PowerVS IBMi Backups Tutorial v1
PowerVS IBMi Backups Tutorial v1
Virtual Server
An IBM Systems Lab Services Tutorial
Cesar Salinas
Vess Natchev
[email protected]
TABLE OF CONTENTS
Introduction................................................................................ 1
Use Cases .................................................................................. 1
Full-system Snapshot and Restore ........................................... 1
Requirements ........................................................................ 3
Diagrams ................................................................................... 5
BRMS integration to move backup to COS ................................. 5
Examples ............................................................................. 29
Introduction
Use Cases
Full-system Backups
Here we will show how to perform full-system saves and restores using
Backup, Recovery and Media Services (BRMS) and IBM Cloud Object
Storage (COS).
Object-level Backups
Components
2
Components to IBM i backup in the Cloud
Requirements
3
Cloud. In our scenario we have created two PowerVS locations, one is Toronto and one
in London datacenters.
Prior to provisioning Power VSI in the PowerVS location, you will need to create at least
one subnet. You can have as many subnets as you require in each PowerVS location
service on which you can provision your Power VSIs.
This solution requires that product IBM Cloud Storage Solutions for i is
installed on the system and the following BRMS PTF(s) must be
installed on a stand-alone BRMS system or on all the systems in a
BRMS network:
• 7.3 - SI61153
• 7.2 - SI61152
• 7.1 - SI61151
IBM Cloud Storage Solutions for i Direct Link (DL) Reverse Proxy
Server Support. Support Google Cloud Storage.
5733ICC V1R2M0
4
BRMS Recovery Reports
Diagrams
5
IBM i Backup in the Cloud
6
Create PowerVS location Services and Subnet(s)
7
Under Select Region, choose your region. You are limited to only one service per region.
8
Your PowerVS location service will now appear under the Services tab.
You will repeat this process to create a second PowerVS location service. In our case we
have two PowerVS location services, one in London and one in Toronto.
Next you will need to click on the PowerVS location Service you created and provision a
subnet to be used by your Power VSI servers.
9
Choose “Subnets” from the menu on the left.
10
There should be a VLAN ID associated with the subnet.
At this point, you will need to open a Support Ticket with Power System to request that
the subnet be configured to allow local communication between any Power VSI you
create in this PowerVS location service. Provide your PowerVS location service location,
and your subnet in the ticket.
Without this step, the Power VSI you create will not be able to ping between each other
even if they are on same subnet in the same PowerVS location.
11
Click on the service for datacenter in which you have created a PowerVS location power
service. In this case we will choose Toronot01 PowerVS location service.
Since we have already provisioned several VSIs, we see the list show above. If you are
creating VSIs for the first time, your list will be empty.
Press “Create Instance” on upper right-hand side.
12
This is where you provision AIX or IBM i VSIs.
Choose a name for your VSI, i.e., AIX-72-Tor01 and select how many VSIs you need to
configure. The names of the VSI will be appended with a “-1”, “-2” etc. if you select
more than one VSI.
You may leave VM pruning and SSH key as is since the VSIs will have no passwords
when you create them for the first time. You will need to create a password via the OS
command.
13
On IBM i VSIs, scroll down and choose “IBM i Cloud Storage Solutions” in order to
perform backups using Backup Recovery and Media Services (BRMS) to Cloud Object
Storage (COS).
14
Choose the subnet you wish to attach, and the press “Attach”
Choose “Public networks” if you wish to attach to a public network, and change to “On”
15
Now check the box “I agree to the ….” And press “create Instance” in lower right-hand
side.
Your VSI is now being provisioned.
16
Select “Direct Link Connect on Classic”.
17
Now choose “Order Direct Link Connect” from top right-hand side.
18
In our case, we decided to use Local Routing and then order a Vyatta Gateway in each
PowerVS location and provision a GRE tunnel end-to-end.
19
After this is complete, you will then be contacted by support and
requested to complete and answer some questions in an attached
document and send it back as attachment to the same ticket.
After this step is complete, support will request that you open a new
IBM support ticket and address it to the Power System. Include the
information in the original DL ticket. This new ticket will be sent to the
PowerVS location support to configure their side of the DL connection.
This should be the last step before DL communication works. You can
test your connection by pinging IBM Cloud Linux/Windows VSI from
your Power VSIs and in reverse.
We used the official IBM cloud procedure to configure the reverse-proxy server.
https://cloud.ibm.com/docs/direct-link?topic=direct-link-using-ibm-cloud-direct-link-to-
connect-to-ibm-cloud-object-storage
• You will need to first provision a Centos VSI in IBM cloud with both public and
private interface.
• Login to the VSI.
• Upgrade your operating system OS (yum update).
• Install the EPEL repository (yum install epel-release).
• Install NginX (yum install nginx).
