Distribution of ASM & Failgroups

Similar Questions

• Distribution of ASM and disk required

  If I use the external RAID 1, the minimum disks required are two. Now
  If I combine the DSO (interleaving) and external RAID 1 (mirroring only), how many records to a minimum is necessary to achieve the purpose of distribution of ASM + RAID 1 mirror writing? OR, another way, implement WITHOUT Raid on disks, instead, let the ASM mirror and strip, in this case, what are the minimum number of disks must be used?
  
  
  Thank you
  
  Scott

  If you want to mirror and scratch using asm, then you must use at least 4 disks.

  Check this: http://url4u.co/15FTGm

• ASM failgroup

  Hello
  
  11 GR 1 material, is it possible to change the failgroup that ASM disk is assigned to the after it is created, or do I have to remove the Group and add it specifying the correct failgroup?
  
  Thank you
  Justin

  Justin,

  The best place to find it would be Oracle docs.
  http://download.Oracle.com/docs/CD/B28359_01/server.111/b28286/statements_1006.htm#BABGCDFE

  And looking at it, its clear that you must re-create the group in order to change the redundancy.

  HTH
  Aman...

• Is have RAID (1 + 0) so that Failgroups for all ASM ASM starts is a well-known best practice?

  Dear Experts,

  Is have RAID (1 + 0) so that Failgroups for all ASM ASM starts is a well-known best practice?

  Having both is the best practice

  • RAID (1 + 0)
  • ASM Failgroups for all ASM starts

  Thank you

  IVW

  You can create groups of ASM disks with normal or high redundancy or specify external redundancy. It depends on your reliability requirements and storage performance. Remember that ASM is not RAID and redundancy is the base file. ASM uses alternating devices. Oracle is generally recommended to have redundancy of storage. If you have RAID redundancy at the level of hardware controller or storage, Oracle recommends to configure ASM disks with external redundancy groups. Redundancy of the DSO is only between groups of disk failure and by default, each device's own failure group. Of course, you always want to make sure that you do not rely on redundancy of data between logical devices or partitions residing on a single physical unit. Compared to external RAID redundancy redundancy ASM using will give the DBA more control and transparency on the setting of the underlying.

• Question about ASM a Failgroups

  Hello everyone,
  
  I have an ASM Diskgroup that is short on space. The diskgroup has NORMAL redundancy and three groups of failure with two disks of each. All the disks are the same size (50G).
  Now the diskgroup is almost full and I have to add space. And I have two raw devices of 36G each available.
  
  I read that recommmends Oracle for all groups of the failure of the same size. So use it means I can't do these two free disks for ASM?
  Because if I create another Failgroup with them, the Group would have a different size than the others. And if I add them to a group (if it is possible at all), this group would also have a different size.
  
  Thanks for your thoughts.
  
  Mario

  The Diskgroup has three default groups: FG1, FG2, FG3.

  in this case you have 3 discs 2 you need to add the disc in each behalf can be 36 GB but all should have the same size disc.

  Seems clear, but on the other hand that means once I start with 50G disks, I always have to use 50 G discs in the future. What happens if they are out of stock etc. ?

  same size of DISKGROUP not discs. so 3 disc may be 36 GB each.

• Disks / start transposition ASM

  Hi all

  Db1 size: 700 GB +.

  Size of DB2: 2.8 TB +.

  + ASM_DG_DATA is > for CRD files

  + ASM_DG_FRA is > for rman backups / archive files in logs/EXPDP...

  > > All recommended options?

  How many drives / starts, we can keep transposition ASM (for good e/s)?

  I suggest you consult the documentation:

  https://docs.Oracle.com/CD/E11882_01/server.112/e18951/asmprepare.htm#OSTMG11000

  You can find more useful 'Recommendations for the preparation of storage'.

  For example:

  A minimum of four (Oracle ASM disks) LUNS of equal size and performance is recommended for each disk group.

  Using Oracle ASM mirroring redundancy when using no hardware RAID, or when you need host-based volume management features.

  
  

  I suggest to keep in mind that ASM is not RAID. RAID uses a fixed allocation size and reflects the data at the level of the block device. ASM reflects files by using a size of distribution variable patterns of files, which gives an advantage of performance when writing to newspapers in recovery, for example.  You can combine ASM and RAID to increase performance and features, but what you choose depends on your budget and your requirements, which is anyone's guess.

