Thread: restore from tape
Howdy:
Running PostgreSQL 7.2.1 on RedHat Linux 7.2 kernel 2.4.7-10.
I'm revamping a lot of our backup process and trying to figure
out new methods for a lot of the ways we do business here.
For example, if I want to restore a table from a previous
dump, I would copy the file from where it's currently being
saved (a backup server that has an NFS mounted filesystem)
and do something like 'psql -U <name-of-table> -d <database>.
Now, (after about a year of begging and pleading) I have a
tape drive (Dell PowerVault 120T DLT4000 Tape Autoloader)
and I need to create a better process for backup and
recovery.
I am currently looking for a few things:
* New backup suggestions
* Any suggestions as to doing a pg_dump directly to
tape (if anyone else has done the same) and if
data recovers well.
+ An outstanding issue is that large tables (>2 Gig)
will terminate before completing
* Suggest low cost / free backup solutions (what can I say?
we're broke!) for an autoloader
+ I've heard about 'Amanda', but, I don't have the
space to spool files and THEN back them up
That's about all for now. Thanks!
-X
Shaunn, I've placed a number of scripts at: http://services.csl.co.uk/postgresql/pgdbbackup.tar.gz The db_backup script backups up a database by first dumping to a temporary file and then to tape along with index info. This index info becomes useful in db_listtape and db_restorespecific. Naturally a bit of specifics in there for our internal use, but not too much and should be useful all the same. Has code to cope with both remote databases and remote tape devices. It is designed to be used via an encapsulating GUI, but simple to work without it... Lee. Johnson, Shaunn writes: > Howdy: > > Running PostgreSQL 7.2.1 on RedHat Linux 7.2 kernel 2.4.7-10. > > I'm revamping a lot of our backup process and trying to figure > out new methods for a lot of the ways we do business here. > > For example, if I want to restore a table from a previous > dump, I would copy the file from where it's currently being > saved (a backup server that has an NFS mounted filesystem) > and do something like 'psql -U <name-of-table> -d <database>. > > Now, (after about a year of begging and pleading) I have a > tape drive (Dell PowerVault 120T DLT4000 Tape Autoloader) > and I need to create a better process for backup and > recovery. > > I am currently looking for a few things: > > * New backup suggestions > > * Any suggestions as to doing a pg_dump directly to > tape (if anyone else has done the same) and if > data recovers well. > + An outstanding issue is that large tables (>2 Gig) > will terminate before completing > > * Suggest low cost / free backup solutions (what can I say? > we're broke!) for an autoloader > + I've heard about 'Amanda', but, I don't have the > space to spool files and THEN back them up > > > That's about all for now. Thanks! > > -X > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> > <HTML> > <HEAD> > <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=US-ASCII"> > <META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2653.12"> > <TITLE>restore from tape</TITLE> > </HEAD> > <BODY> > > <P><FONT SIZE=2>Howdy:</FONT> > </P> > > <P><FONT SIZE=2>Running PostgreSQL 7.2.1 on RedHat Linux 7.2 kernel 2.4.7-10.</FONT> > </P> > > <P><FONT SIZE=2>I'm revamping a lot of our backup process and trying to figure</FONT> > <BR><FONT SIZE=2>out new methods for a lot of the ways we do business here.</FONT> > </P> > > <P><FONT SIZE=2>For example, if I want to restore a table from a previous</FONT> > <BR><FONT SIZE=2>dump, I would copy the file from where it's currently being</FONT> > <BR><FONT SIZE=2>saved (a backup server that has an NFS mounted filesystem)</FONT> > <BR><FONT SIZE=2>and do something like 'psql -U <name-of-table> -d <database>.</FONT> > </P> > > <P><FONT SIZE=2>Now, (after about a year of begging and pleading) I have a</FONT> > <BR><FONT SIZE=2>tape drive (Dell PowerVault 120T DLT4000 Tape Autoloader)</FONT> > <BR><FONT SIZE=2>and I need to create a better process for backup and </FONT> > <BR><FONT SIZE=2>recovery.</FONT> > </P> > > <P><FONT SIZE=2>I am currently looking for a few things:</FONT> > </P> > > <P><FONT SIZE=2>* New backup suggestions</FONT> > </P> > > <P><FONT SIZE=2>* Any suggestions as to doing a pg_dump directly to</FONT> > <BR><FONT SIZE=2> tape (if anyone else has done the same) and if </FONT> > <BR><FONT SIZE=2> data recovers well.</FONT> > <BR><FONT SIZE=2> + An outstanding issue is that large tables (>2Gig)</FONT> > <BR><FONT SIZE=2> will terminate before completing</FONT> > </P> > > <P><FONT SIZE=2>* Suggest low cost / free backup solutions (what can I say?</FONT> > <BR><FONT SIZE=2> we're broke!) for an autoloader</FONT> > <BR><FONT SIZE=2> + I've heard about 'Amanda', but, I don't havethe </FONT> > <BR><FONT SIZE=2> space to spool files and THEN backthem up</FONT> > </P> > <BR> > > <P><FONT SIZE=2>That's about all for now. Thanks!</FONT> > </P> > > <P><FONT SIZE=2>-X</FONT> > </P> > > </BODY> > </HTML>