• Start nginx (systemctl start nginx or just nginx)
• To allow service to run after reboot: systemctl enable nginx
20
https://cloud.ibm.com/docs/direct-link?topic=direct-link-using-ibm-cloud-direct-link-to-
connect-to-ibm-cloud-object-storage
This command will generate a file called acstest.key in your present directory.
Next:
Item in blue is from the previous command output file name.
Items in red you will need to provide with your own information.
openssl req -new -newkey rsa:2048 -nodes -keyout acstest.key -out acstest.csr
21
Please enter the following 'extra' attributes
to be sent with your certificate request
A challenge password []:test123
An optional company name []:
[root@centos-reverseproxy-tor01-fg ~]#
Next:
Items in blue are two files generated from previous command.
Now this command will generate the .crt file which you need.
openssl x509 -req -sha256 -days 365 -in acstest.csr -signkey acstest.key -out
acstest.crt
• cp acstest.key /etc/pki/tls
• cp acstest.crt /etc/pki/tls
• Edit nginx.conf file and add the new acstest.key and acstest.crt file to
the path in the file.
• Proxy_Path: use the private endpoint of COS at IBM cloud.
o https://s3.private.us-east.cloud-object-storage.appdomain.cloud;
• Save the file
22
The final nginx.conf looks like this. This file looks different than what is in the above
IBM link. We had to add additional fields to make it work for IBM i COS interface via
this reverse-proxy. Items shown in Red are the ones which may need to be updated.
This also now works for AIX, so we will use this nginx.conf file.
Restart nginx.
• nginx -t
output:
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful
23
https://cloud.ibm.com/docs/cloud-object-storage?topic=cloud-object-storage-uhc-hmac-
credentials-main
Enter your credentials from the COS HMAC. Make sure when u create a new credential
under Service Credential in COS GUI, you choose Advanced Option and check Include
HMAC Credential check box.
The new credential will now show aws credential too which you used above.
{
"apikey": "2x7rTtJuYFuivMKR3C7iP3Mnausq81t6A42GZoNt6FVb",
"cos_hmac_keys": {
"access_key_id": "e111daebd3…….",
"secret_access_key": "58f6de7f965ef528edc2e9…21036c8d623de"
},
"endpoints": "https://control.cloud-object-
storage.cloud.ibm.com/v2/endpoints",
"iam_apikey_description": "Auto-generated for key e111daeb-d379-42ff-
aa1e-2c8b6994c71……….
}
24
Accessing the VM/LPAR from your PC without Direct Link
• QLMTSECOFR to 0
25
• On Windows you can use PuTTY to run this:
• Open PuTTY and for the Host Name use the public IP
address of the VM/LPAR in the Cloud.
• Create a new name for your Session and Save.
• Select SSH>Tunnels
26
— Add the following new forwarded ports one at a time:
Source port Destination
2001 localhost:2001
2004 localhost:2004
2005 localhost:2005
2010 localhost:2010
449 localhost:449
50000 localhost:23
8470 localhost:8470
8471 localhost:8471
8472 localhost:8472
8473 localhost:8473
8474 localhost:8474
8475 localhost:8475
8476 localhost:8476
— Select Sessions and Save again
— Now you can Open your new PuTTY Session
— Login with your IBM i User ID and password
— Leave your PuTTY Session running
— If you close the PuTTY Session you will be
disconnected from the IBM i. Just minimize
the PuTTY session.
27
— You can now start your IBM i Client Solutions (ACS)
5250 session.
— Destination Address = localhost
— Destination Port = 50000
— Screen Size = 27x132
28
Chapter 2: Implementation
Use Cases
Examples
Legal notices
#
# (C) COPYRIGHT International Business Machines Corp. 2020
# All Rights Reserved
# Licensed Materials - Property of IBM
#
# US Government Users Restricted Rights - Use, duplication or
# disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
#
29
# THE SOURCE CODE EXAMPLES PROVIDED ARE ONLY INTENDED TO ASSIST IN THE
# DEVELOPMENT OF A WORKING SOFTWARE PROGRAM. IN ADDITION, THE SOURCE
# CODE EXAMPLES MAY NOT COMPILE AND/OR BIND SUCCESSFULLY AS WRITTEN.
##
# INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THE SOURCE CODE
# EXAMPLES, BOTH INDIVIDUALLY AND AS ONE OR MORE GROUPS, "AS IS" WITHOUT
# WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT
# LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
# PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE
# OF THE SOURCE CODE EXAMPLES, BOTH INDIVIDUALLY AND AS ONE OR MORE GROUPS,
# IS WITH YOU. SHOULD ANY PART OF THE SOURCE CODE EXAMPLES PROVE
# DEFECTIVE, YOU (AND NOT IBM OR AN AUTHORIZED DEALER) ASSUME THE ENTIRE
# COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
# No warranty is given that the contents of the source code examples, whether
# individually or as one or more groups, will meet your requirements or that
# the source code examples are error-free.
#!/bin/bash
################
IFS=":" read -ra ADDR <<< "${CLOUD_CRN}"
CLOUD_INSTANCE_ID=${ADDR[7]}
CLOUD_URL=(${ADDR[5]}.power-iaas.cloud.ibm.com)
30
## FIRST WE GET THE TOKEN FROM THE CLOUD IAM SERVICE USING THE API KEY
sleep 30
Pre-Conditions:
The body of the Snapshot, Restore and Clone (PVM and Volume) API
would have to be modified with user defined values.
