2-node RAC with 10 local drives each

Similar Questions

• Single node ESXi with large local storage

  I have a HP DL180 G6 loaded with 2 160 GB drives, in the opposite direction. I'm also 6 2 TB disks in a RAID5 configuration including a spare drive (approximately 7.4 to useful).

  What is the best way to take full advantage of the direct storage of this machine?  SAN/NAS isn't in the budget.

  Currently, I have install ESXi sitting on the 160 GB mirrored drives and thought I'd get to use all of the 7.4 to other readers as a local data store, but only the instalation disk ESXi shows upward, the other readers that have been configured to RAID5 but do not appear in vsphere.

  Thank you.

  What RAID controller do you use? E200/P400? There BBWC/FBWC?

  To be able to use the full 7.4 to with ESXi 4.1 you must divide the RAID in several logical volumes, smaller than the maximum supported size 2 TB less than 512 bytes. To do this, boot from the CD/DVD of SmartStart and run the ACU (Array Configuration Utility). Once completed, start the new ESXi and use the "Add storage" Wizard to create VMFS storage of data on the logical volumes. Depending on the maximum size of virtual disk that you plan to use, you should consider to choose the appropriate block size while creating VMFS data warehouses.

  André

• two node Oracle RAC with two Oracle databases completely different on each node?

  We currently have two different applications that use Oracle databases.  Current configuration is two different oracle homes dbhome_1, dbhome2 with separate mounting LINUX points containing the Oracle homes and *.dbf files on a single host server.

  Previously, the idea was to set up a second server identical host configured as a 'data guard' for both of these databases.  Now, management wants to implement a cluster of CARS of two nodes, with each node running two databases.  With the Standard edition.

  I only did CARS of high availability and scalability for a single database.  Is it possible to run two completely different databases Oracle on a two-node RAC cluster?  I am inclined to think that they it has divided into

  two clusters RAC to two nodes on four small servers host.  Any suggestions or comments would be appreciated?

  Is it possible to run two completely different databases Oracle on a two-node RAC cluster?

  The short answer is Yes... it is possible.  It will work without any problem, as long as you have the resources available to an additional database.

  I am inclined to think that they it has divided into two clusters RAC to two nodes on four small servers host.

  There is another option.  In the past I have two databases on a RAC cluster that has been implemented for reasons of cost of license / (I don't think that this applies to itself, but you will need to check).

  However having seen on separate hosts that you have only potentially interruptions of service on a database if you decide to do something like upgrading from one of the databases and thus upgrade the GI as well.  This would mean interruptions in service for all your databases using the GI.

• all my data is all save on the local disk c, how to share the data with the other drive, local drive d.

  all my data records on the local disk c, how to share the data with the other drive, local drive d.

  Hi Jasonbichard,

  1. what type of drive is D? Is - this another partition on the same disk?

  2 Windows operating system you are using?

  You can change the location of the disk to save the data in the d: instead of C: and check if it helps.

  a. navigate to the location (username) C:\Users\.
  b. right click on the folder that you want to change the location, and then select Properties.
  c. click on the location tab and change the location to D: drive.

  d. click on apply and Ok.

• Two-node Rac (11204) Cluster node 1 has failed to stop with error: impossible to get in touch with the loan of cluster service. duty to force stop

  Newly build Production environment (don't use it yet)

  OS: Linux redhat 64-bit 2.6.18

  Cluster version: 11.2.0.4

  This environment clusterware installed last December, we are trying to install oracle RDBMS, so try to first stop the crs.  However on node 1, the v$ asm_diskgroup shows nothing under total_mb /free_mb on diskgroup OCR.  and stop the crs shows: impossible to get in touch with the loan of cluster service.

  Alerts cluster log shows below:

  2014-03-25 03:50:01.429:

  [crsd (8608)] CRS-1013: the location of the OCR in ASM disk group is inaccessible. Details in u00/app/11.2.0.4/grid/log/oprd100/crsd/crsd.log.

  2014-03-25 03:50:01.433:

  [crsd (8608)] CRS - 0804:Cluster Ready Service interrupted due to the Oracle Cluster registry error [PROC-26: error when accessing the physical storage]

  ]. Details at (: CRSD00111 :) in u01/app/11.2.0.4/grid/log/orpd100/crsd/crsd.log.)

