PDDO Duplications fail with status 84 and PDDO Replication job returns 'Allowed memory size of 268435456 bytes exhausted'

Article:TECH148556  |  Created: 2011-01-19  |  Updated: 2012-02-26  |  Article URL http://www.symantec.com/docs/TECH148556
Article Type
Technical Solution


Issue



PDDO Duplications fail with status 84, 

Netbackup Job details  shows error ' cannot copy image from disk, bytesCopied = 18446744073709551615'.  

PureDisk Replication job shows 'Fatal error: Allowed memory size of 268435456 bytes exhausted Fatal error: Allowed memory size of 268435456 bytes exhausted'.


Error



Example of Job Details  showing error

7/12/2010 12:33:17 PM - requesting resource LCM_NBPD-STU01
7/12/2010 12:33:18 PM - Info nbrb(pid=2108) Limit has been reached for the logical resource LCM_NBPD-STU01   
7/12/2010 12:34:44 PM - granted resource LCM_NBPD-STU01
7/12/2010 12:34:45 PM - begin Duplicate
7/12/2010 12:34:45 PM - started process RUNCMD (5508)
7/12/2010 12:34:45 PM - ended process 0 (5508)
7/12/2010 12:34:46 PM - requesting resource NBPD-STU01
7/12/2010 12:34:46 PM - reserving resource @aaaaf
7/12/2010 12:34:46 PM - reserved resource @aaaaf
7/12/2010 12:34:46 PM - granted resource MediaID=@aaaap;DiskVolume=PureDiskVolume;DiskPool=nbpd-01;Path=PureDiskVolume;StorageServer=nbpd-01;MediaServer=me-01
7/12/2010 12:34:46 PM - granted resource NBPD-STU01
7/12/2010 12:34:48 PM - requesting resource @aaaaf
7/12/2010 12:34:49 PM - granted resource MediaID=@aaaaf;DiskVolume=PureDiskVolume;DiskPool=nbpd-01;Path=PureDiskVolume;StorageServer=nbpd-01;MediaServer=me-01
7/12/2010 12:34:51 PM - Info Duplicate(pid=5508) Initiating optimized duplication from @aaaaf to @aaaap     
7/12/2010 12:34:52 PM - started process bpdm (3864)
7/12/2010 12:35:06 PM - begin writing
7/12/2010 12:35:27 PM - Critical bpdm(pid=3864) sts_copy_extent failed: error 2060014 operation aborted      
7/12/2010 12:35:27 PM - Critical bpdm(pid=3864) image copy failed: error 2060014: operation aborted     
7/12/2010 12:35:28 PM - Error bpdm(pid=3864) cannot copy image from disk, bytesCopied = 18446744073709551615    
7/12/2010 12:35:28 PM - Critical bpdm(pid=3864) sts_close_handle failed: 2060022 software error       
7/12/2010 12:35:29 PM - Info nbme-dr-01(pid=3864) StorageServer=PureDisk:nbpd-01; Report=PDDO Stats for (nbpd-01): scanned: 0 KB, stream rate: 0.00 MB/sec, CR sent: 0 KB, dedup: 0.0%, cache hits: 0 (0.0%)
7/12/2010 12:35:47 PM - Error bpduplicate(pid=5508) host me-01 backup id DC-02_1291474800 optimized duplication failed, media write error (84).
7/12/2010 12:35:48 PM - Error bpduplicate(pid=5508) Duplicate of backupid DC-02_1291474800 failed, media write error (84).

PDDO Replication Job

*** Start: Replication Prepare ***
*** Error Message ***
shared.webserviceclient.2132
severity: 6
server: 1737000000
source: Replication_MBE_CLI
description:
 Webservice did not return valid information. No error nor valid output was found. Please check webservice on url: https://nbpd-01/spa/ws/ws_agent.php?login=<masked>&passwd=<masked>&isServerAgent=0&agentVersion=6.0.0.0&osType=12&macAddress=&hostName=%5BLBQ%5DR%5BRBQ%5D%20PDDO%20%28ag2%5BCO%5Dstp1737%29&locationName=Unknown%20location&departmentName=Unknown%20department&isvirtual=1&sourceStoragepoolId=1737&sourceAgentId=2&restorepassword=&type=2&osvariantid=10003&osextensions=Service%20Pack%202&__version=6.0.0.0&action=registerAgentMirror
Webservice returned:

Fatal error: Allowed memory size of 268435456 bytes exhausted (tried to allocate 71 bytes) in /opt/pdspa/mgmtclass/User.php on line 61

*** End ***
Agent Jobstep analysis: exitcode 1, status 3, progress 0.

Environment



NetBackup 7.0

PureDisk 6.6.0.3


Cause



Please be aware there can be several reasons why a duplication job can fail with status 84, this is one of them. 

The problem is caused by too many LDAP users being registered.

 


Solution



Make sure that there are no backups running, then stop PureDisk services:

# /etc/init.d/puredisk stop
# /etc/init.d/vxatd stop

Stop ldap to avoid any possible corruption

# service ldap stop      

Execute the following command to check the number of LDAP users:

# /opt/pdag/bin # slapcat | grep agent_ | wc -l

If this returns a number in the 5 digit range  eg 88348,  then  LDAP has too many users registered.

Solution is to install EEB20 'Rock Solid'   which resolves the problem via a design change.


Supplemental Materials

SourceETrack
Value2113788
Description

Most SLP fail. Memory exhausted in User.php, due to too many LDAP users 




Article URL http://www.symantec.com/docs/TECH148556


Terms of use for this information are found in Legal Notices