Re: changer problems with 2.6.1
On Thu, Feb 19, 2009 at 9:07 AM, stan wrote: > So, in addition to thanks, I have a question. Since this _is_ the direction > we have chosen to go in, I would like to put the requisite perl modules in > this tree also. Is there a way to get the build process to do this? Sure thing! See --with-amperldir: --with-amperldir[=PATH] Where amanda's perl modules are installed; default: installsitelib --without-amperldir Install amanda's perl modules in $amlibdir/perl So the default is to install them in perl's own directory, but you can specify where to put them, or just say --without-amperldir to put them in (in your case) /opt/amanda/lib/amanda/perl/. Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
Good morning. Thanks to all you patient help, we seem to be very close to getting the new client working! Thanks. We mad a bad decision years ago to store all of the Amanda related stuff in /opt/amanda. This would include the Amado software, configs, and history. This makes upgrading a bit tricky. Having said that, I think we are stuck with this, as it would be a huge change to redo it at the moment, so we are going to document the upgrade procedure required by this. So, in addition to thanks, I have a question. Since this _is_ the direction we have chosen to go in, I would like to put the requisite perl modules in this tree also. Is there a way to get the build process to do this? -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 05:27:16PM -0500, Dustin J. Mitchell wrote: > On Wed, Feb 18, 2009 at 5:22 PM, Frank Smith wrote: > > Quick and dirty fix is to add the directory where the Amanda libraries > > are to your /etc/ld.so.conf and run ldconfig. > > Better (IMHO) option is to recompile Amanda either statically linked > > or with runtime library paths. Otherwise you risk the problem of other > > programs potentially linking in the wrong libraries (although probably > > not an issue if the only ones there are the Amanda libraries, since the > > libam* namespace is not too likely to collide with other common libraries). > > Thanks, Frank! The runtime library paths suggestion is a good one. You can > use > ./configure .. LDFLAGS="-R/opt/amanda-2.6.1_2009mmdd" .. > for that, if my memory of Solaris hasn't failed me. > The server is on Linux, but I will work through the solution, given this clue. Thanks. -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 05:37:16PM -0500, Dustin J. Mitchell wrote: > On Wed, Feb 18, 2009 at 4:49 PM, stan wrote: > > This is a real problem for established (longtime) users of Amanda that have > > many, many machines - many flavors of OS, some very old hardware and > > mission critical production software that can't be replaced. > ... > > It appears as though a lot of useful, and interesting new functionality is > > being added to Amanda, and I will regret not being able to take advantage > > of this, but our job is to keep the control systems operating, everything > > else must come second to that. > > I certainly do not want to be responsible for boilers running dry -- > that sounds dangerous! > > As you indicate in your final paragraph, this was a decision we made > some time ago, after much deliberation. I'm keen on the idea of > "freezing" and stripping down a perl-free, Glib-free version of the > Amanda client for cases such as yours. Certainly, nothing we're doing > precludes this option -- client/server backward compatibility is > paramount -- but we don't have the resources to create and maintain > such a mini-client. > Yes, I think keeping the client build requirments as simple as possible is a very good idea. If I still had the resources I once did (sigh) I would probably do that as in "in house" project, and put the changes back into the mainstram code. But we don't have those resources anymore. Thanks for your help. -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 4:49 PM, stan wrote: > This is a real problem for established (longtime) users of Amanda that have > many, many machines - many flavors of OS, some very old hardware and > mission critical production software that can't be replaced. ... > It appears as though a lot of useful, and interesting new functionality is > being added to Amanda, and I will regret not being able to take advantage > of this, but our job is to keep the control systems operating, everything > else must come second to that. I certainly do not want to be responsible for boilers running dry -- that sounds dangerous! As you indicate in your final paragraph, this was a decision we made some time ago, after much deliberation. I'm keen on the idea of "freezing" and stripping down a perl-free, Glib-free version of the Amanda client for cases such as yours. Certainly, nothing we're doing precludes this option -- client/server backward compatibility is paramount -- but we don't have the resources to create and maintain such a mini-client. This might make an interesting project for someone here - any takers? Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 5:22 PM, Frank Smith wrote: > Quick and dirty fix is to add the directory where the Amanda libraries > are to your /etc/ld.so.conf and run ldconfig. > Better (IMHO) option is to recompile Amanda either statically linked > or with runtime library paths. Otherwise you risk the problem of other > programs potentially linking in the wrong libraries (although probably > not an issue if the only ones there are the Amanda libraries, since the > libam* namespace is not too likely to collide with other common libraries). Thanks, Frank! The runtime library paths suggestion is a good one. You can use ./configure .. LDFLAGS="-R/opt/amanda-2.6.1_2009mmdd" .. for that, if my memory of Solaris hasn't failed me. Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
Quick and dirty fix is to add the directory where the Amanda libraries are to your /etc/ld.so.conf and run ldconfig. Better (IMHO) option is to recompile Amanda either statically linked or with runtime library paths. Otherwise you risk the problem of other programs potentially linking in the wrong libraries (although probably not an issue if the only ones there are the Amanda libraries, since the libam* namespace is not too likely to collide with other common libraries). Frank stan wrote: > On Wed, Feb 18, 2009 at 03:23:15PM -0500, Dustin J. Mitchell wrote: >> On Wed, Feb 18, 2009 at 3:06 PM, stan wrote: >>> OK, now we have uncovered what is probably the root cause of my problems: >> Yay! >> >>> It exists and seems to have a set of permissions that I would think would >>> work. >> Try running 'ldd' on it -- perhaps it cannot find the amanda libraries >> it links against? > > Seems that you are on the right track: > > ama...@amanda:/opt/amanda.2.6.1_02172009/sbin$ ldd > /usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so > libamglue.so => not found > libamanda-2.6.1.so => not found > libm.so.6 => /lib/libm.so.6 (0x2b7c9f05f000) > libgmodule-2.0.so.0 => /usr/lib/libgmodule-2.0.so.0 > (0x2b7c9f2e) > libdl.so.2 => /lib/libdl.so.2 (0x2b7c9f4e3000) > libgobject-2.0.so.0 => /usr/lib/libgobject-2.0.so.0 > (0x2b7c9f6e8000) > libgthread-2.0.so.0 => /usr/lib/libgthread-2.0.so.0 > (0x2b7c9f92c000) > librt.so.1 => /lib/librt.so.1 (0x2b7c9fb3) > libglib-2.0.so.0 => /usr/lib/libglib-2.0.so.0 > (0x2b7c9fd3a000) > libnsl.so.1 => /lib/libnsl.so.1 (0x2b7ca0007000) > /lib/libresolv.so.2 (0x2b7ca022) > libpthread.so.0 => /lib/libpthread.so.0 (0x2b7ca0436000) > libc.so.6 => /lib/libc.so.6 (0x2b7ca0651000) > /lib64/ld-linux-x86-64.so.2 (0x4000) > > Any thoughts as to how to correct this? -- Frank Smith fsm...@hoovers.com Sr. Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 03:23:15PM -0500, Dustin J. Mitchell wrote: > On Wed, Feb 18, 2009 at 3:06 PM, stan wrote: > > OK, now we have uncovered what is probably the root cause of my problems: > > Yay! > > > It exists and seems to have a set of permissions that I would think would > > work. > > Try running 'ldd' on it -- perhaps it cannot find the amanda libraries > it links against? Seems that you are on the right track: ama...@amanda:/opt/amanda.2.6.1_02172009/sbin$ ldd /usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so libamglue.so => not found libamanda-2.6.1.so => not found libm.so.6 => /lib/libm.so.6 (0x2b7c9f05f000) libgmodule-2.0.so.0 => /usr/lib/libgmodule-2.0.so.0 (0x2b7c9f2e) libdl.so.2 => /lib/libdl.so.2 (0x2b7c9f4e3000) libgobject-2.0.so.0 => /usr/lib/libgobject-2.0.so.0 (0x2b7c9f6e8000) libgthread-2.0.so.0 => /usr/lib/libgthread-2.0.so.0 (0x2b7c9f92c000) librt.so.1 => /lib/librt.so.1 (0x2b7c9fb3) libglib-2.0.so.0 => /usr/lib/libglib-2.0.so.0 (0x2b7c9fd3a000) libnsl.so.1 => /lib/libnsl.so.1 (0x2b7ca0007000) /lib/libresolv.so.2 (0x2b7ca022) libpthread.so.0 => /lib/libpthread.so.0 (0x2b7ca0436000) libc.so.6 => /lib/libc.so.6 (0x2b7ca0651000) /lib64/ld-linux-x86-64.so.2 (0x4000) Any thoughts as to how to correct this? -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 03:23:15PM -0500, Dustin J. Mitchell wrote: > On Wed, Feb 18, 2009 at 3:06 PM, stan wrote: > > You'll need to stick to older versions of Amanda for such systems, or > install