Before running the Restore API, the PVM instance would have to be
SHUTOFF.
-H "authorization: <AuthToken>" \
31
-H "content-type: application/json" \
-H "crn: <CRN>" \
Expected Response:
{"snapshotID":"65ea39fd-cab6-46b3-b88c-3c28479ab019"}
-H "authorization: <AuthToken>" \
-H "content-type: application/json" \
-H "crn: <CRN>"
Expected Response:
"action": "snapshot",
"creationDate": "2020-04-13T08:51:21.000Z",
32
"description": "Snapshotfor VM1",
"lastUpdateDate": "2020-04-13T08:51:54.000Z",
"name": "VM1-SS",
"percentComplete": 100,
"pvmInstanceID": "7397dc00-f328-4bfb-bef2-27200ca42cb9",
"snapshotID": "65ea39fd-cab6-46b3-b88c-3c28479ab019",
"status": "available",
"volumeSnapshots": {
"398344bb-a64d-4fd5-b3cd-14ddfea6dd0e": "72f07383-ca5b-
46a0-94a2-3d1e7a7faceb",
"7a7a5b6e-1177-400a-82a4-0784957bbe75": "33f91096-f204-
4ed2-8110-c497a258c29c"
Restore to Snapshot:
-H "authorization: <AuthToken>" \
-H "content-type: application/json" \
-H "crn: <CRN>" \
-d "{\"forceRestore\":\"false\"}"
33
Expected Response:
"action": "restore",
"creationDate": "2020-04-13T08:51:21.000Z",
"lastUpdateDate": "2020-04-13T08:55:28.000Z",
"name": "VM1-SS",
"pvmInstanceID": "7397dc00-f328-4bfb-bef2-27200ca42cb9",
"snapshotID": "65ea39fd-cab6-46b3-b88c-3c28479ab019",
"status": "available",
"volumeSnapshots": {
"398344bb-a64d-4fd5-b3cd-14ddfea6dd0e": "72f07383-ca5b-
46a0-94a2-3d1e7a7faceb",
"7a7a5b6e-1177-400a-82a4-0784957bbe75": "33f91096-f204-
4ed2-8110-c497a258c29c"
34
Create Volume Clone:
curl -X POST \
-H 'content-type: application/json' \
-H 'crn: <CRN>' \
-d '{
"displayName": "PerfClone",
}'
Expected Response:
"clonedVolumes": {
"6342e6a9-716d-4686-b644-7f089bceb332": "fd99a7ae-3e15-
4f7e-af79-f5637e9a27f8",
"8461389f-e8fb-403f-8f48-81edcc9ef46f": "16ed7611-26cc-
4b93-945d-760cd6a52c58"
35
Create Cloud Resource – using a Direct Link Private
Endpoint (Recommended)
o CRTS3RICC
o Access key id: Copy this from your Cloud Object Storage
Resource
36
b. Click on the pull down associated with your Key
name
a. Click on Buckets
37
o Resource URI: Copy this from your Cloud Object Storage
Resource
a. Click on Buckets
38
NOTES:
Direct - Direct endpoints can accept requests from within the VPC and
charges are assessed on outgoing bandwidth.
2. Load and Apply PTF Direct Link (DL) Reverse Proxy Server
Support. Support Google Cloud Storage.
39
4. Display the Data Area QICC/QICCS3PRXY
o Verify the IP address
http://localhost:2001/QIBM/ICSS/Cert/Admin/qycucm1.ndm/main0
40
d. Click Create New Certificate Store.
e. Select *SYSTEM, and then click Continue.
In this Example, we had already created a Certificate Store for 5250 SSL
connections.
41
e. Save and rename the new text file with a “.crt”
extension.
Example: labser.crt
42
f. Select “Actions” then “Upload”.
i. Select OK.
43
4) Import the certificate into the *SYSTEM keystore:
a. In IBM Navigator for i, open the Digital Certificate
Manager.
b. In the navigation pane, click Select a Certificate Store.
c. Select *SYSTEM, and then click Continue.
44
Continue. The navigation pane displays tasks that you can
perform with the *SYSTEM keystore.
g. Enter the path and file name of the certificate file that you
copied to the IBM i computer in Step 2, and then click
Continue.
45
h. Enter a label for the certificate, and then click Continue.
The certificate is imported into the *SYSTEM keystore.
Click OK. Do not log out of the Digital Certificate Manager.
