Home -> Mailing Lists -> Oracle-L -> RE: CVU reporting error when checking for ASM storage

RE: CVU reporting error when checking for ASM storage

From: Hameed, Amir <Amir.Hameed_at_xerox.com>
Date: Tue, 27 Jan 2015 16:48:25 +0000
Message-ID: <AF02C941134B1A4AB5F61A726D08DCED0E1376F2_at_USA7109MB012.na.xerox.net> 





Hi Kevin,


I am working in the lab at the moment and trying to figure out the best way to implement ASM. I worked with Riyaj on the CVU issue and after he validated my configuration, I proceeded with the grid installation with OCR and Voting disk placement on ASM, which is where the CVU was giving me the error. It turned out that the CVU error was a false positive. At the moment, my ASM disks for OCR and Voting disks are not partitioned, which is where I had a lot of confusion because of the different feedback that I was getting. So, it does not seem that the block devices have to be partitioned on Linux.
I do see recommendations from Nitin Vengurlekar in his ASM book in section “ASM Storage Device Configuration”:



The alignment issue would be solved if we could start the ASM disk at block 0 of the LUN, but that does not work on some operating systems (Solaris, in particular). On Linux, you could start the ASM disk at block 0, but then there is a chance an administrator would run fdisk on the LUN and destroy the ASM disk header. Therefore, we always recommend using a partition rather than starting the ASM disk at block 0 of the LUN.



So, it seems that on Linux, ASM disk can start at block zero and does not have to be partitioned. But it might be a good idea to partition it just in case we are dealing with rookies in the data center and someone decides to reuse the disk. I am still deciding whether to partition disks or not.



From: Kevin Closson [mailto:ora_kclosson_at_yahoo.com]
Sent: Tuesday, January 27, 2015 11:30 AM
To: Hameed, Amir; Riyaj Shamsudeen; david.barbour1_at_gmail.com
Cc: dmarc-noreply_at_freelists.org; oracle-l_at_freelists.org
Subject: Re: CVU reporting error when checking for ASM storage



Hi Amir,



   So are you still having these difficulties? Did you go with a single partition perhaps just to ensure proper alignment?






From: "Hameed, Amir" <Amir.Hameed_at_xerox.com<mailto:Amir.Hameed_at_xerox.com>>
To: Riyaj Shamsudeen <riyaj.shamsudeen_at_gmail.com<mailto:riyaj.shamsudeen_at_gmail.com>>; "david.barbour1_at_gmail.com<mailto:david.barbour1_at_gmail.com>" <david.barbour1_at_gmail.com<mailto:david.barbour1_at_gmail.com>>
Cc: "dmarc-noreply_at_freelists.org<mailto:dmarc-noreply_at_freelists.org>" <dmarc-noreply_at_freelists.org<mailto:dmarc-noreply_at_freelists.org>>; "oracle-l_at_freelists.org<mailto:oracle-l_at_freelists.org>" <oracle-l_at_freelists.org<mailto:oracle-l_at_freelists.org>>
Sent: Thursday, January 22, 2015 10:09 AM
Subject: RE: CVU reporting error when checking for ASM storage



I would like to add that I got the same CVU error when I tried the same command on a partitioned disk. I am not sure if I am the only one who has seen this CVU error or others have also encountered it.




From: Riyaj Shamsudeen [mailto:riyaj.shamsudeen_at_gmail.com]
Sent: Thursday, January 22, 2015 12:29 PM
To: david.barbour1_at_gmail.com<mailto:david.barbour1_at_gmail.com>
Cc: Hameed, Amir; dmarc-noreply_at_freelists.org<mailto:dmarc-noreply_at_freelists.org>; oracle-l_at_freelists.org<mailto:oracle-l_at_freelists.org>
Subject: Re: CVU reporting error when checking for ASM storage



There is no requirement to partition the disk, AFAIK, for Grid install. Amir and I worked this problem offline, and it seems that CVU has a bug. Proceeding with that useful "ignore all" button during Grid installation, completes successfully.



However,  in Solaris platform, disk must be partitioned though.



Cheers



Riyaj Shamsudeen


Principal DBA,


Ora!nternals -  http://www.orainternals.com<http://www.orainternals.com/> - Specialists in Performance, RAC and EBS
Blog: http://orainternals.wordpress.com/

Oracle ACE Director and OakTable member<http://www.oaktable.com/>
Co-author of the books: Expert Oracle Practices<http://tinyurl.com/book-expert-oracle-practices/>, Pro Oracle SQL, <http://tinyurl.com/ahpvms8> Expert RAC Practices 12c.<http://tinyurl.com/expert-rac-12c> Expert PL/SQL practices<http://tinyurl.com/book-expert-plsql-practices>




On Fri, Jan 16, 2015 at 12:26 PM, David Barbour <david.barbour1_at_gmail.com<mailto:david.barbour1_at_gmail.com>> wrote:
Amir - You missed what Mladen and Tim are pointing out.  /dev/<whatever>/dm-50 is a block device.  What Mladen refers to as "scusify", Tim shows with the output:



udev_rules_get_name: add symlink 'disk/by-path/pci-0000:00:0d.0-scsi-1:0:0:0-part1'



Partition the disk using parted or (deprecated) fdisk.  You can make the name available either through udev rules or via friendly names in the multipath.conf file.



