cancel
Showing results for 
Search instead for 
Did you mean: 

Restoring data from an exchange backup without contact to the original server

contra04
Level 5

Hi catostropic monstrous events going on.

 

I need to restore some exchange databases, which are disk images on local drives on my master server.  Dont need to go to tape.

 

The server has been rebuilt, and has no client installed on it, and I do not have an option to install it at present.

what does one do when trieng to restore data for clients that no longer exist?

Is an exchange netbackup client a different isntall than the regular client ?

 

How can I point to an image and just restore it to a location of my choice? the backup and restore client on the master server asks for a rsetore client and type....

 

from the command line...

bpimagelist -d 01/27/2012 -policy mbx1-store -U

No entity was found.

 

Any help would be extremely appreciated !

35 REPLIES 35

contra04
Level 5

 

 The clear host cache command doesnt seem available on windows? 

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -clear_host_cache

 bpclntcmd: unrecognized option -clear_host_cache

bpclntcmd: -sv

bpclntcmd: -pn

bpclntcmd: -self

bpclntcmd: -hn <hostname>

bpclntcmd: -server <NBU master>

bpclntcmd: -ip <ipaddress>

bpclntcmd: -gethostname

bpclntcmd: -is_local_host <hostname>

bpclntcmd: -check_vxss

bpclntcmd: -check_vxss_with_host <hostname>

bpclntcmd: -get_pbx_port [<hostname>]

bpclntcmd: -get_remote_host_version <hostname>

bpclntcmd: -get_local_client_patch_version

bpclntcmd: -get_local_server_patch_version

bpclntcmd: -sanclient <0|1>

bpclntcmd: -reverse_name_lookup [allowed|restricted|prohibited]   

The test seems correct though: All ip addresses etc, names

 

 C:\Program Files\VERITAS\NetBackup\bin>bptestbpcd -client ixbackup-lon -verbose

1 1 1

10.1.104.78:1564 -> 10.1.2.186:13724

10.1.104.78:1565 -> 10.1.2.186:13724

PEER_NAME = harry

HOST_NAME = ixbackup-lon

CLIENT_NAME = ixbackup-lon

VERSION = 0x07000000

PLATFORM = win_x64

PATCH_VERSION = 7.0.0.0

SERVER_PATCH_VERSION = 7.0.0.0

MASTER_SERVER = harry

EMM_SERVER = harry

10.1.104.78:1570 -> 10.1.2.186:13724 

 

 BPCD on the client: 

 09:04:07.393 [3076.1912] <2> setup_debug_log: switched debug log file for bpcd

09:04:07.393 [3076.1912] <2> bpcd main: VERBOSE = 5

09:04:07.393 [3076.1912] <2> logparams: bpcd 

09:04:07.393 [3076.1912] <2> bpcd main: Got socket for input 480

09:04:07.393 [3076.1912] <2> ReadKeyfile: keyfile C:\Program Files\Veritas\NetBackup\bin\keyfile.dat does not exist

09:04:07.393 [3076.1912] <2> process_requests: offset to GMT 0

09:04:07.393 [3076.1912] <2> logconnections: BPCD ACCEPT FROM 10.1.104.78.1564 TO 10.1.2.186.13724

09:04:07.393 [3076.1912] <2> process_requests: setup_sockopts complete

09:04:07.393 [3076.1912] <2> bpcd peer_hostname: Connection from host harry (10.1.104.78) port 1564

09:04:07.393 [3076.1912] <2> bpcd valid_server: comparing harry and harry

09:04:07.393 [3076.1912] <4> bpcd valid_server: hostname comparison succeeded

09:04:07.580 [3076.1912] <2> process_requests: output socket port number = 1

09:04:08.143 [3076.1912] <2> vnet_receive_network_socket: ../../libvlibs/vnet_vnetd.c.1257: hash_str1: 80f90887ea0e4b498d1a803137771862

09:04:08.143 [3076.1912] <2> vnet_receive_network_socket: ../../libvlibs/vnet_vnetd.c.1258: hash_str2: 958b9de22526c53d4c1c859e5c6e6a3f