46
c. In the Add application page, in the Application ID field,
type IBM_QICC.
d. Select Application description and in the field type IBM
Cloud Storage Solutions for i.
47
6) Add the certificate authorities to the Cloud Storage Solutions
trust list:
a. In the Digital Certificate Manager navigation pane, click
Manage Applications> Define CA trust list.
b. Select Client, and then click Continue.
48
c. Select IBM Cloud Storage Solutions for i, and then click
Define CA Trust List.
Note: Skip this step if you are already using a Private Endpoint.
Jump to:
OR
Object-level Backups
This Example show how to use a Public - Public Endpoint if you don’t
have a Private Endpoint.
49
2) Create a cloud resource by running the following command:
2) CRTS3RICC
5) Access key id: Copy this from your Cloud Object Storage
Resource
50
a. Click on Service credentials
a. Click on Buckets
51
a. Click on Buckets
NOTES:
52
Direct - Direct endpoints can accept requests from within the VPC and
charges are assessed on outgoing bandwidth.
53
Full-System Backups from the Cloud
The following save data must be restored from physical media before BRMS can begin
restoring save data directly from the cloud:
• product IBM Backup, Recovery and Media Services for i and BRMS save
information is required before automatic recovery can be performed
BRMS provides specific control groups that can be used to automatically save this data to
media in the cloud and the cloud media can be used to create physical media. The control
groups will create cloud media that is formatted so it can be downloaded and burned
directly to physical optical media. All remaining data on the system can be backed up to
media in the cloud and restored directly from the cloud without a need to create physical
media.
Control group QCLDBIPLnn can be used to do full backups of all data that must be
recovered from physical media. Likewise, QCLDBGRPnn can be used to do cumulative
incremental saves of the data that must be recovered from physical media.
NOTE: The Journaled objects control group field must be changed to *YES for a
QCLDBGRPnn control group before the control group is used to do an incremental
backup. Run the WRKCTLGBRM command and change the Journaled objects field by
specifying option 8=Change attributes for QCLDBGRPnn.
Control group QCLDBSYSnn can be used to do a full backup of the data which can be
restored directly from the cloud. Likewise, control group QCLDBUSRnn can be used to
do cumulative incremental backups of the data which can be restored directly from the
cloud.
54
NOTE: The Journaled objects control group field must be changed to *YES for a
QCLDBUSRnn control group before the control group is used to do an incremental
backup. Run the WRKCTLGBRM command and change the Journaled objects field by
specifying option 8=Change attributes for QCLDBUSRnn.
It is critical to run the cloud control groups in the correct order otherwise all necessary
media information will not be available to do a recovery. The control groups that
produce media that will be burned to optical must be run after the control groups that
produce media that can be recovered directly from the cloud. For example:
This example will show you how to run the Full-System Backups from the Console.
55
2. Sign On using your Dedicated Service Tools (DST) User and password.
- OR -
1) Select Next
2) Press PF18
NOTES:
56
• The Console will timeout if inactive after 5 minutes. You may have to end your
Console browser and start a new Console connection.
• If you see a Break Message during the Backup, press Enter once to bring you
back to the screen where you entered the STRBKUBRM command so you can see
the progress of the backup.
4. Display QSYSOPR MSGQ, type command DSPMSG QSECOFR and look for
the following messages:
57
5. Change subsystems to process for Control Group QCLDBSYS01
o Find QCLDBSYS01
58
6. Run the First backup from the console
7. Check the backup for errors. Its normal to have some errors:
o Objects not saved. (Some objects are not required for the recovery).
o Media not transferred. (You will complete this step manually after the
Second backup).
8. Check subsystems after the backup completes. You should only have subsystem
QCTL in a status of RSTD. If not, end all subsystems again.
59
9. Change BRMS Control Group QCLDBIPL01
60
10. Run the Second backup from the console
61
11. Check the backup for errors. It’s normal to have some errors:
o Objects not saved. (Some objects are not required for the recovery).
o Media not transferred. (Will complete this manually after the Second
backup).
12. Now you will Identify the Volumes used the BOTH backups QCLDBSYS01 and
QCLDBIPL001 and transfer to IBM Cloud Object Storage (COS).
o If the Status in Failed, this is normal. We will transfer the volumes in the
next step.
62
15. Transfer the Volumes to IBM Cloud Object Storage (COS)
o Run command STRMNTBRM
o You will see the Volume Name, Status, and Complete% for each file
transfer. Wait until all Volumes have Successfully been completed to
proceed to the next step.
16. Verify that all the Volumes used for the Full-System Backup no longer have a
status of *TRF
o WRKMEDBRM TYPE(*TRF)
o You should not see any Volumes in the list
17. As with other recoveries that are performed using BRMS, a recovery report is
used to assist with successful recoveries from save media that has been transferred
to the cloud. To generate a report for recovery from the cloud, run the following
command:
NOTES:
63
cloud. If objects are on volumes that were not included in the recovery report,
they will be listed in a missing objects Attention section near top of the report.