  2014-03-25 03:50:02.123:

  [ohasd (12490)] CRS - 2765:Resource 'ora.crsd' failed on the server "orpd100".

  2014-03-25 03:50:03.407:

  [crsd (8623)] CRS-1013: the location of the OCR in ASM disk group is inaccessible. Details in u01/app/11.2.0.4/grid/log/orpd100/crsd/crsd.log.

  2014-03-25 03:50:03.411:

  [crsd (8623)] CRS - 0804:Cluster Ready Service interrupted due to the Oracle Cluster registry error [PROC-26: error when accessing the physical storage]

  The NEWSPAPER of the ASM alerts:

  Wed Mar 25 03:21:49 2014

  WARNING: Waited 15 seconds IO to write to the disk of PST 1 in Group 1.

  WARNING: Waited 15 seconds to write IO to disk of PST 2 in Group 1.

  WARNING: Waited 15 seconds IO to write to the disk of PST 1 in Group 1.

  WARNING: Waited 15 seconds to write IO to disk of PST 2 in Group 1.

  Wed Mar 25 03:21:49 2014

  NOTE: process _b000_ + asm1 initiating (21071) 1.1807368888 disk offline (OCR_0681_2EF4) with mask 0x7e in Group 1

  NOTE: process _b000_ + asm1 initiating (21071) 2.1807368889 disk offline (OCR_0681_2EF5) with mask 0x7e in Group 1

  NOTE: check PST: grp = 1

  GMON seeking ways of disc for Group 1 to 5 pid 27, DiSo 21071

  ERROR: read no quorum within Group: required 2 found 1 records

  NOTE: audit PST grp 1 fact.

  NOTE: start the PST update: grp = 1, dsk = 1/0x6bba42b8, mask is 0x6a, op = clear

  NOTE: start the PST update: grp = 1, = 2/0x6bba42b9 dsk, mask is 0x6a, op = clear

  GMON update modes of disk for the Group 1 to 6 for 27, DiSo 21071 pid

  ERROR: read no quorum within Group: required 2 found 1 records

  Wed Mar 25 03:21:49 2014

  NOTE: disassembly of the cache (not clean) Group 1/0x35AAB27B (OCR_DATA)

  WARNING: For disk offline OCR_0681_2EF4 0x7f mode failed.

  2 node rac cluster is normal.  We were able to stop the crs without force.

  What should I look at to understand what is happening here?

  Thanks to adavnace.

  WARNING: For disk offline OCR_0681_2EF5 0x7f mode failed.

  NOTE: e-mail CKPT suspend pins Unix process pid: 21073, image:

  No default value is 1 M

• Problem when executing the processors in the nodes with disabled local storage

  Hello world
  
  We test the coherence JSR107 (a.k.a. JCache) implementation of Yannis (https://github.com/yannis666/Coherence-JSR-107) and we found the following issue when you work with a local storage where distributed cache node is disabled (using - Dtangosol.coherence.distributed.localstorage = false). Yannis implementation uses several EntryProcessors for the execution of operations get() or put () in the cache. For example, the following code is used to update a value in the cache:
  
  / public class PutProcessor implements InvocableMap.EntryProcessor {}
  
  @Override
  public Object process (input InvocableMap.Entry) {}
  BEntry entry = (BinaryEntry) BinaryEntry;
  return bEntry.isPresent ()? bEntry.getBinaryValue (): null;
  }
  
  }
  
  And put operatino is taken from the following code:
  
  namedCache.invoke (key, CPU)
  
  where processor argument contains a reference to the PutProcessor object.
  
  
  When we call JSR107 performs surgery put in the cache and code above we get following exception:
  
  java.lang.NullPointerException
  to com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.partitionedService.PartitionedCache$ Storage.invoke (PartitionedCache.CDB:10)
  at com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.partitionedService.PartitionedCache.onInvokeRequest(PartitionedCache.CDB:52)
  to com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.partitionedService.PartitionedCache$ InvokeRequest.run (PartitionedCache.CDB:1)
  at com.tangosol.coherence.component.net.message.requestMessage.DistributedCacheKeyRequest.onReceived(DistributedCacheKeyRequest.CDB:12)
  at com.tangosol.coherence.component.util.daemon.queueProcessor.service.Grid.onMessage(Grid.CDB:34)
  at com.tangosol.coherence.component.util.daemon.queueProcessor.service.Grid.onNotify(Grid.CDB:33)
  at com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.PartitionedService.onNotify(PartitionedService.CDB:3)
  at com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.partitionedService.PartitionedCache.onNotify(PartitionedCache.CDB:3)
  at com.tangosol.coherence.component.util.Daemon.run(Daemon.CDB:42)
  at java.lang.Thread.run(Thread.java:662)
  
  Of course we have previously started another node in the cluster with local storage enabled.
  