a "private" copy of perl against which you build Amanda. > This is a real problem for established (longtime) users of Amanda that have many, many machines - many flavors of OS, some very old hardware and mission critical production software that can't be replaced. btw, we have been using amanda for over 10 years now and have about 60 clients, 1 server. The reason for attempting this upgrade at all is because we put in a new server, upgraded its amanda version & even at 2.5.x it 'broke' the recover function on the older clients (but backup was still working fine). Thought we might as well get the whole system updated to compatible server/client versions. For example, we have the following OSes currently in service: Linux 32 & 64 bit, many flavors FreeBSD from v4.x thru 7.x OpenBSD v3.9 thru current Solaris 5, 8, 10 HP-UX 10.20 And many of these machines are control systems in a manufacuring plant (process control in a paper mill & power generation real-time control applications) that can't be upgraded to a newer version because the control software is frozen to a particular platform/version. ie. we ran into one client today, Solaris 8 client with this build error for 2.6.1 client only build. btw, this client controls the water treatment plant. This plant processes thousands of gallons of h2o a day to be used in boilers. checking for perl... /usr/bin/perl checking for perl version greater than or equal to 5.6.0... no configure: error: Amanda requires at least perl 5.6.0 AW0550# perl --version This is perl, version 5.005_03 built for sun4-solaris Copyright 1987-1999, Larry Wall Looks like the last good build was 2.5.1p2... AW0550# ls .. amanda-2.4.1p1 amanda-2.5.0amanda-2.5.1p2.tar amanda-2.4.3b4 amanda-2.5.0.taramanda-2.6.1 amanda-2.4.4p2 amanda-2.5.1p2 amanda-2.6.1.tar So if it is looking like 2.6.x is going to be a lost cause for us, can you advise on what back version will be our best bet? Will server/clients at 2.5.2p1 cause any of these build problems? Will we need to go back to 2.5.1p2 ? Any ideas? It appears as though a lot of useful, and interesting new functionality is being added to Amanda, and I will regret not being able to take advantage of this, but our job is to keep the control systems operating, everything else must come second to that. -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 3:06 PM, stan wrote: > OK, now we have uncovered what is probably the root cause of my problems: Yay! > It exists and seems to have a set of permissions that I would think would > work. Try running 'ldd' on it -- perhaps it cannot find the amanda libraries it links against? > Can I also register a concern? In the past, we have done work where we created > our own perl modules. One of the problems with that is that when you > upgrade perl, you have to go back and recreate all your own perl modules, > because perl puts these in a directory that is specific to the version. So, > for example on the Ubuntu server, I may accidently break Amanda, just by > doing an apt-get upgrade to isntall a fixed perl :-( Yeah, this is something we've put some thought into. This is the sort of problem that package-management systems should, and have, solved. If you are installing with debs, then the Amanda deb will depend on the version of perl you have installed. If you upgrade that, you'll need to upgrade your Amanda package, too. If you're installing from source, you can use --with-amperldir to put the Amanda perl libraries elsewhere, although I don't know how reliably forward-binary-compatible perl is. > Also, we have clients with no perl installed on them. Is the use of perl > confined to the server? Some of the clients are old versions of Solaris, > HP-UX, etc. and we can't change them because they are production machines > in an industrial manufacturing environment. If they won't work with 2.6.1 > we might have to backtrack to 2.5.2 (or 2.5.1). You'll need to stick to older versions of Amanda for such systems, or install a "private" copy of perl against which you build Amanda. Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 12:28:24PM -0500, Dustin J. Mitchell wrote:
> On Wed, Feb 18, 2009 at 11:30 AM, stan wrote:
> > However, what I am suspecting is that we have been using a "corner case"
> > for years which may not work now. The new design appears to be able to more
> > correctly
> > describe our confguration, so I am suspecting that I need to change the
> > basic config to one using chg-rait, as i posted a minute ago. Or would you
> > suggest that we continue to use the older config, and debug it?
