We've been using legatos b2d option for about a year now and while we're satisfied with its performance as a backup device, we are very disappointed with the artificial limits imposed by the media manager daemon for restore jobs.
My primary concern lies in the fact that the single session available for restores is busy while old save sets are staged to tape, and this blocks all restores including saveset cloning for hours at a time, every day.
Now, I told this to my local EMC shop and was reassured that this problem has been fixed in release 7.2 of the networker server.
However after reading the release notes I realized they were probably refering to the new functionality in the command-line restore program, which now supports direct reads from disk (ie, circumventing the nsrmmd) *but only on clients with a local mount of the adv_file device!*
Obviously this is at best a slight improvement, and in my case it's pretty much useless. Cloning & database restores is where I'm hurting the most, and both of these use the nsrmmd to retrieve data.
Anyway, since this is a Big Problem for us currently I'm hoping that someone here may have some tips on how best to live with this artificial single read limit which nsrmmd imposes on restores from adv_file devices.
My primary concern lies in the fact that the single session available for restores is busy while old save sets are staged to tape, and this blocks all restores including saveset cloning for hours at a time, every day.
Now, I told this to my local EMC shop and was reassured that this problem has been fixed in release 7.2 of the networker server.
However after reading the release notes I realized they were probably refering to the new functionality in the command-line restore program, which now supports direct reads from disk (ie, circumventing the nsrmmd) *but only on clients with a local mount of the adv_file device!*
Obviously this is at best a slight improvement, and in my case it's pretty much useless. Cloning & database restores is where I'm hurting the most, and both of these use the nsrmmd to retrieve data.
Anyway, since this is a Big Problem for us currently I'm hoping that someone here may have some tips on how best to live with this artificial single read limit which nsrmmd imposes on restores from adv_file devices.