Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

How to figure out what volumes an expired save set is on

Status
Not open for further replies.

collective7

Technical User
Mar 12, 2003
44
0
0
US
Our retention period is 3 weeks, they want a restore from 2 months ago. I know i have to scann the tapes in to restore the indexes, but how do i find out what volumes to scann?

PS: No nwadmin gui available, command line only.

NWrkr 5.5.4 on Solaris 8
 
Do you know the save set id? (ssid). If not, you can construct a command-line query to mminfo to help find it, although I generally use a larger query and grep to filter through the data.

mminfo -v -q client='client_name' | grep mm/dd/yy
and troll around with that a little bit. once you've settled on an ssid to recover from, that same report will also tell you which volumes to use.

Noting that you're using version 5, there's the possibility that you may have saveset chains if any of your savesets exceed 2G. I've got a script I use for chasing relations like that around, let me know if you need it.
 
Chapter11,

I'm going to have to buy you dinner dude! Thanks for all your help!

I do have saveset chains.. can you email me your script?

collective_rhythm@yahoo.com

a million thanks
 
eef. Ok, as we are all probably aware, Networker's reporting capability is less than stellar. The script I normally use for chasing down chains happens to be highly dependent on some other stuff I've done that's too lengthy to really go into here. (specifically, I export networker's data into mysql).

So, here is a version of my script that's designed (and tested) to work with networker without any localized extensions. The problem is that the pssid field in the media database isn't indexed on, and that's one of the fields you have to use to troll through the media index, so there's a lot of what DBAs call "full table scans" going on. The short version is that it takes a while to run.

calling convention: ssfr.pub <ssid>
method of operation: using the ssid, get the pssid. Keep doing this until you get 0. That will be the head of the ss chain.
Output the record of that ssid. Find the record whose pssid matches the current ssid. if no record is found, we are at the end of the ss chain, otherwise, keep doing this.

ssfr.pub (saveset find related, public version)
[tt]#!/bin/ksh

# check args

if test &quot;${1}&quot; = &quot;&quot;
then
echo &quot;error: must include an ssid&quot;
exit
else
SSID=${1}
fi

# set up vars

WALK=${SSID}
HEAD=1


# step 1: find the beginning of this ss chain
# when pssid=0, we've found the head

while test ${WALK} -gt 0
do
C=$((`mminfo -r pssid -q ssid=${WALK}`))
if test ${C} -gt 0
then
HEAD=${C}
fi
WALK=${C}
done

# step two, output the children searching where pssid=ssid

WALK=${HEAD}

while test ${WALK} -gt 0
do
LINE=`mminfo -v -q ssid=${WALK} 2>/dev/null | grep -v &quot;volume&quot;`
if test &quot;${LINE}&quot; != &quot;&quot;
then
echo &quot;${LINE}&quot;
WALK=$((`mminfo -r ssid -q pssid=${WALK}`))
fi
done[/tt]
 
You know, i tried to play with the mminfo command, read the man page and twisted it around some. But no matter what I do, i just cant get any saveset from before 3/17

If our retention period is 3 weeks, would a saveset from 2/19 be overwritten? (current date is 4/10)
 
Hello,

I copied the script to my machine and start &quot;playing&quot; with it... I'll let you know my findings....

But... can't you easily use recover -S ssid ?

It depends what kind of data it is....

Ron
 
Retention period is the amount of time a saveset will stay in the media database. Once the retention period is exceeded, the saveset is deleted from the media database, and will no longer show up anywhere. You will *have* to rescan the media it was on, and since it's deleted, there's no way of getting Networker to tell you.

The &quot;other&quot; period is browse period, this is the amount of time a saveset's file indexes are stored, allowing a browsable recovery.

At my site, browse period is 30 days, but retention period is much longer - 90 days for incrs, 1 years for weekly fulls, and (effectively) permanent for weekly fulls that are also monthlys.

Without any manual changes to a saveset's retention period, the retention period represents the furthest you can go back. If your retention period is set to 3 weeks, you will never be able to go back further than 3 weeks for restores.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top