  It seems that, as no storage card is present at the node, a NullPointerException is thrown when you try to insert the entry.
  
  Could someone give advice on this problem and how to overcome? This implementation is based on the processor is not compatible with the knots of localstorage cache disabled?
  
  Thank you in advance,
  
  Daniel C.

  Hello

  It sounds like COH-7180 - demand deserialization failure results in a cryptic NullPointerException, that has been fixed in one of the 3.7.1 patches. The real problem is that your entryprocessor cannot be deserialized.

• How to get a list of the mapped local drives on a machine with VB RDP?

  Hi all

  First of all thanks for reading by the problem.

  Challenge:

  To build an application (VB) that lists all local drives on a RDP machine.

  Environment:

  Local Machine XP Professional, Local activated sharing disks

  See the list of drives when I type the \\tsclient in the RDP session.

  What I'm trying?

  Win32_Share,Win32_MappedLogicalDisk. But no luck.

  Kind regards

  Shashidhar

  Hello

  Your question is more complex than what is generally answered in the Microsoft Answers forums. It is better suited for Visual Basic on MSDN.

  Please post your question in the MSDN Forums.

  Diana

  Microsoft Answers Support Engineer

  Visit our Microsoft answers feedback Forum and let us know what you think.

  If this post can help solve your problem, please click the 'Mark as answer' or 'Useful' at the top of this message. Marking a post as answer, or relatively useful, you help others find the answer more quickly.

• Need advice from configuration to the flash cache multi-node RAC/ASM

  We have a database of CARS to work two nodes. We install a new fusion 640 GB flash card in each node. The is a double card, so that it shows up as two devices: / dev/fioa and/dev/fiob on each node.
  We have created two groups of private drives, one on each node. The Group of disks on node 1 is '+ FLASHCACHE' and the disk on node 2 group is "+ FLASHCACHE2".
  
  We have not found specific examples of how to configure the parameters of the init for flashcache on RAC/ASM. We guess that we must do the following. If anyone would be critical and confirm or suggest something different, we would appreciate it very much!
  
  "prod1.db_flash_cache_file ='+ FLASHCACHE/flash1"
  prod2.dg_flash_cache_file = '+ FLASHCACHE2/flash2'
  * .db_flash_cache_size = 600G
  
  Thanks for any advice.
  
  Jim
  
  Published by: jimbev90 on January 5, 2012 08:18

  Hi Jim,.

  prod1.db_flash_cache_file='+FLASHCACHE/flash1'
  prod2.dg_flash_cache_file='+FLASHCACHE2/flash2'
  *.db_flash_cache_size=600G
  

  Your configuration is correct.

  Smart Flash Cache database is not shared across RAC nodes. It is private to each RAC instance. For this reason, each RAC instance must define its own local cache of flash file path. When the flash cache is managed by ASM, a separate diskgroup is necessary for each instance flash cache.

  You can use ALTER SYSTEM for db_flash_cache_size set to zero to disable the flash cache. Alternatively, you can use ALTER SYSTEM to deliver the flash cache to its original size to reactivate it.

  Read this white paper Oracle database Smart Cache Flash
  http://www.Oracle.com/technetwork/articles/systems-hardware-architecture/Oracle-DB-smart-Flash-cache-175588.PDF

  Kind regards
  Levi Pereira

• ORA-27504: error of IPC create context OSD - error on nodes RAC times

  Hello
  
  I did Rman cloning Linux 3 (32-bit) for Linux 5 (64 - bit) with success. Then settings of cluster enabled in pfile (two-node Rac) and autoconfig enabled for two databases ORACLE_HOME because it's the Oracle Applications database.
  
