The BRSB Software Engineering and Product Management teams are proud to announce the availability of NetWorker 8.0 SP4 for production environments.
The NetWorker 8.0 SP4 software contains the following new features and enhancements:
Parallel cloning for Data Domain clone-controlled replication
Parallel cloning is a feature of clone-controlled replication (CCR) for Data Domain systems that enables clone operations to occur concurrently, similar to parallel save set sessions.
By default, NetWorker performs up to 30 parallel concurrent cloning sessions or threads from each source clone pool per CCR operation. This is the recommended setting, but it can be changed as follows:
If you try to clone a save set that must be read from more than one source pool, then NetWorker will queue the backup pools. To avoid this, use a separate CCR operation for each pool.
1. Create a file named nsrcloneconfig, with no file extension, in the following location on the host that runs the CCR.
- On Microsoft Windows systems: C:\Program Files\EMC NetWorker\nsr\debug
- On UNIX systems: /nsr/debug
2. In this file, you can type the following two global and two resource-specific parameters:
max_threads_per_client=max concurrent save sets
nsrclone_resource_name_max_threads_per_client=max concurrent save sets
- max concurrent save sets limits the number of parallel clone sessions that NetWorker will run on any one source client, for example, 15.
- max clients limits the number of source clients that NetWorker will include in the parallel cloning operation, for example, 2.
- nsrclone_resource_name_ is the name of the specific NetWorker Clone resource you will use, added as a prefix to the parameter.
The numeric values multiplied must not exceed 30 or an error message results. If you use both global and resource-specific settings, then the specific settings override the global settings. To disable parallel cloning, set both values to 0. This will result in a queue with sequential cloning only.
Cloning by pools
In order to copy save sets from Data Domain storage to a device, a special pool must be specified. This pool is known as a clone pool. A clone pool must be assigned to a device on the target Data Domain system, where it will be available for use.
There are two main purposes for a clone pool:
- To copy existing deduplicated VTL or CIFS/NFS AFTD save sets to a Data Domain device.
- To copy the existing save sets from one Data Domain device to another Data Domain device, typically at a remote location for disaster recovery purposes.
Below is a list customer reported defects resolved in release 8.0 SP4.
193861 Aborted VSS Fileset is reported as competed in the media database
193421 Savegroup completion report of a successful Save Set all restart job incorrectly reports a number of clients as failed
193420 Allow recovery of VMs which have non-ASCII characters in annotation notes
193091 Update libDDBoost.so to include BOOST 126.96.36.199 library
192814 NetWorker hangs daily
192149 Disable “label” operation in AMM functionality: Data Domain devices unmount with “RPC severe Lost connection to media database”
190700 NMC Group Details backup size is incorrect when the save set spans multiple volumes
189423 Add to repository functionality using Client Push does not work for AIX packages
188816 Some storage nodes not mounting volumes after upgrade
187999 Duplicate resid error in jobd when running a probe based backup
187980 NDMP cloning selects the storage nodes from the affinity lists of the NetWorker server when there are no storage nodes
187655 Nsrjobsd creating core dumps on NetWorker
187640 Completed group shows status as “Interrupted” in NMC Group Details window
187629 Savegrp with probe based backup hangs intermittently, can only be killed with kill -9
187294 Save set all-local does not work on Solaris 11 non-global zone backup
186684 Nsrindexd core dumps on Linux 64-bit server
186668 Unresponsive nsrsnmd on the NetWorker sever causes the server to hang
186583 NDMP recover always selects first storage node defined in the Recover storage node affinity list
186374 Tape drives do not go into service mode even after maximum consecutive errors reached
NW160757 CCR spawns only 5 clone sessions in parallel while schedule cloning 30 save sets
NW160090 Nsrclone core dumps during cloning
NW159896 Expired save sets not deleted from NetWorker Cloud Device
NW159783 Savegroup status not updated properly when a non-critical volume fails during a disaster recovery backup
NW159072 GST backup marked as failed after upgrading the NetWorker server
NW159050 Setting cleaning interval to 300 months or more causes nsrmmgd and nsrd to exit
NW158786 nsrvadp_recover crashes when restoring VMs using VC 5.0.3
NW158768 Restored VM from Image Level backup using VADP is not bootable
NW158649 nsrmmdbd: Force a rebuild of Media database btree on startup of Server
NW158489 Nsrsnmd dies without generating the core files
NW158170 Cannot connect to NetWorker server or distribute NMC configuration after upgrade from NetWorker 7.6.3
NW157911 Nsrjobd query performance is poor when no type is specified
NW157907 Tapes getting marked Full prematurely with message “posix async write: Error 0”
NW157899 Group status remains as “running” when savegroup completes with the message “exited with return code 2”
NW157066 Nsrexecd fails to start with an error indicating it is already running
NW157018 Scheduled clone does not report completion in NMC Monitoring tab
NW157007 NetWorker deletes unexpired bootstrap save sets from disk volumes after touch “force_bootstrap_removal_from_disk”
NW156909 Unable to label new VTL tapes on AIX storage node after upgrade
NW156906 Jobid not being propagated in nsrclone when running CCR of multiple save sets in parallel
NW156890 Scheduled clone resource is re-enabled silently if disabled when clone job is in progress
NW156878 Faulting application: save.exe, Faulting module: libvss2008.dll
NW156699 Nsrd core dumps with NULL clone pool request in broker.c:device_available()
NW156728 Possible data loss after Media Database rebuild – bootstrap size reduced significantly
NW156305 Recover cores during a browse of an NDMP save set’s index when trying to change the time or list the directory contents
NW155951 NMC Group status reports success despite a failed NetWorker backup with input/output error
NW155917 Archiving: public archives is disabled but all users can retrieve files they do not own
NW155575 Recovery goes into a loop when attempting to overwrite busy files
NW155277 Analysis of core dump of nsrjobd
NW155211 gstclreport.bat returns null
NW155061 Nsrd loops on the error “failed to retrieve ssid when building clone rlist”
NW155055 nsrvadp_recover.exe crashes on restore of a VM
NW155015 When selecting multiple clients via NMC to update specific fields, an error is reported and the update fails
NW155003 Networker performance test shows CCR time increases 100% after each 0% incremental backups
NW154984 Clients created using all upper-case characters on a Linux server create a nested directory in the index directory
NW154781 Restarted savegroup does not completes, clients with skip schedule hang
NW154719 Running jobkill does not kill active jobs
NW154524 Checking resource types in the RAP database process is very slow
NW154461 DDS status messages incorrect after upgrade to NetWorker 8.0.2
NW154313 nsrsnmd Unable to initialize device: Request for reserved connection failed for device
NW154259 Running nsrsgrpcomp Group_name -R fails to generate a report for the requested jobID
NW152719 ubcc003a – nsrjobd “Broken pipe” to local addresses
NW151935 A downgrade from AD to local authentication causes user groups not to work
NW150984 Savegrp does not honor the -q option
NW146642 A save set defined with multi-byte characters is garbled in savegrp.log/messages on Windows
If you need a specific OS binary of this release, then make contact with EMC Technical support, and they will for sure give you the right thing you need.
NetWorker packages are available at: https://support.emc.com/downloads/1095_NetWorker
After downloading the necessary binary, make sure it is the good build!