09:04:08.143 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1771: hash_str1: 80f90887ea0e4b498d1a803137771862

09:04:08.143 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1773: hash_str2: 958b9de22526c53d4c1c859e5c6e6a3f

09:04:08.143 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1799: hash_str: 80f90887ea0e4b498d1a803137771862

09:04:08.143 [3076.1912] <2> process_requests: Duplicated vnetd socket on stderr

09:04:08.143 [3076.1912] <2> process_requests: <---- NetBackup 7.0 0 ------------initiated

09:04:08.143 [3076.1912] <2> process_requests: VERBOSE = 5

09:04:08.143 [3076.1912] <2> process_requests: Not using VxSS authentication with harry

Interface: 10.1.2.186 --- 0xb

  Internet Address      Physical Address      Type

  10.1.1.1              ec-30-91-e3-ca-80     dynamic   

  10.1.1.3              00-00-00-00-00-18     dynamic   

  10.1.1.53             00-0f-20-31-e6-6b     dynamic   

  10.1.1.87             00-50-56-af-00-40     dynamic   

  10.1.1.88             00-1e-0b-e9-be-be     dynamic   

  10.1.2.71             00-50-56-af-00-6a     dynamic   

  10.1.2.223            00-50-56-a5-00-0b     dynamic   

  10.1.31.255           ff-ff-ff-ff-ff-ff     static    

  224.0.0.22            01-00-5e-00-00-16     static    

  224.0.0.252           01-00-5e-00-00-fc     static    

09:04:08.159 [3076.1912] <2> process_requests: 

09:04:08.346 [3076.1912] <2> process_requests: BPCD_PEERNAME_RQST

09:04:08.346 [3076.1912] <2> bpcd peer_hostname: Connection from host harry (10.1.104.78) port 1564

09:04:08.674 [3076.1912] <2> process_requests: BPCD_HOSTNAME_RQST

09:04:09.002 [3076.1912] <2> process_requests: BPCD_CLIENTNAME_RQST

09:04:09.221 [3076.1912] <2> process_requests: BPCD_GET_VERSION_RQST

09:04:09.549 [3076.1912] <2> process_requests: BPCD_GET_PLATFORM_RQST

09:04:09.768 [3076.1912] <2> process_requests: BPCD_GET_VERSION_RQST

09:04:09.987 [3076.1912] <2> process_requests: BPCD_PATCH_VERSION_RQST

09:04:10.205 [3076.1912] <2> retrieveLocalPatchVersion: Reading from C:\Program Files\Veritas\NetBackup\bin\version.txt

09:04:10.205 [3076.1912] <2> parsePatchVersionString: parsing = >7.0

<

09:04:10.205 [3076.1912] <2> parsePatchVersionString: theRest = ><

09:04:10.315 [3076.1912] <2> process_requests: BPCD_GET_VERSION_RQST

09:04:10.534 [3076.1912] <2> process_requests: BPCD_PATCH_VERSION_RQST

09:04:10.534 [3076.1912] <2> retrieveLocalPatchVersion: Reading from C:\Program Files\Veritas\NetBackup\version.txt

09:04:10.534 [3076.1912] <2> parsePatchVersionString: parsing = >7.0

<

09:04:10.534 [3076.1912] <2> parsePatchVersionString: theRest = ><

09:04:10.643 [3076.1912] <2> process_requests: BPCD_GET_VERSION_RQST

09:04:10.862 [3076.1912] <2> process_requests: BPCD_READ_HOST_CONFIG_RQST

09:04:11.080 [3076.1912] <2> process_requests: BPCD_GET_STDOUT_SOCKET_RQST

09:04:11.299 [3076.1912] <2> process_requests: socket port number = 1

09:04:11.846 [3076.1912] <2> vnet_receive_network_socket: ../../libvlibs/vnet_vnetd.c.1257: hash_str1: c598d247a7b3600134d56cac95131ee2

09:04:11.846 [3076.1912] <2> vnet_receive_network_socket: ../../libvlibs/vnet_vnetd.c.1258: hash_str2: 01a7887a4a2f40be4b1d452f0a25d431

09:04:11.846 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1771: hash_str1: c598d247a7b3600134d56cac95131ee2

09:04:11.846 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1773: hash_str2: 01a7887a4a2f40be4b1d452f0a25d431

09:04:11.846 [3076.1912] <2> verify_hashes: ../../libvlibs/vnet_vnetd.c.1799: hash_str: c598d247a7b3600134d56cac95131ee2

09:04:11.846 [3076.1912] <2> process_requests: Connected on output socket

09:04:11.846 [3076.1912] <2> process_requests: Skipping shutdown of send side of stdout.

