cancel
Showing results for 
Search instead for 
Did you mean: 

VMware restore trying to use PBX and vnetd, when there is no Netbackup client on guests

PaulStevens
Level 4
Partner

Netbackup 9.1.0.1

Environment is predominantly VMware.  Windows team just built a new guest to replace an old one and asked me to restore some files from oldserver onto newserver.  Not a full machine recovery, just restore a couple directories.

I backup up clients with VMware policies, so there is no Netbackup client on the individual guests.  I am successfully backing up old server and new server.

When I queued up the restore, I selected old server as the source, new server as the destination, and policy type VMware. Then I selected last night's backup from the date dropdown and selected the directories they asked for.

But when I kick off the restore I am getting status 58 - Cannot connect to client.

Why is it trying to connect via pbx and vnetd when the policy type is VMware and there is no netbackup client on the guests?

Mar 1, 2023 5:04:04 PM - Info bsoalbnbu001 (pid=205249) StorageServer=PureDisk:bsoalbnbu001; Report=PDDO Stats for (bsoalbnbu001): read: 172156 KB, CR received: 48233 KB, CR received over FC: 0 KB, dedup: 0.0%
Mar 1, 2023 5:04:04 PM - requesting resource  @aaaac
Mar 1, 2023 5:04:04 PM - granted resource  MediaID=@aaaac;DiskVolume=PureDiskVolume;DiskPool=dp_disk_bsoalbnbu001;Path=PureDiskVolume;StorageServer=bsoalbnbu001;MediaServer=bsoalbnbu001
Mar 1, 2023 5:04:20 PM - Info bpbrm (pid=205533) connect failed STATUS (18) CONNECT_FAILED
Mar 1, 2023 5:04:20 PM - Info bpbrm (pid=205533)  status: FAILED, (42) CONNECT_REFUSED; system: (111) Connection refused; FROM 0.0.0.0 TO bsobccmalb001 10.30.67.34 bpcd VIA pbx
Mar 1, 2023 5:04:20 PM - Info bpbrm (pid=205533)  status: FAILED, (42) CONNECT_REFUSED; system: (111) Connection refused; FROM 0.0.0.0 TO bsobccmalb001 10.30.67.34 bpcd VIA vnetd
Mar 1, 2023 5:04:20 PM - Error bpbrm (pid=205533) cannot connect to bsobccmalb001, Transport endpoint is not connected (107)
Mar 1, 2023 5:04:20 PM - Info tar (pid=0) done. status: 58: can't connect to client
Mar 1, 2023 5:04:20 PM - Error bpbrm (pid=205533) client restore EXIT STATUS 58: can't connect to client

 

5 REPLIES 5

davidmoline
Level 6
Employee

Hi @PaulStevens 

The simple answer is the way you have initiated the restore it is attempting to connect to a NetBackup agent on the target machine. NetBackup uses PBX for all network communication for current version of NetBackup (and vnetd is the fallback when PBX doesn't work). 

What you need to do is one of the following:

  1. Restore to a staging host and move the files across to the target server.
  2. Use instant access (if available) to allow the VM to access the source VMDK and retrieve the files from there.
  3. Install a NetBackup agent into the target VM
  4. Use VMware agentless restore (available using the WebUI).

Regards
David

trying to login to webui but   https://primaryserver/webui/login gets a webpage cannot be found.  https://primaryserver does get to the appliance gui, but get a lot of errors trying to retrieve the information.   the hardware section says hardware is healthy, but error retrieving alert info, storage info, etc.    

Hi @PaulStevens 

This is a problem you should resolve - as webservices are required for other things to happen correctly. 

You could check the logs (/usr/openv/wmc/webserver/logs) to see if you can identify why the web service is unavailable. 

A reboot of the appliance may also help (or possibly a simple restart of NetBackup services). 

Regards
David

I think there is some crazy stuff happening because we moved the server into a new network domain.

Appliance was initially built as bsoalbnbu001.csc-rsr-m001.com, then added secondary interface to another VLAN named .bsoc.local.  When all the clients were moved to .bsoc.local we disabled the initial network interface.  I removed the old search domains and added the new one as well as the routing and nameservers.  We are still backing up the clients, but the hostname of the server is still the original and I have been told we can't change that without reimaging.  the nameservers in the new domain have us listed as bsoalbnbu001.bsoc.local  So we are getting errors that the name doesn't match what is in the certificate.

Long story short, can we reimage without losing our backups in the MSDP?  There is no tape involved, everything is on disk on the appliance.

Hi @PaulStevens 

It is possible to reimage and appliance without resetting the backup storage. I will also say that it is risky to do this and should only be done in the last resort. 

Either way if you want to go down this route (or even explore alternatives to a reimage) I would open a support case and request assistance.

And yes - the name change is probably most of the cause of the issues - you may be able to have some luck if you can create a CNAME alias for the old name (bsoalbnbu001.csc-rsr-m001.com) that points to the new name (worth attempting as a start).

Cheers
David