needs. There is no exact answer for your diskgroup related question.. Again it depends.. If you choose normal redundancy,you will have a 2 way mirror. So yout 16 disks will be like 8 disks in size. If you choose high redundancy,you will have a 3 way mirror... (more secure than normal redundancy, but you will have less disks in size.) As for AU related question -> 4MB recommended for Exadata. Actually, it depends on your database type.. For OLTP , 4MB is recommended. However, for Data warehouses, you may consider having bigger AUs. (8MB, 16MB and so on) |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
See my blog post -> https://ermanarslan.blogspot.com/2014/12/rac-11203-rac-installation-on-2-node.html
It is a good and detailed additional reference for you. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
In reply to this post by ErmanArslansOracleBlog
Ok thanks for clarification
What about external option? Is it when no mirroring is to be done? |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
when you choose normal or high redundancy, ASM does the mirroring.
But if you choose external redundancy, you rely on the Store level raid. So, when you choose external redundancy, you tell ASM to not to do the mirroring.. you can use external redundancy, if you do the mirroring/striping etc.. on the storage layer. . (RAID) |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Ok thanks for advise.
Whenever I installl database, should I choose General Purpose or Data Warehousing if the db will be used for datawarehousing? Suppose I install the database using ASM with disk group "DATA" containing 4 disks. Later, I create a new disk group DATA2 with another 4 disks. Will the datafiles automatically stipped across the new 4 disks? Thanks, Roshan |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
1) Yes.
If you will use it as a DWH, choose DWH. But ofcourse you will have to do more than that.. (after creating the database, you may need to configure it more to provide the datawarehousing you need) 2)As for your Diskgroup related question, the answer is NO! The redundancy is done based on the diskgroup. Not across the diskgroups. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Hi Erman,
thanks a lot for the clarifications. Can you please advise if I create another disk group and the first disk group is filled, will the database write the data on the new disk group? Thanks, Roshan |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
No.
You put your datafiles in diskgroups. A datafile can be in a single diskgroup. So if your diskgroup is filled, you need to increase its size by adding a new disk to that diskgroup. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Hi Erman,
I have configured 1 disk group with 4 disks each size 250 GB. Suppose I need to increase the size to 20T, how will I increase the size? Adding another 76 disks of 250G? |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
It depends on your choice.
You can also resize your current asm disks and then more asm disks with the updated size. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Hi,
if I add additional disks in a disk group, will ASM rebalancing automatically take place or should I do it manually so that disk is striped across the additional disks also |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
Yes.
This is from Oracle Support. Altering a diskgroup (add, drop, undrop, resize) will trigger an automatic rebalance operation according to ASM_POWER_LIMIT initialization parameter if no POWER clause specified in the alter command. If the optional POWER clause is specified, ASM will rebalance the diskgroup using the integer value to override the value that the ASM_POWER_LIMIT initialization parameter. |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
Many thanks for your support
![]() |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
1294 posts
|
In reply to this post by ErmanArslansOracleBlog
Hi Erman,
I have resized my Oracle ASM disks from 250 Gb to 500 GB as well as adding 4 new disks. But when I query SQL> set linesize 1000 SQL> break on report on disk_group_name skip 1 compute sum label "Grand Total: " of total_MB used_MB on report SQL> SQL> SELECT 2 name GROUP_NAME 3 , round ( total_mb) TOTAL_MB 4 , round ( (total_mb - free_mb) ) USED_MB 5 , round ( free_mb) FREE_MB 6 , ROUND((1- (free_mb / total_mb))*100,2 ) USED 7 , (100 - ROUND((1- (free_mb / total_mb))*100,2)) FREE 8 FROM 9 v$asm_diskgroup 10 ORDER BY 11 name 12 / GROUP_NAME TOTAL_MB USED_MB FREE_MB USED FREE ------------------------------ ---------- ---------- ---------- ---------- ---------- DATA 1023968 1018400 5568 99.46 .54 ---------- ---------- Grand Total: 1023968 1018400 I see the size of the disks is still 250 GB. Can you please advise? Regards, Roshan |
Loading... |
Reply to author |
Edit post |
Move post |
Delete this post |
Delete this post and replies |
Change post date |
Print post |
Permalink |
Raw mail |
Administrator
5731 posts
|
This is a new issue , please create a new issue for this Roshan.
Also, please provide the detailed info.. (how did you resized the ASM disks?) Probably, you didn't make ASM to recognize the disk rezise operation that you did in OS and/or Storage level.. |
Free forum by Nabble | Edit this page |