>
> I think your case is actually pretty typical. The advantage of using
> chg-rait is that you can employ chg-zd-mtx to operate a tape robot for
> your physical tapes. If you have no such robot, then the chg-multi
> solution is simpler. Also, the chg-rait shell script seems to be a
> bit buggy. I'm about to commit a new Amanda::Changer::rait, but
> obviously that's not in 2.6.1 :(
>
> > Now Here is the output of a failed run with 2.6.1 using the 2.6.1 chg-multi:
> >
> > arguments -> -info
> > Note: setting posteject to a default of true
> > Exit -> 2 25 1
> > arguments -> -slot current
> > Note: setting posteject to a default of true
> > Exit -> 2 chg-multi: slot is empty:
> (etc.)
>
> chg-multi is running 'amdevcheck' on each device to find out if there
> is a volume loaded, and apparently deciding that nothing is loaded.
> What is the output of
> amdevcheck DailyDump "rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}"
OK, now we have uncovered what is probably the root cause of my problems:
ama...@amanda:/opt/amanda.2.6.1_02172009/sbin$ ./amdevcheck DailyDump
"rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}"
Can't load '/usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so' for
module Amanda::Types: libamglue.so: cannot open shared object file: No such
file or directory at /usr/lib/perl/5.8/DynaLoader.pm line 225.
at /usr/local/share/perl/5.8.8/Amanda/Types.pm line 11
Compilation failed in require at
/usr/local/share/perl/5.8.8/Amanda/Device.pm line 10.
Compilation failed in require at ./amdevcheck line 24.
BEGIN failed--compilation aborted at ./amdevcheck line 24.
ama...@amanda:/opt/amanda.2.6.1_02172009/sbin$ vi +24 amdev
So the perl script cannot load the required module. Yet:
ama...@amanda:/usr/local/share/perl/5.8.8/auto/Amanda/Types$ ls -l
/usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so
-rwxr-xr-x 1 root root 199639 2009-01-28 10:25
/usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so
It exists and seems to have a set of permissions that I would think would
work.
file seems to like it:
ama...@amanda:/usr/local/share/perl/5.8.8/auto/Amanda/Types$ file
/usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so
/usr/local/share/perl/5.8.8/auto/Amanda/Types/libTypes.so: ELF 64-bit LSB
shared object, x86-64, version 1 (SYSV), not stripped
ama...@amanda:/usr/local/share/perl/5.8.8/auto/Aman
I have to admit, I have never had perl dynamically load a shared object
library, so I don't quite know where to go at this point.
Can I also register a concern? In the past, we have done work where we created
our own perl modules. One of the problems with that is that when you
upgrade perl, you have to go back and recreate all your own perl modules,
because perl puts these in a directory that is specific to the version. So,
for example on the Ubuntu server, I may accidently break Amanda, just by
doing an apt-get upgrade to isntall a fixed perl :-(
Also, we have clients with no perl installed on them. Is the use of perl
confined to the server? Some of the clients are old versions of Solaris,
HP-UX, etc. and we can't change them because they are production machines
in an industrial manufacturing environment. If they won't work with 2.6.1
we might have to backtrack to 2.5.2 (or 2.5.1).
--
One of the main causes of the fall of the roman empire was that, lacking
zero, they had no way to indicate successful termination of their C
programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 11:30 AM, stan wrote:
> However, what I am suspecting is that we have been using a "corner case" for
> years which may not work now. The new design appears to be able to more
> correctly
> describe our confguration, so I am suspecting that I need to change the
> basic config to one using chg-rait, as i posted a minute ago. Or would you
> suggest that we continue to use the older config, and debug it?
I think your case is actually pretty typical. The advantage of using
chg-rait is that you can employ chg-zd-mtx to operate a tape robot for
your physical tapes. If you have no such robot, then the chg-multi
solution is simpler. Also, the chg-rait shell script seems to be a
bit buggy. I'm about to commit a new Amanda::Changer::rait, but
obviously that's not in 2.6.1 :(
> Now Here is the output of a failed run with 2.6.1 using the 2.6.1 chg-multi:
>
> arguments -> -info
> Note: setting posteject to a default of true
> Exit -> 2 25 1
> arguments -> -slot current
> Note: setting posteject to a default of true
> Exit -> 2 chg-multi: slot is empty:
(etc.)
chg-multi is running 'amdevcheck' on each device to find out if there
is a volume loaded, and apparently deciding that nothing is loaded.
