And what failures do you get?

Listed below. Going to adv_file. But also fails going to tape. Majority of times it works. When it fails all we get is listed below:

nsrsnap: Printing savecmd=nsrsnap_save after parsing
nsrsnapck: Diagnostic: nsrsnapck completed successfully.
S:\
S:\: Preparing...
S:\: Requesting snapshot...
S:\: Saving...
S:\:
S:\: Error :Unknown error
nsrsnap_save: Backup Failed.
nsrsnap_save: Backup operation failed.
09/11/06 18:45:25 savegrp: failed.
09/11/06 18:45:25 savegrp: :index

(hostname removed)

Failure always seem to occur right at the end after backing up all the data.

Single save set defined. Have tried breaking it down into smaller chunks - same issue.

L9 debugging doesn't reveal much either other then this on the client in one of the trace files:

E:\rb_ps_mantle_221\nsr\snap\brc_mover\BRCopyModule.cpp 2434 11/09/2006 18:43:59 4408:Unit encountered error from remote agent. Error :CopyService failed to do its operation. Reason :nsrbragent: off 5785784 len 93884 file #21

It looks like session this break at mentioned file - reason could be either corruption or disk issue. Apparently snap module is very much sensitive to flare version and EMC tools version used in operation process. You certainly wish to open up this with support as they will inspect versions involved and give you debug binaries to get more data for engineer to troubleshoot.

As far as I know not many users use this feature yet. Most of them were already using BCV scripts before and were happy with it without any idea to spend more money on license fee required for module. At least that was my impression. Not sure if you will find many users using it on this forum as it is still small...

Thanks. We raised a support call with Legato back in June. So far we have been unable to identify the cause of the problem.

The post was more to see if anyone else has had similar issues with PSnap.

can you upgrade to the latest PS 2.2.1 module, there are a lot of LGTpa's for PS
you can try also
increase client parallelism to 8
Please set ps_group_objs = no
set ss_group_by_fs = no
do you get any cores in /nsr/cores ?
check your OS/ulimits

We are running PS 2.2.1. There are two LGTpa's that apply to PS 2.2.1 one for files & dirs > 255 characters on Windows and one for Solaris.

Both client & proxy are Windows 2003 SP1 so the later patch is not applicable. The > 255 patch makes no difference.

What do the two variables do?

Thanks.

I've been doing a bit of research and have discovered those variables are for use with the SAP/Oracle PS module. We are not using that, just trying to get it working backing up a FS.