  VERSION OF DB - 10.2.0.2
  CRS - 11g
  Linux version - Red Hat Enterprise Linux Server version 5.4 (Tikanga)
  Application version - 11.5.10.2
  
  I am new to cars and this database is up and working now, I see that following error messages in the alert log of the two instances at each start
  
  **********
  Thu Jan 20 07:33:35 2011
  Starting ORACLE instance (normal)
  LICENSE_MAX_SESSION = 0
  LICENSE_SESSIONS_WARNING = 0
  Thu Jan 20 07:33:35 2011
  Errors in the /u01/app/oracle/admin/udev/udump/udev1_ora_28252.trc file:
  ORA-27504: create context of OSD IPC error
  ORA-27300: OS dependent operating system: skgxnqtsz failed with status: 0
  ORA-27301: OS Error Message: error 0
  ORA-27302: an error: SKGXN not av
  clsssinit ret = 21
  interconnection there is no OCR information
  WARNING: No cluster interconnect has been specified. According to the
  the communication configured cluster traffic Oracle driver
  may be addressed to the public interface of this machine.
  Oracle recommends that the RAC grouped be configured databases
  with a simple soldier of interconnection for security improved and
  performance.
  SNA system picked latch-free 3
  With the help of LOG_ARCHIVE_DEST_1 parameter value default as/u01/app/oracle/product / 10 g 2, db, dbs, arch
  Autotune undo retention is enabled.
  LICENSE_MAX_USERS = 0
  ****************
  
  Please let me know if I have to put the DB services to the cluster.
  
  Thank you
  Muhammad Tanveer

  Hello

  11.2 or 11.1 clusterware is?

  If it's 11.2 see Please note:

  Pre 11.2 Database issues in environment Infrastructure grid 11 GR 2 (Doc ID 948456.1)
  https://supporthtml.Oracle.com/EP/faces/secure/km/DocumentDisplay.JSPX?ID=948456.1

  Concerning
  Sebastian

• I can connect to my network, but with access "local only".

  Access to the "Local" network only
  I can connect to my network, but with access "local only".  Internet became more intermittent (not sure if that is related or if I guess it) and did not work in several weeks.  Desktop computer is plugged into the router, but have the same problem when it is plugged into the DSL modem.  Other computers on the same router (wireless or other) work very well.  Recently, I removed Mcaffee and installed MS security essentials.  I uninstalled mcaffee of programs and settings then used the mcaffee removal tool to lighten the rest.  I uninstalled and reinstalled MS security essentials.  I thought it might be a firewall issue, but I get error 0x6D9 when I try to start it.  I tried all the steps in the following post, nothing helps.  Any other suggestions?

  ______________________

  You can follow the steps below and check if that helps you solve the problem.

  Method 1

  Try to power cycle the router and the computer and check if it helps.

  On the PC:

  1. Save your work and restart the machine.

  On the router or modem (if wireless printing):

  1. Unplug the router and the modem.
  2. Wait 30 seconds.
  3. Plug in the modem and wait for it to come to the ready state.
  4. Plug in the router.

  After you put cycle check the connection between the router and the computer.

  Method 2

  If the steps above do not help, you mayreset TCP/IP stack. To reset access the link below and either click on "Fix it for me" or follow the instructions to fix it yourself:http://support.microsoft.com/kb/299357
   
  Disable the IP helpdesk:
  1 hold the Windows key and type R, type "services.msc" (without the quotes) and press enter
  2. scroll down to the IP assistance service, right-click on it and select Properties
  3. in the drop-down list box that says "Automatic" or "Manual", set it to disabled and then click 'apply '.
  4. then click on "Stop" to stop the service from running in the current session
  5. click on OK to exit the dialog box

  Method 3

  Disable IPv6 and remove IPv6 virtual cards:
   
  Try to uninstall IPv6 on all interfaces, the removal of virtual cards of IPv6 and reset the TCP/IP stack. To remove the IPv6, go to the properties for each network adapter, and deselect the check box next to the Protocol "Internet Protocol version 6 (TCP/IPv6), which will turn off, or select it and click on uninstall, which withdraw power off the computer.» Then go into Device Manager and remove any 4to6 adapters, adapters miniport WUN or tunnel adapters.
  NOTE: You should do this for each network connection, even if they are disabled.