• Once the recovery report has been verified, the report should be stored in a safe
location so it can be referred to during a recovery.
• On Monday through Saturday you can run daily incremental backups using the
following control groups:
64
Full-System Recovery from the Cloud using IBM i as
an NFS Server
Note: The IBM i IP address and the IBM i service tools server (LAN
console connection) IP address may be the same.
65
Creating an IBM i PowerVSI NFS Server Example:
• From the Cloud, copy the volumes created with the QCLDBIPL01 backup. In our
example we had three volumes.
Example: (Q06990, Q07898, Q08807)
• Resource name = the cloud resource that was created
• Submit to batch = *YES (you can submit all 3 at the same time)
• Cloud file name = Full name of the volume on the cloud
• Display the volumes on the COS
• Cloud Object Storage>Bucket>Object name>Details
• Local file name = Full directory path plus volume name
Note:
BRMS stores media in the cloud as files under directory
QBRMS_XXXXXXXX, where XXXXXXXX is the name of the system that
performed the backups.
66
Use the command CPYFRMCLD
• Go to the next step once all volumes have completed the transfer and show a
status of “Success”
67
Create Virtual Optical Device on IBM i VSI NFS Server
68
ADDIMGCLGE
o For the First volume Q06990
5. Add the next 2 using the same name for the TOFILE (To image file)
69
7. Verify Image Catalog
o WRKIMGCLG
o Select Option 10=Verify for image catalog “SYSIPL”
o Verify type = *LIC
o Sort image catalog = *YES
o Network file server share = *YES
70
9. Work with Object Links
o WRKLNK OBJ('/install/sysipl/*')
o Verify that a BOOTP DIR was created
o Verify that a VOLUME_LIST was created
71
12. Run command to change Object Authority
CHGAUT OBJ('/install/sysipl') USER(*PUBLIC) DTAAUT(*RWX)
SUBTREE(*ALL)
Specify the Alternate source directory where the volumes are stored
72
15. End TCP Server TFTP
ENDTCPSVR SERVER(*TFTP)
16. Start TCP Server TFTP
STRTCPSVR SERVER(*TFTP)
Configure IBM i PowerVSI Client Server (TARGET), NFS Network LAN install
device
4. Select Option 8=Work with service tools user IDs and Devices
73
5. Select F13=Select STS LAN adapter
o Press Next, then F13 on the bottom of the console screen
6. Option 1=Select
74
o Use the same resource used to communicate with the IBM i VSI NFS
Server
Note:
The CMNxx Resource needs to be on the same VLAN (public, direct, or DL) as
your IBM i VSI NSF Server. You also need to End TCP/IP and Vary Off the Line
Description using that CMNxx.
75
Note:
After you press F14 Activate, the addapter will restart, which may not be ready
right away.
76
10. On the Client Server Create Optical Device
o CRTDEVOPT
o Press F4 to prompt
o Local internet address = *SRVLAN
o Remote internet address = the ip address of the IBM i VSI NFS Server
o Network image directory = ‘/install/sysipl’
77
12. Verify you can access the remote image catalog
WRKIMGCLGE IMGCLG(*DEV) DEV(INSTALL)
78
Install Licensed Internal Code LIC on IBM i PowerVSI Client Server (TARGET)
using NFS Network Server
**WARNING**
Before you begin the scratch install on the (TARGET) IBM i VSI, be sure to document all
your current network information. You will have to recreate these after the install in
order to complete the restore:
• CFGTCP
- Work with TCP/IP interfaces
Internet Subnet Line Line
Address Mask Description Type
79
1. Start Network Install
o STRNETINS press F4 to prompt
o Network optical device = INSTALL
o Installation option = *LIC
o Keylock mode = MANUAL
80
4. Select Option 2=Install Licensed Internal Code and Initialize System
81
5. Press F10 to Continue
This took about 15 minutes plus 5 minutes for the IPL to complete.
82
6. Press F10 to Accept the problems and continue
o You will get this screen if you have more disk to add to the ASP
83
QSECOFR upper case for the password
84
9. On the Use Dedicated Service Tools (DST) Menu
o Select Option 4=Work with disk units
• This is when you can add non-configured disk to you system
85
11. Select Option 3=Work with ASP configuration
86
13. Select Option 3=Add units to existing ASPs
87
15. Press F10 to Ignore problems and continue
88
89
17. Verify disk units have been added
o Select Option 1=Work with disk configuration
90
19. Select Option 1=Display disk configuration status
or
20. Press F3=Exit three times and take Option 1. Exit Dedicated Service Tools (DST)
91
Install the Base Operating System IBM i Power VSI Client Server (TARGET) using
NFS Network Server
92
3. Configure the Network Device
o Server IP = the IP address of the SOURCE NFS IBM i VSI Server
o Path Name = the name of the Directory where the image volumes are
located
o Press F10 =Continue
93
6. Press Enter to confirm
94
8. Select Option 1=Restore programs and language objects from the current media
set (this will restore system information from the SOURCE system backup)
o Select Option 2=Keep for Job and output queues
o Select Option 1=Yes for distribute operating system and on available disk
units
95
10. User = QSECOFR
96
11. Change the following:
o Start print writers = N
o Start system to restricted state = Y
o Set major system options = Y
o Define or change system at IPL = Y
97
13. Select Option 3= System value commands
98
15. Select Option 2=Change for each of the System Values
o Change and document any of the changes made from the list below:
values to the new settings when you reach the "Define and Change the
Record the current setting for use after the recovery is complete.