09:04:11.846 [3076.1912] <2> process_requests: Duplicated socket on stdout

09:04:11.846 [3076.1912] <2> process_requests: BPCD_DISCONNECT_RQST

09:04:11.846 [3076.1912] <2> bpcd exit_bpcd: exit status 0  ----------->exiting 
I feel we are getting closer!

contra04
Level 5

argh why is potter trying to connect ! - it shouldnt be involved?! This happens when I initiate the backup, like before

 

 

09:13:41.314 [2272.2904] <2> process_requests: offset to GMT 0
09:13:41.314 [2272.2904] <2> logconnections: BPCD ACCEPT FROM 10.2.20.20.1650 TO 10.1.2.186.13724
09:13:41.314 [2272.2904] <2> process_requests: setup_sockopts complete
09:13:41.314 [2272.2904] <2> bpcd peer_hostname: Connection from host potter.internal.robertwalters.com (10.2.20.20) port 1650
09:13:41.314 [2272.2904] <2> bpcd valid_server: comparing harry and potter.internal.robertwalters.com
09:13:41.314 [2272.2904] <4> bpcd valid_server: potter.internal.com is not a master server
09:13:41.314 [2272.2904] <16> bpcd valid_server: potter.internal.com is not a media server either
 
I have added potter as another master server in the BAR on the client to see if it makes a difference

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

clear_host_cache was introduced with a 7.0.x patch. Included in 7.1.

Your client is resolving incoming IP address from master: 10.2.20.20 to potter.internal.robertwalters.com.

On client do:

bpclntcmd -ip 10.2.20.2

contra04
Level 5

OK that solved that problem - problem has changed which is good! now we cannot access the img on the staging area: - although it looks fine in the screenshot?

 

 

03/02/2012 09:27:39 - begin Restore

03/02/2012 09:27:41 - restoring image mbx1-lon_1327731028

03/02/2012 09:28:20 - restored image mbx1-lon_1327731028 - (media open error(83)); restore time 00:00:39

03/02/2012 09:28:25 - end Restore; elapsed time: 00:00:46

03/02/2012 09:33:02 - connecting

03/02/2012 09:33:07 - connected; connect time: 00:00:05

03/02/2012 09:33:08 - started process bptm (5308)

03/02/2012 09:33:10 - started process bptm (1160)

03/02/2012 09:33:18 - Critical bptm(pid=5308) sts_get_lsu_prop_byname on LSU E:\Staging failed: 2060013 no more entries    

03/02/2012 09:33:18 - Critical bptm(pid=5308) Invalid storage device: E:\Staging no more entries      

03/02/2012 09:33:19 - Critical bptm(pid=5308) failure to open disk at path E:\Staging: plug-in reports error 2060013  

03/02/2012 09:33:21 - Error bptm(pid=1160) The following files/folders were not restored:       

03/02/2012 09:33:22 - Error bptm(pid=1160) UTF - /Microsoft Information Store/Public/        

03/02/2012 09:33:22 - Error bptm(pid=1160) UTF - /Microsoft Information Store/Public/Public        

03/02/2012 09:33:23 - Error bptm(pid=1160) UTF - /Microsoft Information Store/Public/Log files_1327731028       

03/02/2012 09:33:29 - Error bpbrm(pid=4360) client restore EXIT STATUS 83: media open error     

the restore failed to recover the requested files(5)
 
but it seems available?
 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Does the STU exist in the Storage Unit config?

contra04
Level 5

All seems fine - What does stu stand for? sorry Im not very good at this. Thank you very much for your time

Mark_Solutions
Level 6
Partner Accredited Certified

Contra

STU stands for STorage Unit

It looks like the Masters configuration has changed within itself somewhere if it can no longer read its own images - this may require a lot more research to put it all right.