  About disk groups ASM how you create, my answer is that there is no performance gain unless you can redirect i/o to separate physical disks. At a minimum, it is best to separate the DATA and FRA, but otherwise, if all the data is on the same storage device and the path of I/O, additional disk groups will not give you improve performance.

  Unless you know exactly how you size your drive includes and how you want to limit the occupation of space by a specific disk group, several groups of disks have the potential for management fees and complications, for example when executing space in one group, while having "too much" in another.

• That means recommended recommended for an ASM diskgroup to maintain production database?

  Dear Experts,

  That means recommended recommended for an ASM diskgroup to maintain production database?

  + DATA

  to keep the redo log groups redo1a, redo2a, redo3a with 3 members = redo1b, redo2b, redo3b, redo1c, redo2c, redo3c

  to organize 3 controlfiles

  organize storage of data files

  to keep the index tablespaces

  How many disk (as a point of reference)

  + FRA

  Where to place archivelogs

  Thanks in advance,

  Kind regards

  IVW

  Salvation;

  A good practice is:

  + DATA holds:

  • Redo log files, Member 1.
  • Control file.
  • Data files.

  + FRA holds:

  • Redo log files, Member of the 2.
  • Control file
  • Backup of files: archived redo logs, control automatic file backups, spills auto backups, copies, backup sets, flashback logs.

  This is the distribution of files most used; but always distribution will depend on the size of your database and how do you set up your database. Remember that you need to know how your database works. As Aman said in another post, distribution of files depends on how many disk groups you can have, depending on the speed of the disc.

  Kind regards.

  Willy

• Stripping and mirroring in ASM

  Hi all;

  I have a fundamental confusion of ASM.

  Consider that I have an A diskgroup (it contains 3 discs)

  I want to write data of the ABC in A diskgroup.

  Option 1:

  [A]   [B]  [C]| -> stripping happens here.

  Option 2:

  [ABC]   [ABC]  [ABC]-> mirrored two ways that happens here.

  ASM does support stripping and mirroring at the same time?

  Hello

  In the first option that you described, there no mirroring at all. This is the scenario when you define a group of ASM disks with external redundancy (external = do not use ASM mirroring, for example when you configure hardware RAID for redundancy). In this case it will be only interlacing (assuming you have more than 1 asm disk).

  ASM spreads data evenly across all disks in a disk group to optimize the performance and usage. This even distribution of data files eliminates the need for a regular monitoring and optimizing the performance of I/O. You can learn more about interlacing here: https://docs.oracle.com/database/121/OSTMG/GUID-114A8CA6-D757-4A70-9F68-C5133724D7BB.htm

  In the second option, you described a 3 - way mirroring (high redundancy) and 2-way no mirroring (Normal Reundncay) which means you have 3 groups of failure. Read here for more info on the mirror and failure groups implementation: https://docs.oracle.com/cd/B28359_01/server.111/b31107/asmcon.htm#BABHBHIE

  ASM does support stripping and mirroring at the same time?

  The answer is certainly yes. In the image below, for example, you can see an example of asm high redundnacy (3 - way mirroring) disk group, so it has 3 groups of failure.

  Each failure Group 2 discs. ASM spreads data evenly on these discs.

  

• Label of the ASM to name of LUN mapping; Are not even necessary device names when you use ASMLib?

  = Environment details start =.

  OS: Oracle Linux 6.4

  Version Infra grid: 11.2.0.4

  Storage: EMC VMAX

  ASM Lib version:

  # rpm - qa | grep oracleasm

  oracleasm-support - 2.1.8 - 1.el6.x86_64

  oracleasmlib - 2.0.4 - 1.el6.x86_64

  oracleasm version came not in rpm-qa output may be because it is part of the Oracle Linux kernel. But the 2 components above are not

  = Environment details end =.

  We have a 2 node RAC cluster.

  We wanted to add 2 LUNS to our data Diskgroup and asked team storage available to both nodes of the cluster.

  After that they provided, the Linux Admin scanned the LUN and powerpath device names were generated. The powerpath device names like SCSI logic unit number names are incremented sequentially. Because, two additional powerpath LUN used to Node1 to a local file system, the names of powerpath were incrementing differently in the two nodes.