QFRCCVNRST _______________ 0
QLMTDEVSSN _______________ 0
QLMTSECOFR _______________ 0
99
QMAXSIGN _______________ *NOMAX
QPFRADJ _______________ 2
QVFYOBJRST _______________ 1
16. Press F3=Exit, two times or until you get the next screen.
100
17. Press Enter to continue
101
19. At the IBM i Main Menu notice the System name changed
Notice you have the Base IBM i installed and Library QGPL and QUSRSYS are
*BACKLEVEL
102
Recover the BRMS Product and Associated Libraries on the IBM i Power VSI
Client Server (TARGET) using NFS Network Server
103
4. Verify the Catalog, Type, and Directory
o /install/sysipl
o Volumes should be Mounted and Loaded
Note:
You will find your BRMS Recovery Report from file name “QP1ARCY”
“Recovery Report”
VOL(volume-identifier) OPTFILE('')
104
Saved Save ----- ASP ------ Save Save Not Sequence
OPTFILE('/BRMSOPTSAV2248801200706103706I922BRMC/QBRM ')
OPTFILE('/BRMSOPTSAV2248801200706103706I922BRMC/QMSE ')
OPTFILE('/BRMSOPTSAV2248801200706103708I922BRMC/QUSRBRM ')
105
6. Follow Step 005 in the BRMS Recovery Report
- STEP 005 : Recover BRMS Related Media Information
You must recover this information for the BRMS product to accurately
VOL(volume-identifier) OPTFILE('')
MBROPT(*ALL) ALWOBJDIF(*COMPATIBLE)
Saved Save ----- ASP ------ Save Save Not Sequence
INZBRM OPTION(*SETAUT)
INZBRM OPTION(*DEVICE)
WRKDEVBRM
106
- Verify your “INSTALL” device is in the list
Note:
You will find your VOLUMES from the BRMS Report file name “QP1A2RCY”
- OR -
107
---------------------------------------------------------------------
108
__ *COMPATIBLE for the Allow object differences prompt.
10. Press F9= Restore Command Defaults and make the following changes
109
11. Page down once and make the following changes
o System resource management = *NONE (needs to be when you restore to
a different system)
110
Note:
If you get a message “Waiting for reply to message on message queue QSYSOPR. Select
SysReq and type 6 to display QSYSOPR system messages.
111
15. Follow Step 013 in the BRMS Recovery Report
- STEP 013 : Recover Required System Libraries
o Select Option 1=Select for all of the “Saved Item” and press Enter
112
Work with TCP/IP Interfaces and add Internet Address from TARGET IBM i
Before you began the scratch install on the (TARGET) IBM i VSI, you document all your
current network information. You will NOW have to recreate these in order to complete
the restore:
Use the IP Address information you documented from the TARGET IBM i VM created
in previous Step “Install Licensed Internal Code LIC on IBM i PowerVSI Client
Server (TARGET) using NFS Network Server”
113
Subnet mask = 255.255.255.x
Line description = use one of the 3 CLOUDINITx line description that was restored or
use the same one that was documented
3. Find the Resource URI used for the Cloud Object Storage (COS), this is where
you volumes are stored.
o Work with ICC Resources
o Command WRKCFGICC press Enter
o Select Option 5=Display
114
4. Document the Resource URI
5. See if you can ping the Resource URI to verify the connection
- If the ping was successful you have a working network to the Cloud Object
Storage.
- If the ping was unsuccessful, END the IP Interface and Vary off the Ethernet Line.
Change the Ethernet line description to another CMNxx Resource you haven’t
used yet and try to ping again.
115
7. Vary on the virtual tape device:
At this point the recovery, physical media is no longer required, media will be
downloaded from the cloud.
Since the system is in restricted state, TCP/IP must be started so BRMS can transfer the
media.
• STRTCPIFC INTNETADR('nnn.nnn.nnn.nnn')
116
through 255.
• STRSBS SBSD(QCTL)
• STRTCP
-----------------------------------------------------------------------------------------------------------
8. Move volumes from Cloud Object Storage (COS) to the TARGET IBM i VSI
Client
If you volume has a plus (+) to the right, it is part of a serial set.