  Method 4

  Disable the DHCP Broadcast Flag:
  Link: http://support.microsoft.com/default.aspx/kb/928233
  Important: This section, method, or task contains steps that tell you how to modify the registry. However, serious problems can occur if you modify the registry incorrectly. Therefore, make sure that you proceed with caution. For added protection, back up the registry before you edit it. Then you can restore the registry if a problem occurs. For more information about how to back up and restore the registry, click on the number below to view the article in the Microsoft Knowledge Base:
  http://support.microsoft.com/kb/322756.  How to back up and restore the registry in Windows
   
  Windows Vista cannot obtain an IP address from certain routers or some non-Microsoft DHCP servers
   
  To resolve this issue, disable the DHCP BROADCAST flag in Windows Vista. To do this, follow these steps:

  
  1. Click Start, type regedit in the search box, and then click regedit in the list programs.
  2. If you are prompted for an administrator password or for confirmation, type your password, or click on continue.
  3. Locate and then click the following registry subkey:
  4 HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\ {GUID}
  5. in this registry path, click the (GUID) subkey that corresponds to the network adapter that is connected to the network.
  6. on the Edit menu, point to new, and then click DWORD (32-bit) value.
  7. in the new area #1, type DhcpConnEnableBcastFlagToggle and press ENTER.
  8. right click on DhcpConnEnableBcastFlagToggle, then click on modify.
  9. in the value data box, type 1 and then click OK.
  10. close the registry editor.
   
  By setting this registry key to 1, Windows Vista's trying to get an IP address using the BROADCAST flag in DHCP Discover packets. If that fails, he will try to obtain an IP address without using the BROADCAST flag in DHCP Discover packets.
  You can also try uninstalling and reinstalling the driver for the wireless card.

  ________________

  Thanks for any help!

  Hello

  All changes to the software or hardware of the computer?

  Method 1:

  Visit the link below and follow the steps.

  Of network connectivity status incorrectly as 'Local' only on a Windows Server 2008 or Windows Vista-based computer that has multiple network cards

  http://support.Microsoft.com/kb/947041

  Method 2:

  Update the NIC drivers and check.

  Network adapter problems

  http://Windows.Microsoft.com/en-us/Windows-Vista/troubleshoot-network-adapter-problems

• How can I share local drives exclusively between my 4 computers on the network?

  Hello
  Now I'm sharing local drives between my computers setting to share with everyone and which works very well but I'm worried about my brothers and sisters, to access my files.
  I want my 4 computers to share their local hard drives exclusively between them so, how can I do?
  I tried to add permissions for each computer individually but every time I have to try give me a message "the next object is not a domain listed in the Select location dialog box, and is therefore not valid" here is a screenshot of this:
  http://img502.imageshack.us/img502/4425/82018296.PNG
  
  
  And btw my computers all have the same user name and password are.
  Thanks for the help I appreciate it sincerely from the bottom of my heart =)

  In the network and sharing settings / advanced sharing settings / turn sharing protected by password. Use a unique name for the corresponding user password on your four computers.
  Set the Public sharing or disable as your prefer for your public folders and organize your files accordingly.

• Connection speed slow on one node to a node RAC 2 put in place

  Hi all

  Having a little trouble with one of the nodes in a node 2 CARS set up
  We have sent traffic directly to the node in question (using a service that runs only on node 1)

  Forced connection to this node (sqlplus) since an application server remote takes 3 times as long does if we force fitting to the second node RAC.

  If I try to connect locally - connection speed is instantaneous-

  There is NO problem with the server performance

  There is NO problem with the perfromance instance / database - it is only noticed when the initial connection

  I've been in the network of the node and can see nothing wrong - so I was wondering if there was anything you guys might have experienced while the connection time is slow for a single node only (949 ms - compared to 247 ms on node 2)

  If anyone out there is interested or falls on this issue

  The problem has proved to be the mount point options MP Linux Volumes

  I had configured them as part of best practices for RAC / NetApp / Direct NFS - hjowever I also got the binaries for both nodes in order to have greater availability.

  Thereffore I would have had the Mount for NŒUD UNIQUE options.