  Thus, for the same LUNS emcpowerpath names differed in Node2. Linux Admin did not notice this difference. And he told us the names of newly added from Node1 LUNS powerpath that have been

  / dev/emcpowerv

  / dev/emcpowerw

  But in Node2, the same LUN were appointed

  / dev/emcpowert

  / dev/emcpoweru

  The DBA guess powerpath name these 2 LUNS will be same on Node2, label the discs to help

  oracleasm createdisk DATA17/dev/emcpowerv1

  oracleasm createdisk DATA18/dev/emcpowerw1

  After marking in node 1, the DBA ran scandisks in Node2. But while the oracleasm scandisks in Node2 is running, it does not seem to/dev/emcpowerv and/dev/emcpowerw in Node2. Instead, ASMLib is smart enough to read the LUN ID in Node1 and search devices powerpath with the same LUN ID and labels as Node1. What's it like in Node2, / dev/emcpowert is labeled DATA17 and/dev/emcpoweru is labeled DATA18 correctly, as shown below.

  Node1

  Node2

  LUN ID

  Label of the ASM

  / dev/emcpowerv

  / dev/emcpowert

  873F

  DATA17

  / dev/emcpowerw

  / dev/emcpoweru

  629:

  DATA18

  Confirmed using the oracleasm querydisk Pei/dev/emcpower < x > .

  During labelling using the oracleasm createdisk , since we use the LUN name I always thought that asmlib dealt only with the names of devices (Powerpath name in this case). I didn't know that ASMLib treats ID LUN. If ASMLIB faces in LUN ID internally, then you don't need to have the same name for a particular LUN for disks shared in different RAC nodes? Right?

  ASMLIB contains 3 packs. Oracleasmlib, which is distributed under a license of Oracle user space and is not part of the standard GPL software distribution. Oracleasm-support, which is in the standard repository and kernel driver of Oracleasm, which is included in the core Oracle UEK only or distributed by Red Hat to RHCK kernel. The subsequent RHCK driver requires a Red Hat subscription. You need all 3 installed packages.

  ASMLib doesn't care to the SCSI ID or device names. When you configure a device to ASMLIB, he writes a header special disk in the device, the command scandisk read and identify a device. ASMLIB devices are populated under the tree of device/dev/oracleasm and found by ASM using the default ORCL: device chain.

  Don't forget that ASMLIB does not work with multiple devices and if you use DB-Multipath or EMC Powerpath, you need to blacklist the physical devices so ASMLIB will use powerpath or multipath device instead. Otherwise there will be no failover devices like ASMLIB will simply use the first device found and ignore the other. Check the following:

  http://www.Oracle.com/technetwork/topics/Linux/multipath-097959.html

• Resync ASM disk in 12 c

  Hello

  Please can someone tell what is the difference between ASM resynchronization and consolidation of disk drive?

  Are they different? If so, what is re-synchronization happens and when to rebalance?

  Thank you

  Fast ASM disk Resync is used to recovery failure a failure to make temporarily unavailable group is considered transient failures.

  As the disconnection of the cable, controller failures or interruptions of supply disk or host bus adapter current.

  You just put a time (duration) where ASM Diskgroup will be to track all the changes up to recover this failure. If records are retrieved using the last State when, down disc Resync then update all the blocks on the disk recovered from survive failgroup.

  If drive failure is not picked up within the window of time, records are automatically removed from the diskgroup.

  The process of adding these drive broke diskgroup start the rebalancing process.

  Resynchronization occurs when updating the out-of-sync disk track.

  Rebalancing occurs when data are not balanced or when to add new disc on diskgroup.

• Use of the ASM.

  Team,

  You can see listed below use on ASM diskgroup. so, what's the MB all usable for the diskgroup that we can use it?

  ASMCMD > lsdg

  State Type sector Rebal to Total_MB Free_MB Req_mir_free_MB Usable_file_MB Offline_disks Voting_files name of the block

  NORMAL CLIMB N 512 4096 4194304 43008000 7227384 6144000 541692 0 DATA015 N /.

  NORMAL CLIMB N 512 4096 4194304 43008000 7186216 6144000 521108 0 DATA016 N /.

  NORMAL CLIMB N 512 4096 4194304 43008000 7178376 6144000 517188 0 DATA017 N /.

  We need to think about Usable_file_MB? or the Free_MB?