What is the output of
amdevcheck DailyDump "rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}"
If this is "SUCCESS", then you may also want to edit chg-multi
temporarily to echo its $device to the debug file:
421 exit 2
422 fi
echo `_ 'Note: checking amdevcheck $device'` >> $logfile
423 amdevcheck_status $device
424 if [ $? -ne 0 ]; then
425 #
--
Storage Software Engineer
http://www.zmanda.com
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 10:56:24AM -0500, Dustin J. Mitchell wrote:
> On Wed, Feb 18, 2009 at 10:44 AM, stan wrote:
> > Can you make any suggestions as to how to come up with a configuration for
> > the new system that will replicate the existing functionality.
>
> What you have shown here looks fine, and should still be functional,
> although the global tapedev is not necessary when using chg-multi. In
> my earlier email, I suggested some places you could start looking for
> information as to what's happening -- so far all I've seen is "fails
> to be able to find a valid slot". I think you've gotten distracted by
> the 'mt' issue and the _ function, neither of which are likely to be
> relevant to the problem at hand.
Agreed,
However, what I am suspecting is that we have been using a "corner case" for
years which may not work now. The new design appears to be able to more
correctly
describe our confguration, so I am suspecting that I need to change the
basic config to one using chg-rait, as i posted a minute ago. Or would you
suggest that we continue to use the older config, and debug it?
> Please check out the logfiles and
> other sources of clues I mentioned earlier, and post the interesting
> bits here if you need a hand analyzing them.
>
> You'll find that chg-multi logs to "changer.debug" in your debug log
> directory:
> 29 if [ -d "@AMANDA_DBGDIR@" ]; then
> 30 logfi...@amanda_dbgdir@/changer.debug
> 31 else
> 32 logfile=/dev/null
> 33 fi
>
> you can find the value of AMANDA_DBGDIR with
> amgetconf build.AMANDA_DBGDIR
>
> The debug logs from various 'amtape' invocations may also be enlightening.
Right, we found that last night. It looks like it appends the debug lines
for each run onto the same file, right?
It appears that this is the output of a "good" run with the 2.5.2p1 using
the 2.5.1.p2 chg-multi script:
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
Exit -> 1 25 1
Args -> -slot current
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
-> rewind rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}
Exit -> 1 rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}
Args -> -slot next
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
-> rewind rait:{file:/vtapes/DailyDump/vtape2,tape:/dev/nst0}
Exit -> 2 rait:{file:/vtapes/DailyDump/vtape2,tape:/dev/nst0}
Args -> -slot next
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
-> rewind
rait:{file:/vtapes/DailyDump/vtape3,tape:/dev/nst0}
Exit -> 3
rait:{file:/vtapes/DailyDump/vtape3,tape:/dev/nst0}
Args -> -info
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
Exit -> 3 25 1
Args -> -slot current
MT -> /opt/amanda/sbin/ammt -f
Note: setting posteject to a default of "true"
-> rewind
rait:{file:/vtapes/DailyDump/vtape3,tape:/dev/nst0}
Exit -> 3
rait:{file:/vtapes/DailyDump/vtape3,tape:/dev/nst0}
Now Here is the output of a failed run with 2.6.1 using the 2.6.1 chg-multi:
arguments -> -info
Note: setting posteject to a default of true
Exit -> 2 25 1
arguments -> -slot current
Note: setting posteject to a default of true
Exit -> 2 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 3 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 4 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 5 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 6 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 7 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 8 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 9 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 10 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 11 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 12 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 13 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 14 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 15 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 16 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 17 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of true
Exit -> 18 chg-multi: slot is empty:
arguments -> -slot next
Note: setting posteject to a default of tr