  The change was to

  -stop the instance

  -unmoiunt binary files

  -change the following on fstab

            NFS rw, bg, hard, nointr, rsize is 32768, wsize = 32768, tcp, to = 3, timeo = 600, actimeo = 0 0 0

  TO

            NFS rw, bg, hard, rsize = 32768, wsize = 32768, to = 3, nointr, timeo is 600, tcp

  -start the instance and repeat on node 2

• voltage of local drives...

  Hello

  Maybe someone could give me a hint on this!

  I want to, for a short time only, add four hard drives 146 GB each to one of my servers to ESX 4.0. I would like to create a store of data on all four disks. Unfortunately I can't create one, then spread with the other three in the GUI. I guess because they are local disks. I used rather vmkfstools and spans two partitions, I created first using the GUI. Up to present, so good, but I cover more than two?

  Is there a way to cover or expand all four local drives in a data store?

  Thanks for your help

  Harro Nehslen

  You should be able to do that, if I understand the question correct.

  I have created 8 x 1 TB LUN and presented to an ESX host.

  Then, I created a data store and all LUNS to end up with a single TB 8 data store added. (I know that this is not recommended, but it works)

  If the server see a 1 to LUN or a 140 GB drive, should be the same.

  The only difrence will be LUN access through the same HBA and im not sure what is your hardware configuration.

  I guess this is a server with a Raid controller card installed.

  Create a single logical Volume of the 4 disks using raid with RAID 5 preference

  The ESX should see 1 big Disk and you create your data store

  Concerning

  If you find this or any other answer useful please consider giving points by checking the answer useful or appropriate. Thank you.

• My computer with two hard drives and 2 starting systems has the car crash in the solid state. When I got the new drive and tried to restore from time capsule, I lost some data. How can I see what backups are on the time capsule? My laptop is also backed u

  My computer with two hard drives and 2 starting systems has the car crash in the solid state. When I got the new drive and tried to restore from time capsule, I lost some data. How can I see what backups are on the time capsule? My laptop is also saved to the time capsule. Everything is in 'Data' when I look through the Airport utility. I think that the information that I've lost are still there since I was a Capsule 3 to.

  Time Machine backup to a sparsebundle.

  The sparsebundle will use the share name of computer. So, it should be obvious who is your Macbook and that is your Macpro.

  Open the sparsebundle in Finder and keep open the subfolders until you get to the backup of each separate hard drive...

  You has of course set the exclusion in Time Machine?

  By default, Time Machine will exclude your second drive unless you have included it. Did you?

  Once you get to this level during the backup, you can do a full restore manually whenever you want.

• I have lost my original photos from my local drive if I optimized my iCloud library?

  I moved my photo (900 GB) library on an external drive. The value in pictures new external player as the new library system library. Now when I turn on iCloud it tells me I have to optimize my photos. I'm afraid of losing my originals from my local drive. Optimization of my library will erase my originals from my local computer? I have to delete my old library in iCloud that has already gone through the pain staking the download process (759GB) in order to reuse my iCloud? I try to avoid to keep my computer another 3 months. Simply cross ICloud cannot check the new library with my iCloud library and move on. How can I start again using my iCloud? Please give me some advice or assistance. Thank you

  The value in pictures new external player as the new library system library. Now when I turn on iCloud it tells me I have to optimize my photos.

  What is your external drive only has 1 TB of storage?  You have found the warning because the 900 GB in iCloud and the size of your current library on the external hard drive will not fit on the disc. If you have a disc, use a drive twice the size of your library of Photos, so you will not miss storage in the years to come. Any drive that you use with pictures must have Mac OS Extended (Journaled) file system.

  iCloud library does not know, the pictures are identical. It will download all the photos in the library moved to iCloud and check if they are duplicates. And it only allows you to introduce you, if you use 'Optimize', if the combined size of the photos in iCloud and on your Mac is larger than the available storage.

  And Yes, the "optimize" option can cause some originals will be deleted. It is completely dynamic.  You can turn this option off again directly after that you turned the library on the external drive as your iCloud photo library for re-download all original, and before Photos was able to start optimization. Do it only if you are sure, the library will fit.

  If you do not all projects in your photo library and did not use the function "Faces" and has invested a lot of work by naming the faces, you could simply start with an empty library on your external drive and let the photos to download icloud to the empty library. It will be faster to download again and the merger of the library moved in your library to iCloud. The only downside, the tags of faces will be missing and your books and other projects. If you need to keep a backup copy of the library has moved.