------=_Part_5876_1486628.1116414547830
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
I restored a 1GB catalog from an IBM LTO2 just last week. It took less than=
=20
1 minute to get the data off the tape.
Why are you using the -dhost parameter? Is the catalog restore running on a=
=20
media server, not the master?
Based on what worked for me, and what you're doing differently, try these :
- On the bprecover command, use device path /dev/rmt/11m (rather than 11cbn=
)
- Don't shutdown NetBackup before running the bprecover. I don't think I di=
d=20
(but I might be wrong)
Good luck,
Dean
Are you running this on a media sr
On 5/17/05, Marek Gorka <marek.gorka AT sun DOT com> wrote:
>=20
> Dear Helpful People,
>=20
> I do my CATALOG backup to tapes. I've set a testing procedure do restore=
=20
> one
> of the images I backup to tape. The image to restore is
> /opt/openv/netbackup/db directory.
>=20
> When I switch off the NBU, manually (robtest) mount the catalog backup=20
> tape
> into LTO2 drive and do
>=20
> bprecover -l -tpath /dev/rmt/11cbn
>=20
> I get the full listing of the images backed up.
>=20
> I "mv /opt/openv/netbackup/db /opt/openv/netbackup/nodb" and start the
> restore process:
>=20
> bprecover -r 1 -tpath /dev/rmt/11cbn -dhost SERVERNAME -v
>=20
> It tells me that /opt/openv/netbackup/db is being restored. Correct. In=
=20
> fact
> no activity (monitored by "iostat -xnz 2 | grep rmt") can be seen on
> /dev/rmt/11cbn. Nothing actually happens. I've left this process to last=
=20
> for
> 2 hours and nothing happened - though the "mt -f /dev/rmt/11cbn status"
> gives me "Drive is busy" message.
>=20
> The image to be restored is 200MB. I had to cancel this test after 2 hour=
s
> as this is a pruduction environment.
>=20
> Do you have any comments on that? Is it possible for the drive to scan=20
> this
> tape so terribly long? Or anything else?
>=20
> I expected (this proces to last less than 15 minutes.
>=20
> best regards,
>=20
> --
> Marek Gorka
> Client Solutions
> Sun Microsystems Poland
> _______________________________________________
> Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>
------=_Part_5876_1486628.1116414547830
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
I restored a 1GB catalog from an IBM LTO2 just last week. It took less than=
1 minute to get the data off the tape.<br>
<br>
Why are you using the -dhost parameter? Is the catalog restore running on a=
media server, not the master?<br>
<br>
Based on what worked for me, and what you're doing differently, try these :=
<br>
- On the bprecover command, use device path /dev/rmt/11m (rather than 11cbn=
)<br>
- Don't shutdown NetBackup before running the bprecover. I don't think I di=
d (but I might be wrong)<br>
<br>
Good luck,<br>
Dean<br>
<br>
<br>
Are you running this on a media sr<br><div><span class=3D"gmail_quote">On 5=
/17/05, <b class=3D"gmail_sendername">Marek Gorka</b> <<a href=3D"mailto=
:marek.gorka AT sun DOT com">marek.gorka AT sun DOT com</a>>
wrote:</span><blockquote =
class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, 204, 204); m=
argin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Dear Helpful People,<br><br>I do my CATALOG backup to tapes. I've set a tes=
ting procedure do restore one<br>of the images I backup to tape. The image =
to restore is<br>/opt/openv/netbackup/db directory.<br><br>When I switch of=
f the NBU, manually (robtest) mount the catalog backup tape
<br>into LTO2 drive and do<br><br>bprecover -l -tpath /dev/rmt/11cbn<br><br=
>I get the full listing of the images backed up.<br><br>I "mv /opt/ope=
nv/netbackup/db /opt/openv/netbackup/nodb" and start the<br>restore pr=
ocess:
<br><br>bprecover -r 1 -tpath /dev/rmt/11cbn -dhost SERVERNAME -v<br><br>It=
tells me that /opt/openv/netbackup/db is being restored. Correct. In fact<=
br>no activity (monitored by "iostat -xnz 2 | grep rmt") can be s=
een on
<br>/dev/rmt/11cbn. Nothing actually happens. I've left this process to las=
t for<br>2 hours and nothing happened - though the "mt -f /dev/rmt/11c=
bn status"<br>gives me "Drive is busy" message.<br><br>The i=
mage to be restored is 200MB. I had to cancel this test after 2 hours
<br>as this is a pruduction environment.<br><br>Do you have any comments on=
that? Is it possible for the drive to scan this<br>tape so terribly long? =
Or anything else?<br><br>I expected (this proces to last less than 15 minut=
es.
<br><br>best regards,<br><br>--<br>Marek Gorka<br>Client Solutions<br>Sun M=
icrosystems Poland<br>_______________________________________________<br>Ve=
ritas-bu maillist - <a href=3D"mailto:Veritas-bu@mail=
man.eng.auburn.edu">
Veritas-bu AT mailman.eng.auburn DOT edu</a><br><a
href=3D"http://mailman.eng.aubu=
rn.edu/mailman/listinfo/veritas-bu">http://mailman.eng.auburn.edu/mailman/l=
istinfo/veritas-bu</a><br></blockquote></div><br>
------=_Part_5876_1486628.1116414547830--
|