  Thank you

  Mir

  Hello

  • REQUIRED_MIRROR_FREE_MBindicates the amount of space that must be available in a disk group to restore full redundancy after worse failure that can be tolerated by the disk group. The amount of space shown in this column takes into account mirroring effects. The value is calculated as follows: for normal redundancy disk group, the value is the total gross space for all disks in the largest group of failure. The largest group of failure is the one with the largest total raw capacity. For example, if each disc in its own group of default, the value would be the size of the largest capacity disk. For a disk group high redundancy, the value is the gross total space for all disks in the two largest groups of failure.
  • USABLE_FILE_MBthe amount of free space, adjusted for mirroring, which is available for the new files to restore redundancy after a disk failure. USABLE_FILE_MBis calculated by subtracting REQUIRED_MIRROR_FREE_MB from the total free space in the disk group and then adjusting the value of mirroring. For example, in a normal redundancy disk group where by default the mirror files use disk space equal to twice their size, if there's 4 GB of usable file space, then USABLE_FILE_MB is equal to about 2 GB. You can then add a file that is up to 2 GB in size.
  • FREE_MBis the unused capacity of the Group disk in megabytes, without taking into account any imbalance data. Therefore, there may be situations where the value of the FREE_MB column shows unused capacity but because one of the ASM disks is full, database entries fail because of the imbalance in the disk group. Make sure that you initiate a manual rebalancing to force the same distribution of data which translates a precise presentation of the values in the FREE_MB column.

  See: http://docs.oracle.com/cd/B28359_01/server.111/b31107/asmdiskgrps.htm#CHDGGGAD

• dissappearing asm space

  Oracle 11.2.0.4 SE a 64-bit

  Standalone ASM 11.2.0.4 - no CARS, no grouping

  Oracle Linux 5.6

  Database had two DGs in ASM.  External redundancy, another normal redundancy with two failure groups.  A group of failure was on local SSD on the server. We have been short on space and limited by these local drives, so created a third DG in order to replace the constraint...  This one disc, external redundancy.  Disks to look like this.  The SMALL DG is the one who is forced to the internal SSD.  DG DATA2 is his replacement.

  GROUP_NUMBER DISK_NUMBER SIZE_GB REDUNC DISK_NAME FAILGROUP GROUP_NAME

  ------------ ----------- ----------- ---------- ------ -------------------- ---------------

  3 LARGE EXTERN 0 950 LARGE_0000 LARGE_0000

  4 SMALL 0 149.863281 NORMAL SMALL_FIOA SSD_GROUP

  4 SMALL 1 149.863281 NORMAL SMALL_FIOB SSD_GROUP

  4 SMALL 3 149.011719 NORMAL SMALL_FIOC SSD_GROUP

  4 SMALL 6 150 NORMAL SAN_GROUP SMALL_0006

  4 SMALL 7 150 NORMAL SAN_GROUP SMALL_0007

  4 SMALL 8 150 NORMAL SAN_GROUP SMALL_0008

  5 DATA2 EXTERN 0 500 DATA2_0000 DATA2_0000

  Yesterday, I finished moving all the data and temporary files from SMALL to DATA2.  The only files left on the SMALL are control files, which I hope to move today.

  Now the question.  I have a task that runs every morning at 0630 to gather various metrics of the use of disk file system and ASM. The result of this work is recorded and used to fill a daily tracking spreadsheet. Yesterday, I did an extra run after I finished moving all the data files (but before you move the temporary files). At this point, the SMALL DG showed 337,8 GB of usable space.  After that, I created a new temp TS (TEMP2) with its files on DATA2 and dropped the old a little.  I capture info use disk at this stage.

  This morning I got a big surprise.  The SMALL DG relates now only 74.4 GB of usable space.  I double checked its contents using "asmcmd" and verified that the only file on it's a db control file and its alias.

  The command to capture its use ASM is

  asmcmd lsdg | awk ' {print $13 "" $2 "" $7 "" $8 "" $10}' > > $logfile

  with this result:

  Name Type Total_MB Free_MB Usable_file_MB

  DATA2 / 512000 154640 154640 EXTERN

  LARGE / EXTERN 972800 33915 33915

  SMALL / NORMAL 920308 OF NTA 305895 76147

  Output power of "lsdg;

  State Type sector Rebal to Total_MB Free_MB Req_mir_free_MB Usable_file_MB Offline_disks Voting_files name of the block

  EXTERN MOUNTED N 512 4096 1048576 512000 154640 0 154640 0 N DATA2.

  EXTERN MOUNTED N 512 4096 1048576 972800 33915 0 33915 0 N GRAND.

  NORMAL CLIMB N 512 4096 1048576 920308 OF NTA 305895 153600 76147 0 N SMALL.

  I don't know if this will solve the problem, but perhaps that it was tyring:

  ALTER DISKGROUP SMALL CHECK REPAIRS;

• Assignment of names manually in ASM files

  Version of grid infrastructure: 11.2.0.4

  RDBMS Version: 11.2.0.4

  When you have an ASM filesystem, you can add a file of data using Method1 or methode2 below.

