Re: amrecover hangs: bzip issue?

Ok, well I found the problem.. the broken pipe in the log file was me ctrl-c ing amrecover because it was taking way too long. What I found out was happening was that my actual backup file on the tape was 144 GB, so tar was taking forever to go through the tar file to find the actual file needed for the recover (all in all about 4 hours). I guess i need to figure out how to chunk things up a bit better.
-Kris

On Wed, 2004-08-04 at 15:31, Frank Smith wrote:

--On Wednesday, August 04, 2004 15:06:13 -0700 Kris Vassallo <kris AT linuxcertified DOT com> wrote:

> So looking into some logs I see a problem that looks like its whats
> causing everything to break.
> Inside of the log amidxtaped.20040804140128.debug, I am seeing some gzip
> errors. I have included a snippet of the log below which includes the
> error towards the bottom. Anyone know whats going on!?
> 
> amidxtaped: time 0.000: Ready to execv amrestore with:
> path = /usr/sbin/amrestore
> argv[0] = "amrestore"
> argv[1] = "-h"
> argv[2] = "-p"
> argv[3] = "file:/backup/amandadumps/tape01"
> argv[4] = "^venus.berkeley-da.com$"
> argv[5] = "^/home$"
> argv[6] = "20040728"
> amrestore:   0: skipping start of tape: date 20040728 label DailySet101
> amrestore:   1: skipping bda2.berkeley-da.com._var_ www.20040728.1
> <http://www.20040728.1> 
> amrestore:   2: skipping bda1.berkeley-da.com._home.20040728.2
> amrestore:   3: restoring venus.berkeley-da.com._home.20040728.0
> 
> gzip: stdout: Connection reset by peer
> Error 32 (Broken pipe) offset 2063728640+32768, wrote 0

Looks like something bombed when it hit 2GB, perhaps something in the
chain is not largefile aware?

I did hit a post on google that said although Sun claimed that bzip2 in
Solaris 8 was largefile aware, it actually wasn't, and they needed to
install a newer bzip2.  Try doing some testing with tar and bzip2 and
see if they can both handle files > 2GB.

Frank

> amrestore: pipe reader has quit in middle of file.
> amidxtaped: time 145.758: amrestore terminated normally with status: 2
> 
> Thanks, 
> Kris
> 
> 
> On Tue, 2004-08-03 at 16:01, Kris Vassallo wrote: 
> 
> I am still trying to get amrecover to work, but now I have a different
> issue which is that amrecover is hanging. I am recovering files on the
> client which is the same machine as the server and I am using a disk-tar
> backup. Here is what I typed in...
> amrecover> add .bash_profile
> Added /qa/.bash_profile
> amrecover> lcd /home
> amrecover> extract
> amrecover: warning: using /dev/null as the tape device will not work
> 
> Extracting files using tape drive /dev/null on host localhost.
> The following tapes are needed: DailySet101
> 
> Restoring files into directory /home
> Continue [?/Y/n]?
> 
> Extracting files using tape drive /dev/null on host localhost.
> Load tape DailySet101 now
> Continue [?/Y/n/t]? t
> New tape device [?]: localhost:file:/backup/amandadumps/tape01
> Using tape "file:/backup/amandadumps/tape01" from server localhost.
> Continue [?/Y/n/t]? Y
> 
> I told it to use the file /backup/amandadumps/tape01 which is what
> amanda knows as DailySet101. Anywho, at this point it just hangs until i
> ctrl-c it. Any ideas whats causing the hang?
> 
> 
>