In the mean time - OnTrack can actually read NetBackup Images

If you copy all of the fragments relating to this backup to an area of disk on your OnTrack server it should be able to just open them up and read all of the Exchange Information within them - check your OnTrack docs - I have a cutomer that does this all the time so I know it shoudl work (assuming you have the same type of OnTrack)

Hope this helps

contra04
Level 5

I have the latest ontrack and a full liscence - I have already copied the image files to the ontrack server, but as far as I can see in the documentation the way it improts them is by emulating an exchange server, then you send the restore to the ontrack server which is what I am doing.

 

I thought the img files created by netbackup were proprietary and cannot be decompressed by third parties?  Ontrack can certainly import edb files, there is no option to import img files.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Seems the master/media server has more than one NIC with more than one hostname.

Please run the following from cmd:

nbemmcmd -listhosts -verbose

You will probably need to do a 'nbemmcmd -machinealias -addalias .....'

>nbemmcmd -machinealias -help
NBEMMCMD, Version:7.1
Help requested.
Usage:
nbemmcmd -machinealias [-brief] [-emmname <string>] [-emmport <unsigned integer>]
    [-addalias -alias <string> -machinename <string>]

contra04
Level 5

 

Thank you for your suggestions.
 
 
 
-Mark_Solutions - I found that option in ontrack and it complains that no "valid informations stores were found" when pointing it to the IMG file on the master server.
 
-Marianne  I will try to disable the second nic (isolated network)/ restart the services and try again.
 
C:\Program Files\VERITAS\NetBackup\bin>nbemmcmd -listhosts -verbose
 
NBEMMCMD, Version:7.0
The following hosts were found:
harry
        MachineName = "harry"
        MachineDescription = ""
        MachineNbuType = server (6)
harry
        ClusterName = ""
        MachineName = "harry"
        GlobalDriveSeed = "VEND:#.:PROD:#.:IDX"
        LocalDriveSeed = ""
        MachineDescription = ""
        MachineFlags = 0xa4
        MachineNbuType = master (3)
        MachineState = active for tape and disk jobs (14)
        NetBackupVersion = 7.0.0.0 (700000)
        OperatingSystem = windows (11)
        ScanAbility = 5
potter
        ClusterName = ""
        MachineName = "potter"
        LocalDriveSeed = ""
        MachineDescription = ""
        MachineFlags = 0xa4
        MachineNbuType = media (1)
        MachineState = active for tape and disk jobs (14)
        MasterServerName = "harry"
        NetBackupVersion = 7.0.0.0 (700000)
        OperatingSystem = windows (11)
        ScanAbility = 5
Command completed successfully.

contra04
Level 5

Ok with just one nic, on the master servers original IP address we still get issues connecting to the Staging Unit.  However I am sure it works fine for duplication to tape ??!

 

 03/02/2012 10:44:23 - begin Restore

03/02/2012 10:44:25 - restoring image mbx1-lon_1327731028

03/02/2012 10:45:16 - restored image mbx1-lon_1327731028 - (media open error(83)); restore time 00:00:51

03/02/2012 10:45:21 - end Restore; elapsed time: 00:00:58

03/02/2012 10:49:47 - connecting

03/02/2012 10:49:52 - connected; connect time: 00:00:05

03/02/2012 10:49:53 - started process bptm (5184)

03/02/2012 10:50:07 - started process bptm (5104)

03/02/2012 10:50:15 - Critical bptm(pid=5184) sts_get_lsu_prop_byname on LSU E:\Staging failed: 2060013 no more entries    

03/02/2012 10:50:15 - Critical bptm(pid=5184) Invalid storage device: E:\Staging no more entries      

03/02/2012 10:50:16 - Critical bptm(pid=5184) failure to open disk at path E:\Staging: plug-in reports error 2060013  