Notice the Location of the volumes, they should still be in the Cloud Object Storage
(COS) Location.
9. Select Option 8=Move (all the volumes at the same time) press Enter
117
10. Change the Storage location to *HOME and press Enter
WRKSTSICC STATUS(*ALL)
118
12. Follow Step 014 in the BRMS Recovery Report
- STEP 014 : Reset BRMS Device and Media Library Information
- INZBRM OPTION(*DEVICE)
- WRKDEVBRM
- You should see Virtual Tape Device “TOR1CLDTAP *VRTTAP”
119
Recover IBM Product Libraries on the IBM i Power VSI Client Server (TARGET)
using Cloud Object Storage (COS)
The Restore Command Defaults should be used to specify the correct Device
parameter and change the Create parent directories prompt back to *NO.
-----------------------------------------------------------------------
2. Review the list of Recovery Items and Remove any that have already been
restored.
o Select Option 4=Remove and press Enter
▪ You can also see that the “Volume Serial” is from the optical
media
▪ Once you remove the items, they will drop off the list
120
3. Change the Recovery Defaults
o Select F9=Recovery defaults
o Device = TOR1CLDTAP
o Option = *ALL
o Data base member option = *ALL
o Restore spooled file data = *NEW
o Allow object differences = *COMPATIBLE
4. Page down
o System resource management = *NONE
o Create parent directories = *NO
121
o Press Enter to return to “Select Recovery Items”
122
Recover User Libraries on the IBM i Power VSI Client Server (TARGET) using
Cloud Object Storage (COS)
2. Review the list of Recovery Items and Remove any that have already been
restored.
o Select Option 4=Remove and press Enter
▪ You can also see that the “Volume Serial” is from the optical
media
▪ Press F11=Object View (will show you which Control Group
created the saved item)
▪ Remove items created by “QCLDBIPL01”
▪ Once you remove the items, they will drop off the list
123
3. Select the saved items
o Review the list again
o Select Option 1=Select for each item
or
o Select F16= Select (this will select all the items for you)
o Press Enter to recover the saved items
124
Recover Document Library Objects on the IBM i Power VSI Client Server
(TARGET) using Cloud Object Storage (COS)
125
Recover Directories and Files on the IBM i Power VSI Client Server (TARGET)
using Cloud Object Storage (COS)
2. Review the list of Recovery Items and Remove any that have already been
restored.
o Select Option 4=Remove and press Enter
▪ You can also see that the “Volume Serial” is from the optical
media
▪ Press F11=Object View (will show you which Control Group
created the saved item)
▪ Remove items created by “QCLDBIPL01”
▪ Once you remove the items, they will drop off the list
126
3. You can see the progress of the “processing objects” and “completed objects” as
they get restored
Complete Final Steps in BRMS on the IBM i Power VSI Client Server (TARGET)
UPDPTFINF
RSTAUT USRPRF(*ALL)
127
3. Follow Step 027 in the BRMS Recovery Report
- STEP 027 : Verify System Information
- This will restore your System Values
128
• Check network, software license keys, etc..
129
- STEP 031 : Perform IPL
Press F16=Confirm
130
11. Select Option 7=Start a service tool
131
13. IPL mode:
o Select Option 2=Normal
o Press F8 to set the IPL attributes and restart the system
132
14. Press Enter to confirm
133
Object-level Backups
NOTES:
1. The automatic transfer will begin after the first volume switch if
the save spans media
3. When using virtual optical media, the automatic transfer will only
start once the save job is completed.
Control groups that have a QCLDB prefix are created by BRMS for each
cloud resource defined on the system. The QCLDB control groups
should not be changed by the user but users can tailor control groups
that will automatically transfer media to the cloud, see section User
control groups for automatic transfers to the cloud -
QCLDUxxxxx below for more information.
Note: When using IBM Cloud Storage Solutions for i V1.2.0, a full
system backup requires that the data from QCLDBIPLnn and
QCLDBGRPnn control groups are backed up using a cloud connector
resource that does not use encryption and/or compression. To use
cloud connector resources with encryption and/or compression for a
full system save scenario, you must create two different sets of cloud
connectors to the same resource. One control group pair using
compression and/or encryption and another pair using neither
compression nor encryption to back up all the system data needed for
a Disaster Recovery.
134
The control groups QCLDBSYSnn and QCLDBUSRnn can be backed up
with a cloud connector resource using any configuration.
NOTES:
135
3. Do not copy the QCLDBIPLxx and QCLDBGRPxx control
groups and modify the copies to use encryption or compression
connectors. They will not work in a Disaster Recovery scenario.
o INZBRM OPTION(*DATA)
2. The pre-defined cloud control groups that BRMS created for the
cloud resource should be setup to work as defined.
