Preston,
Send me a copy of your brochure too.
We are running 7.0 on Solaris and doing some disk staging using a Legato
staging policy.
I had some simple cloning/staging scripts I used at a remote site to do
a weekly clone of mondays savesets in the past but would be interested
in what you have available.
A couple of problems we have with a storage-node with 6 adv_file
filesystems of ~1.7TB each and 6 corresponding staging policies we
invariably have a filesystem go full which pauses any savestreams going
there and then because we only have 2 tape drives available exclusively
for the staging of oldest savesets back to tape until a lower threshold
of ~70 percent is reached.
I'm going to try going with DDS next so that we share 6 of the 7
tapedrives we have in the library across the NetWorker server and a
storage node to help the staging work more reliably. I find myself
manually kicking it off.
How does NetWorker decide which adv_file device to go to - will it go
with the most full one or what ?
Thanks - John
-----Original Message-----
From: Legato NetWorker discussion [mailto:NETWORKER AT LISTSERV.TEMPLE DOT EDU]
On Behalf Of Preston de Guise
Sent: Monday, May 23, 2005 11:28 PM
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Subject: Re: [Networker] Problem with Staging in 7.2
Hi Carl,
> Thanks for the tip, Preston; I had recently concluded that an
> "upgrade" to 7.1.3 would be my next step, if I don't find an answer.
>
> Actually, I've since determined that it appears that it's the 5.x
> flavor savesets that are messing up the staging. Once I manually
> staged off all the Exchange savesets from a 2.0-006 Module, the
> automatic staging kicked in like clockwork.
>
> I assume that a Hotfix similar to the one to recover >2GB savesets is
> required for cloning/staging, but at least I've got a workaround.
Actually I've not seen a hotfix come out for that one, though I've seen
the problem occur with NetWare savesets. - in general it can happen
whenever you have a continuation saveset.
<might be construed as an ad - apologies in advance>
In the end I got somewhat tired of some of the limitations of the Legato
staging policies and wrote (on behalf of my company) an enhanced stage
manager program which resolves all sorts of issues - including the
problem you described there by "stepping around" those sorts of savesets
- i.e., isolating them and staging them individually. The stage manager
addressed a lot of conventional issues with NetWorker staging including:
- Device contention - NetWorker just stages "whenever", even if the disk
backup devices are busy;
- All or nothing - NetWorker stages everything that's eligible in one
hit, making it very annoying if you hit a bad saveset 99% through a 1TB
stage; ours instead does percentage based staging, freeing up disk space
faster;
- Volume cleaning;
- Bad saveset detection - NetWorker has some "feature" which results in
a (rare) situation where you can have a saveset referenced in the media
database which does not exist on the disk backup unit - this can result
in a situation where nsrd coredumps (unpatched) or nsrmmd aborts the
staging (patched)
- ...and so on...
(If you're interested in knowing more, let me know and I'll throw you a
brochure off-list for all the features - if you're staging a lot it may
help you out.)
</might be construed as an ad - apologies in advance>
Cheers,
-Preston.
--
Note: To sign off this list, send a "signoff networker" command via
email to listserv AT listserv.temple DOT edu or visit the list's Web site at
http://listserv.temple.edu/archives/networker.html where you can also
view and post messages to the list. Questions regarding this list should
be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listserv.temple DOT edu or visit the list's Web site at
http://listserv.temple.edu/archives/networker.html where you can
also view and post messages to the list. Questions regarding this list
should be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