By the way - welcome back Mladen.




On Fri, Jan 16, 2015 at 2:13 PM, Hameed, Amir <Amir.Hameed_at_xerox.com<mailto:Amir.Hameed_at_xerox.com>> wrote:
Thanks Mladen.


I am following Red Hat’s note https://access.redhat.com/solutions/272153 titled “How to create Oracle ASM disks using DM Multipath devices in Red Hat Enterprise Linux?” to define UDEV rules on block devices. What Tim has documented uses SCSI device names in the rule instead of Device Mapper. This is the first time I am doing ASM. I am going to carry on with the installation to see how it goes and it might be an issue with CVU. This is what is listed in Red Hat’s document.



    
  
  1.   udevadm info --query=all --name=/dev/mapper/mpathN |grep -i DM_UUID

  
  2.   ACTION=="add|change", ENV{DM_UUID}=="mpath-<UUID>", SYMLINK+="oracleasm/asm01", GROUP="dba", OWNER="oracle", MODE="0660"






From: oracle-l-bounce_at_freelists.org<mailto:oracle-l-bounce_at_freelists.org> [mailto:oracle-l-bounce_at_freelists.org<mailto:oracle-l-bounce_at_freelists.org>] On Behalf Of Mladen Gogala
Sent: Friday, January 16, 2015 2:47 PM


To: oracle-l_at_freelists.org<mailto:oracle-l_at_freelists.org>
Subject: Re: CVU reporting error when checking for ASM storage



Amir, you cannot just add Linux block devices to ASM. You need to "scsify" them. The best brief description is on Tim Hall's ORACLE-BASE site:



http://oracle-base.com/articles/linux/udev-scsi-rules-configuration-in-oracle-linux-5-and-6.php






On 01/16/2015 02:00 PM, Hameed, Amir wrote:
Hi,


I am installing Oracle Grid 12.1.0.2 on RHEL 6.5. When the Prerequisite Checks phase of OUI, it reported PRVF -5150 on ASM disks. When run CVU on the ASM disks, it reports an error as shown below. Below is an out from CVU with debugging enabled when it was run for one RAC node only:
CV_TRACELOC=/tmp/cvutrace


./runcluvfy.sh comp ssa -n <node-name> -s /dev/oracleasm/grid/asmgrid02




Verifying shared storage accessibility



Checking shared storage accessibility...



ERROR:  /dev/oracleasm/grid/asmgrid02


Storage operation failed




Shared storage check failed on nodes "usa0300lx566"



Verification of shared storage accessibility was unsuccessful on all the specified nodes.





The CVU trace file shows the following error:
OUTPUT><CV_ERR><SLOS_LOC>CVU00101</SLOS_LOC><SLOS_OP></SLOS_OP><SLOS_CAT>OTHEROS</SLOS_CAT><SLOS_OTHERINFO>Cannot locate disk for '/dev/dm-50'</SLOS_OTHERINFO></CV_ERR><CV_VRES>0</CV_VRES><CV_LOG>Exectask:getstorage success.</CV_LOG><CV_CMDLOG><CV_INITCMD>/tmp/CVU_12.1.0.2.0_oracle/exectask -getstorage /dev/oracleasm/grid/asmgrid02 </CV_INITCMD><CV_CMD>realpath /dev/oracleasm/grid/asmgrid02</CV_CMD><CV_CMDOUT> /dev/dm-50</CV_CMDOUT><CV_CMDSTAT>0</CV_CMDSTAT><CV_CMD>stat /dev/dm-50</CV_CMD><CV_CMDOUT> stat.st_mode:61B0</CV_CMDOUT><CV_CMDSTAT>0</CV_CMDSTAT><CV_CMD>fopen /proc/partitions</CV_CMD><CV_CMDOUT>    8        0  291991552 sda



I am using udev rules and the rule defined is shown below:
ACTION=="add|change", ENV{DM_UUID}=="mpath-360000970000192606068533031464134", SYMLINK+="oracleasm/grid/asmgrid02", GROUP="asmadmin", OWNER="oracle", MODE="0660"



The Device alias exists as shown below:


$ ls -ltr /dev/oracleasm/grid/asmgrid02





The multipath device has proper permissions set:
cd /dev/oracleasm/grid


$ ls -ltr




total 0


lrwxrwxrwx 1 root root 11 Jan 16 13:37 asmgrid02 -> ../../dm-50



$ ls -ltr ../../dm-50




brw-rw---- 1 oracle asmadmin 253, 50 Jan 16 13:37 ../../dm-50



Any idea what might be wrong here?



Thanks,


Amir




--



Mladen Gogala



Oracle DBA



http://mgogala.freehostia.com<http://mgogala.freehostia.com/>




--
http://www.freelists.org/webpage/oracle-l


Received on Tue Jan 27 2015 - 17:48:25 CET












Original text of this message