Forum Discussion

M-NYC's avatar
M-NYC
Level 3
11 years ago

UNC path backups are failing

Win 2008 r2 x64

NBU 7.1

Accessing the UNC path from the NBU Master/Media server " netbackup-01ny"

Large Shares:

\\file-03ny\redirect-adm$\Dead-My-Docs

\\file-03ny\redirect-adm$\RedirectFolders

Attached is bpbrm log

 

I am trying to backup, over the network via UNC path a couple of large file systems. We're eventually moving away from this method of backup but I need it for a little while.

 

I did the following to get basic UNC path backups working (changing the account that the services run as):

http://www.symantec.com/connect/blogs/netbackup-error-when-using-unc-path-storage-unit-absolute-pathname

 

At this point I can backup small shares or single files via UNC path but I get the following error when I am doing larger files.

 

2/25/2014 10:56:22 AM - Info nbjm(pid=5940) starting backup job (jobid=974) for client netbackup-01ny, policy VNX-5500-Redirect, schedule Monthly-Full  
2/25/2014 10:56:22 AM - Info nbjm(pid=5940) requesting STANDARD_RESOURCE resources from RB for backup job (jobid=974, request id:{6DF81998-AFE8-4E6C-9058-49A84F1420EE})  
2/25/2014 10:56:22 AM - requesting resource Netbackup-01NY--Scalar50--SAS--LTO4--Libraries
2/25/2014 10:56:22 AM - requesting resource netbackup-01ny.hq.hview.com.NBU_CLIENT.MAXJOBS.netbackup-01ny
2/25/2014 10:56:22 AM - requesting resource netbackup-01ny.hq.hview.com.NBU_POLICY.MAXJOBS.VNX-5500-Redirect
2/25/2014 10:56:22 AM - granted resource netbackup-01ny.hq.hview.com.NBU_CLIENT.MAXJOBS.netbackup-01ny
2/25/2014 10:56:22 AM - granted resource netbackup-01ny.hq.hview.com.NBU_POLICY.MAXJOBS.VNX-5500-Redirect
2/25/2014 10:56:22 AM - granted resource ABC073
2/25/2014 10:56:22 AM - granted resource IBM.ULTRIUM-HH6.000
2/25/2014 10:56:22 AM - granted resource netbackup-01ny-hcart2-robot-tld-0
2/25/2014 10:56:23 AM - Info bpbrm(pid=4444) telling media manager to start backup on client     
2/25/2014 10:56:23 AM - estimated 0 Kbytes needed
2/25/2014 10:56:23 AM - Info nbjm(pid=5940) started backup job for client netbackup-01ny, policy VNX-5500-Redirect, schedule Monthly-Full on storage unit netbackup-01ny-hcart2-robot-tld-0
2/25/2014 10:56:23 AM - mounting ABC073
2/25/2014 10:56:24 AM - Info bptm(pid=5964) using 262144 data buffer size        
2/25/2014 10:56:24 AM - Info bptm(pid=5964) using 64 data buffers         
2/25/2014 10:56:24 AM - Info bpbrm(pid=5528) sending bpsched msg: CONNECTING TO CLIENT FOR netbackup-01ny_1393343783     
2/25/2014 10:56:24 AM - Info bpbrm(pid=5528) start bpbkar32 on client         
2/25/2014 10:56:24 AM - Info bpbkar32(pid=3096) Backup started           
2/25/2014 10:56:24 AM - Info bpbrm(pid=5528) Sending the file list to the client      
2/25/2014 10:56:24 AM - connecting
2/25/2014 10:56:24 AM - connected; connect time: 00:00:00
2/25/2014 10:56:59 AM - mounted; mount time: 00:00:36
2/25/2014 10:56:59 AM - positioning ABC073 to file 42
2/25/2014 10:57:35 AM - positioned ABC073; position time: 00:00:36
2/25/2014 10:57:35 AM - begin writing
2/25/2014 10:57:37 AM - Error bpbrm(pid=5528) socket read failed, An existing connection was forcibly closed by the remote host.  (10054)
2/25/2014 10:57:37 AM - Error bpbrm(pid=5528) could not send server status message       
2/25/2014 10:57:37 AM - Info bpbrm(pid=4444) child done, status 13         
2/25/2014 10:57:37 AM - Info bpbrm(pid=4444) sending media manager msg: STOP BACKUP netbackup-01ny_1393343782      
2/25/2014 10:57:42 AM - Error bpbrm(pid=4444) cannot send mail to mike.eid@erlworldwide.com,backup.alerts@erlworldwide.com,shukri.mahmud@erlworldwide.com        
2/25/2014 10:57:42 AM - Info bpbrm(pid=4444) child done, status 13         
2/25/2014 10:57:42 AM - Info bpbrm(pid=4444) sending media manager msg: STOP BACKUP netbackup-01ny_1393343783      
2/25/2014 10:57:56 AM - Info bpbrm(pid=4444) media manager for backup id netbackup-01ny_1393343783 exited with status 150: termination requested by administrator
2/25/2014 10:57:56 AM - Error bpbrm(pid=4444) cannot send mail to mike.eid@erlworldwide.com,backup.alerts@erlworldwide.com,shukri.mahmud@erlworldwide.com        
2/25/2014 10:57:56 AM - Info bpbrm(pid=4444) sending media manager msg: TERMINATE        
2/25/2014 10:57:56 AM - end writing; write time: 00:00:21
file read failed(13)