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 10:44:48AM -0500, stan wrote: > On Tue, Feb 17, 2009 at 11:33:12PM -0500, Dustin J. Mitchell wrote: > > On Tue, Feb 17, 2009 at 11:06 PM, stan wrote: > > OK, a few pieces of Amanda history may be relevant to your investigations > > here. OK, so I have looked at the web pages you pointed me too`. Let me explain the way that the existing system works. Using the "backup" config, the "changer" checks the label on each of the vtapes, till it finds the matching label on the tape in the drive, or fails to. Assuming case 1 then amanda sets the "slot" to the slot that matches the label on the tape, and the backup begins. Looks to me like what we have to do in the new design is use chg-rait(?), with 3 devices, one to throw away the parity that we don't need, chg-disk, for the vtapes, and ?? for the single physical slot tapedrive itself. Does this sound right? What changer should we use for the tape? -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 10:44 AM, stan wrote: > Can you make any suggestions as to how to come up with a configuration for > the new system that will replicate the existing functionality. What you have shown here looks fine, and should still be functional, although the global tapedev is not necessary when using chg-multi. In my earlier email, I suggested some places you could start looking for information as to what's happening -- so far all I've seen is "fails to be able to find a valid slot". I think you've gotten distracted by the 'mt' issue and the _ function, neither of which are likely to be relevant to the problem at hand. Please check out the logfiles and other sources of clues I mentioned earlier, and post the interesting bits here if you need a hand analyzing them. You'll find that chg-multi logs to "changer.debug" in your debug log directory: 29 if [ -d "@AMANDA_DBGDIR@" ]; then 30 logfi...@amanda_dbgdir@/changer.debug 31 else 32 logfile=/dev/null 33 fi you can find the value of AMANDA_DBGDIR with amgetconf build.AMANDA_DBGDIR The debug logs from various 'amtape' invocations may also be enlightening. Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
On Tue, Feb 17, 2009 at 11:33:12PM -0500, Dustin J. Mitchell wrote:
> On Tue, Feb 17, 2009 at 11:06 PM, stan wrote:
> OK, a few pieces of Amanda history may be relevant to your investigations
> here.
>
> First, Amanda no longer assumes it is using a tape drive (that is,
> something which 'mt' might be able to operate). This is a part of the
> Device API which was introduced in 2.6.0. The only point at which
> changer scripts need to interact with a device is to determine whether
> it's ready or not -- in chg-zd-mtx and chg-manual. For this, these
> scripts now use amdevcheck. This has brought its own set of issues,
> but is fully functional and does not involve 'mt' in any way.
>
> Second, there are a number of other changes to the changer API in
> 2.6.1 -- see amanda-changers(7) and
> http://wiki.zmanda.com/index.php/Changer_API for a taste. Among lots
> of other interesting action, we're deprecating and removing all of the
> old shell scripts. At the moment, the new Changer API is *always*
> used, via a C-to-perl (and old-to-new) wrapper script
> (changer-src/chg-glue.pl[1]) and a perl-to-shell (and new-to-old)
> wrapper script (perl/Amanda/Changer/compat.pm). There's room for
> problems at several points along this chain. I'm busily at work
> rewriting changers into the new API, and rewriting the Amanda taper in
> Perl so that it can utilize the new API directly. You can see the new
> Amanda::Changer::rait at
> http://github.com/djmitche/amanda/tree/chg-rait-pl/perl/Amanda/Changer/rait.pm
> (and, if you're willing to build Amanda from source, you may be able
> to use this to fix your chg-multi problems).
>
> I woul suggest looking into the source of chg-multi to see where it
> puts its debugging information, and potentially adding more debugging
> echo's as necessary to see what it's doing. Also look at the taper
> debug log to see what commands it's sending to the changer. You may
> be able to run simpler operations using 'amtape', rather than running
> an entire amdump just to replicate the failure.
>
> Good luck, happy hunting, and, if you've got some Perl-fu, please
> consider coding up a new-style Amanda::Changer::multi!
>
So, it appears as though we have broken existing functionality, at some
point in these changes. Let em describe what we have that is working, but
no longer works, and perhaps you can enlighten me as to how this
configuration _should_ work in the new system.
We have a RAIT. It consists of 25 vtapes, on one side, and an LTO
tape drive (not a changer) on the other side. Actually we have 3
configurations, the RAIT is used for backups, and we have 2 other
configurations for restore. One of them uses the vtape side of the RAIT, and
the other uses the tapes. The idea is that the first recourse for restoring
would be from the vtape, because it is faster, but we can fall back to
tape, if necessary. We would never attempt a restore from the RAIT for
obvious reasons.