  Name the files manually using the method 1 (non - OMF) is not recommended. But I don't remember why? Anyone know?

  
  

  -Method1 (non - OMF)

  ALTER TABLESPACE CSE_TBS add DATAFILE '+ UIM_DATA/eimprod/datafile/cse_tbs_03.dbf' size 10g autoextend on next 100 m MAXSIZE 32767 m;

  -Methode2 (OMF leaving decide the file naming)

  ALTER TABLESPACE CSE_TBS add DATAFILE '+ UIM_DATA' size 10g autoextend on next 100 m MAXSIZE 32767 m;

  ASM starts is only designed to manage Oracle data files. Whether or not, you use OMF ASM knows Oracle databases and put in the right place including the optimum settings for the distribution and duplication, without worrying. When you specify the name of the data file, ASM will always create a data using OMF file, but it will create an additional ALIAS according to the name you specified. I'm not sure if not using OMF is really recommended, but it can really be any advantage and quite confusing.

• Redundancy of the Normal ASM - three stores, three groups of failure

  Hello

  I have a question about groups of failure in ASM. I have an ASM diskgroup with normal redundancy and records of three different preparations. Each storage disks are in separate groups of failure (see screen below). I know what ASM allocates to a point as it allocates a copy of copy and the primary mirror and mirror copy is placed on a different disk that is part of a group of different failure. And it is understood for the configuration with two storage (or controllers) and two different groups of failure. But what with configuration where are three storages and failure groups? My question is: If I failed three groups if new extensions are allocated are mirrored are placed on the other two discs are it is a part of a group of different failure? Or maybe this kind of configuration (with three groups of normal redundancy ASM failure) is not supported?

  

  I know what ASM allocates to a point as it allocates a copy of copy and the primary mirror and mirror copy is placed on a different disk that is part of a group of different failure. And it is understood for the configuration with two storage (or controllers) and two different groups of failure. But what with configuration where are three storages and failure groups? My question is: if I have three groups of failure, if new extensions are allocated are mirrored are placed on the other two discs are it is a part of a group of different failure? Or maybe this kind of configuration (with three groups of normal redundancy ASM failure) is not supported?

  No matter how much you failgroup. Normal redundancy requires at least 2 Failgroup.

  Oracle will allocate primary measure in a failgroup and place the mirror of the scopes in an another failgroup.

  The order does not matter since you always have a copy in a different failgroup.

  See example below (d = data extended)

  FG1 |   FG2 | FG3

  ========|===============|=========
  

  d1      |   d1_copy:

  |   d2          | d2_copy

  d3_copy |               | d3

  When you create a Diskgroup without specifying the failgroup, Oracle will automatically create a failgroup to each diskgroup. So, if you create a normal with normal redundancy Diskgroup and ASMDISK 10, ASM will create 10 failgroups.

  You have 3 warehouses and 3 Failgroup

  If you loss storage (h/f) ASM will tolerate because you measure mirror in any other failgroup (h, l, square).

  If you loss both storage (h/f) or two different storage disk/controller at the same time ASM will not tolerate. (Need to restore a backup database)

  Your point of failure: you must make sure that the two storages/controllers/disks will still be available.

  If your concern is: can I have a situation where two storage can do fail (or cut) at the same time. Then, you need high redundancy, because this configuration will tolerate you two storages loss (h/f) at the same time and Diskgroup will remain active.

• file data and the ASM disks.

  Hi experts,

  We have a situation where one of the data file has been corrupted. Can we find the related ASM disk... no request to find that the drive associated with asm (physical disk) will be useful...

  because for example I need to find the disk related asm for the data file: +disk_group_1/my_dir/my_file.dbf

  Kind regards

  Spengler

  Hi Spengler,

  In ASM we saw exactly the same as database.

  What you do is,

  Your file "+ disk_group_1/my_dir/my_file.dbf" is an alias, so go to this directory through asmcmd and ls - lt and get the original physical file named ..

  + Now check where the distribution of the is distributed for this file.

  To do this, run the command UESA (read-only)

  UESA - diskstring ''-'' dump - noimage

  This will generate a new directory.

  on the inside you will see 2 files.

  + Now open the .map file and find this filenumber, it is like F00000NNN

  Where NNN corresponds to file_number

  + See related disc of the map as N00n file, then look for this file report.txt N00n.

  -This will give you disc associated with this asm file.

  Kind regards
  Loriette