 

I increased the timeout periods:

Host Properties>Master Servers>Timeouts

Client connect timeouts:  1000 seconds

Backup star notify timeout: 500 seconds

File browse timeout: 10,000 seconds

Client read timeout: 1000 seconds

Backup end notify timeouts: 500 seconds

Media server connect timeout: 60 seconds

 

 

Clients>Master Servers>Timeouts

Client read timeout: 10,000 seconds

File browse timeout: 10,000 seconds

 

 

Host Properties>Master Servers>Client Attributes>netbackup-01ny.hq.hview.com

Windows Open File Backup: off

 

7.1.x and Earlier
Backing Up
Backup and Recovery
NetBackup
Windows Server (2003-2008)
  • sdo's avatar
    sdo
    11 years ago
    See this: http://www.symantec.com/business/support/index?page=content&id=TECH166496

14 Replies

Sort By
  • sdo's avatar
    sdo
    Moderator
    11 years ago
    See this: http://www.symantec.com/business/support/index?page=content&id=TECH166496
  • sri_vani's avatar
    sri_vani
    Level 6
    11 years ago

    (bpbrm readline: socket read failed, An existing connection was forcibly closed by the remote host.  (10054)

    10053 and 10054 are related to network timeout issues

     

    Please make Client Read out time as 3600 and try running the backup again

  • Mark_Solutions's avatar
    Mark_Solutions
    Level 6
    11 years ago

    It fails almost instantly so timeouts do not look to be the issue

    I am assuming that the account you have set the client service (and maybe the netbackup legacy client network service) to use has full rights to those locations?

    Do the smaller shares you backup point to "real" shares?

    I am wondering if admin shares will not get passed to NetBackup correctly

    If possible back up the share rather than the admin share... so something like the below (no $):

    \\file-03ny\docs\Dead-My-Docs

    \\file-03ny\folders\RedirectFolders

    rather than

    \\file-03ny\redirect-adm$\Dead-My-Docs

    \\file-03ny\redirect-adm$\RedirectFolders

  • revarooo's avatar
    revarooo
    Level 6
    11 years ago

    The error occurs to quickly to be a read timeout.

    Have you tried running bpbkar -nocont on the UNC path?

     

    http://www.symantec.com/business/support/index?page=content&id=TECH17541

  • M-NYC's avatar
    M-NYC
    Level 3
    11 years ago

    I tried both the admin share and regular shares. When I didn't see a difference I put it back to the admin shares.

    I've done this type of backup months ago on a different NBU master server. So I am pretty sure that admin shares are ok.

     

    The account used for the services has full access to the volume. That part has been confirmed by a couple people over here.

  • M-NYC's avatar
    M-NYC
    Level 3
    11 years ago

    I increased all the time outs to 10,000 but it did not make a difference.

  • Mark_Solutions's avatar
    Mark_Solutions
    Level 6
    11 years ago

    Ok - setup bpbrm, bpcd and bpbkar logs and run it again - then post thise logs as attachments so that we can try and spot something

    Could we also see the policy output: bppllist policyname -U

    Thanks

  • sdo's avatar
    sdo
    Moderator
    11 years ago

    I agree with revaroo - try a bpbkar -nocont on the UNC path - to isolate whether the issue is with reading the file system.

  • Scott_Austin's avatar
    Scott_Austin
    Level 3
    11 years ago

    Also set the client_read and client_connect time outs to 900 seconds each. Enable the bpbkar log on the client and run the bpbkar to null. I would say over 80% of the time the problem is the client with UNC backups.