136
3. To simplify this example, copy one of the pre-defined control
groups:
o WRKCTLGBRM TYPE(*BKU)
o WRKLIB LIB(QUSRTOOL)
137
• Specify New library = CLDLIB
o WRKCTLGBRM TYPE(*BKU)
▪ Seq - 10
o WRKCTLGBRM TYPE(*BKU)
138
• page down once
139
6. Run the control group:
o WRKMEDBRM FILEGRP(QCLDUUSR01)
o DLTLIB LIB(CLDLIB)
o WRKMEDIBRM CTLGRP(QCLDUUSR01)
140
10. Verify that the library was restored:
o DSPLIB LIB(CLDLIB)
11. Verify that the library was restored from the Cloud Object
Storage Location
o WRKSTSICC STATUS(*ALL)
• Status – Success
141
o WRKMEDBRM FILEGRP(QCLDUUSR01)
o CLRLIB LIB(CLDLIB)
o WRKLIB LIB(CLDLIB)
• Specify 12=Display
o WRKMEDIBRM CTLGRP(QCLDUUSR01)
142
15. Restore objects with *ERR (Obj Dtl =*NO)
143
16. Restore objects with *OBJ (Obj Dtl =*YES)
• press Enter
• press Enter
o DSPLIB LIB(CLDLIB)
144
NOTES:
145
• WRKMEDBRM option 8=Move
• MOVMEDBRM
• STRMNTBRM MOVMED(*YES)
This scenario will save a single empty library to virtual tape media,
manually transfer the media to the cloud and then automatically
transfer media back to the system during a restore.
o CRTS3RICC
146
9. Configure BRMS objects required to use the cloud resource:
o INZBRM OPTION(*DATA)
147
13. Create an image catalog:
o WRKIMGCLG
o ADDIMGCLGE IMGCLG(TOR1CLDTAP)
FROMFILE(*NEW) TOFILE(*GEN) VOLNAM(TORCLD)
148
15. Load the image catalog on the device:
o LODIMGCLG IMGCLG(TOR1CLDTAP)
DEV(TOR1CLDTAP)
o WRKIMGCLG
149
o WRKDEVBRM
o WRKCLSBRM TYPE(*MED)
150
▪ specify TOR1CLDTAP for the Class field
151
o WRKMEDBRM VOL(TORCLD)
o WRKPCYBRM TYPE(*MED)
152
20. Create a library to save to the cloud:
o CRTLIB LIB(CLDLIB)
o WRKMEDBRM VOL(TORCLD)
153
▪ specify TOR1CLD for the Storage location field
154
23. Delete the library:
o DLTLIB LIB(CLDLIB)
o DSPLIB LIB(CLDLIB)
155
NOTES
• When control groups are run that have a name with a QCLD
prefix will, BRMS will automatically create and transfer media to
the cloud. For more information regarding automatic cloud
transfers see Automatic Transfers of Media to Cloud Storage.
• If using user created images, BRMS will not manage these image
catalog entries. As such, when the data is transferred to the
cloud, the images will not be cleared or deleted. Users will need
to manage these themselves. it is recommended
using CHGIMGCLGE with the parameter IMGCLG(*MIN) to
shrink the size down as much as possible.
156
Chapter 3: Troubleshooting
WRKMEDBRM TYPE(*TRF)
Most cloud transfers will be started by BRMS but run in jobs in subsystem
QICCSBS. To determine which jobs have been used to process cloud transfers for
BRMS run the following command:
WRKSTSICC STATUS(*ALL)
157
Use the job information from the failed transfer to work with the job. The job log will
contain information about the transfer:
WRKJOB JOB(015970/QSECOFR/L000000010)
option 5=Display
WRKLNK /QIBM/ProdData/QICC
158
You can verify the IP address your using for the Direct Link (DL)
Private connection and the COS Bucket.
159
o Local file name = /tmp/plume.log (verify it is in your directory
path)
o Cloud file name = test1cs (this can be any name, don’t forget
to cleanup file in COS if successful after your test)
Restore problems
CPF41B0 – Incorrect image catalog name specified.
Check the current location of the volume BRMS is trying to restore from in
WRKMEDBRM. If the volume does not have a status of *ACT and the location is not
the cloud location, run STRMNTBRM MOVMED(*YES) on the source and remote
system and verify the volume has a status of *ACT and the location is the cloud. Retry
the restore.
160
Chapter 4: Additional Resources
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd
=2&ved=2ahUKEwjypYrCkMXpAhUCH80KHV69D5wQFjABegQIAxAB&ur
l=https%3A%2F%2Fwww.ibm.com%2Fsupport%2Fknowledgecenter%
2Fssw_ibm_i_73%2Ficc%2Ficcumstpdf.pdf&usg=AOvVaw01p3eqw7Tu
0eqXHGmqWLsX
https://helpsystemswiki.atlassian.net/wiki/spaces/IWT/pages/1656422
70/BRMS+and+Cloud+Storage+Solutions
161