03/02/2012 10:50:18 - Error bptm(pid=5104) The following files/folders were not restored:       

03/02/2012 10:50:18 - Error bptm(pid=5104) UTF - /Microsoft Information Store/Public/        

03/02/2012 10:50:19 - Error bptm(pid=5104) UTF - /Microsoft Information Store/Public/Public        

03/02/2012 10:50:19 - Error bptm(pid=5104) UTF - /Microsoft Information Store/Public/Log files_1327731028       

03/02/2012 10:50:25 - Error bpbrm(pid=5420) client restore EXIT STATUS 83: media open error     

the restore failed to recover the requested files(5)
 
All the articles I have read regarding this point to issues with staging units that have been removed. However my one is there, a local disk in netbackup and looks fine.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

You need an 'extra pair of eyes' to try and find out what is wrong here.

Best to log a support call and ask for a WebEx session.

nbemmcmd output lists master as 'harry' and media server as 'potter'.

Earlier today the master server seemed to connect to the client as 'potter'?

Are they 2 different machines or 2 names linked to 2 NICs on the master?

What happens if you do 'bpclntcmd -self' on the master?

Do you have bprd log on the master? Please post as attachment.

Another option would be to promote tape copy to primary.

contra04
Level 5

Hi Marianne, I have a support case open and they are stumped.  ive been through 3 engineers so far.  All of them webexed in.  Generally I find the support from you, mark_solutions and andy welburn is far superior to the engineers that have been assigned to my last 3 cases.  In every instance the forums have solved the issue, not the support engineers.

 

They are two seperate machines, in two different buildings.

 

 

C:\Documents and Settings\adm-cuttinn>bpclntcmd -self
gethostname() returned: harry
host harry: harry.internal.com at 10.1.104.78 (0x4e68010a)
aliases:
 
 
I dont have a bprd log on the server, but I have now created the folder.
 
The backup is losing value now that it is a week old, so I think we are going to use a san snapshot from a few days earlier than the saturday to go ahead with. That and Im off to New Zealand for 3 weeks !
 
If you or anyone thinks of anything let me know, And Ill check it out in three weeks. Thanks very much all.
 
Nick

Mark_Solutions
Level 6
Partner Accredited Certified

Contra04

In view of the length of this thread please give us a very quick summary of everything that has happened so that we can review where we are ..

1. Master server name - ever changed? which IP changes have happened and has DNS been fully updated

2. Where was the disk backup originally done (master or media)

3. Where was the tape copy originally done (master or media)

4. Has the disk storage unit (Master at E:\Staging) always been there

5. When you ping from Master to Client in a command line (and vice versa does it return the correct IP address and do they report the short name of FQDN

Anything else that has happened of relevance

Thanks

contra04
Level 5

1) no hostname never changed.

It went from a teamed nic on 10.1.104.88, to a single NIC on 10.1.200.20.

That was when we attemped the restore to the exchange server, that we moved onto the same vlan.

Then we added NIC2 back onto its original address. I removed all traces of the 10.1.200.20 address from all hosts files, and dns servers. I also disabled the 10.1.200.20 NIC from being able to update DNS.

Then I began to attempt to restore it to the "Emulated" exchange server on a spare server running Ontrack power Controls. Same errors about not being able to access the E:\staging

Then I disabled the second NIC, and looked for any tracs of it in dns / hosts bp commands etc on the client and the host.  From this point I only care about the spare server running ontrack getting the data.

It was then the same network setup as it has been for the last few years, except running on one NIC, not a team

(2) It is sitting on a staging disk, local to the server.  It has not been copied to tape yet.

(3) no tapes involved yet

(4) Yes no changes there for a few years

(5) yes reverse and forward looks are fine.

I have now managed to mount the IMG file DIRECTLY onto Ontract Import.  They key is it MUST be on a local disk - so rather than using SMB, I copyied it over the LAN and it is mounting as we speak :)

 

Thank you for your help

Mark_Solutions
Level 6
Partner Accredited Certified

Ok - I think I did say to copy it over to the OnTrack server earlier - lost track now

See how it goes and update us