There is a main amanda.conf, and a very few things are overridden in it by
the specific configs. For the RAIT, the overrides are as follows:
includefile "/opt/amanda/etc/amanda.conf.main"
tpchanger "chg-multi"
changerfile "/opt/amanda/etc/DailyDump/chg-multi.conf"
tapetype RAIT
tapedev "{file:/vtapes/DailyDump,tape:/dev/nst0}"
The settings in chg-multi.conf look like this:
multieject 0
gravity 0
needeject 0
ejectdelay 0
statefile /opt/amanda/etc/DailyDump/changer-status
firstslot 1
lastslot 25
slot 1 rait:{file:/vtapes/DailyDump/vtape1,tape:/dev/nst0}
slot 2 rait:{file:/vtapes/DailyDump/vtape2,tape:/dev/nst0}
slot 3 rait:{file:/vtapes/DailyDump/vtape3,tape:/dev/nst0}
slot 4 rait:{file:/vtapes/DailyDump/vtape4,tape:/dev/nst0}
slot 5 rait:{file:/vtapes/DailyDump/vtape5,tape:/dev/nst0}
slot 6 rait:{file:/vtapes/DailyDump/vtape6,tape:/dev/nst0}
slot 7 rait:{file:/vtapes/DailyDump/vtape7,tape:/dev/nst0}
slot 8 rait:{file:/vtapes/DailyDump/vtape8,tape:/dev/nst0}
slot 9 rait:{file:/vtapes/DailyDump/vtape9,tape:/dev/nst0}
slot 10 rait:{file:/vtapes/DailyDump/vtape10,tape:/dev/nst0}
slot 11 rait:{file:/vtapes/DailyDump/vtape11,tape:/dev/nst0}
slot 12 rait:{file:/vtapes/DailyDump/vtape12,tape:/dev/nst0}
slot 13 rait:{file:/vtapes/DailyDump/vtape13,tape:/dev/nst0}
slot 14 rait:{file:/vtapes/DailyDump/vtape14,tape:/dev/nst0}
slot 15 rait:{file:/vtapes/DailyDump/vtape15,tape:/dev/nst0}
slot 16 rait:{file:/vtapes/DailyDump/vtape16,tape:/dev/nst0}
slot 17 rait:{file:/vtapes/DailyDump/vtape17,tape:/dev/nst0}
slot 18 rait:{file:/vtapes/DailyDump/vtape18,tape:/dev/nst0}
slot 19 rait:{file:/vtapes/DailyDump/vtape19,tape:/dev/nst0}
slot 20 rait:{file:/vtapes/DailyDump/vtape20,tape:/dev/nst0}
slot 21 rait:{file:/vtapes/DailyDump/vtape21,tape:/dev/nst0}
slot 22 rait:{file:/vtapes/DailyDump/vtape22,tape:/dev/nst0}
slot 23 rait:{file:/vtapes/DailyDump/vtape23,tape:/dev/nst0}
slot 24 rait:{file:/vtapes/DailyDump/vtape24,tape:/dev/nst0}
slot 25 rait:{file:/vtapes/DailyDump/vtape25,tape:/dev/nst0}
slot 24 rait:{file:/vtapes/DailyDu
Re: changer problems with 2.6.1
On Wed, Feb 18, 2009 at 10:22 AM, stan wrote: >> Interesting -- did you submit patches? > > No, we never had the time to figure out what problem had been introduced. > But, I suspose we must do this now, as we cannot make 2/`.6.1 work at all, > the old script does not work with it. I'm not surprised :( >> I'm curious - what do you need mt for? > > I thought that chg-multi used it, no? Nope (see below). >> > echo `_ '$var'` >> $logfile >> > >> > I (nor my sh interperter) understand these. Looks to me like we are trying >> > to execute something called + which does not exist. >> >> It's a rather ill-advised attempt at internationalization. It's not >> pretty, especially in shell. > > It's a syntax erro in any Bourne relate shell I am aware of the back ticks > say run this command, and substiture the output of it into the coammand > line. Thus we are trying to excute a command called _ which does not exist. > I have checked my bash, and korn shell books, and niether has any reference > to a construct like this. What is it "supposed" to do? It's a function defined in chg-lib.sh, which is sourced at the top of chg-multi. It's the shell equivalent of _(), which is (loosely put) the "translate this string" function used for internationalization. >> 51 if ! error=try_find_mt; then >> 52 echo $error >> 53 exit 2 >> 54 fi >> >> but the result ($MT) is never used. I'd rip that out if I weren't >> about to delete the whole file :) > > I had looked at that. Sorry to have distracted you :) > I'll comment on this ina seperate message. Sounds good. Dustin -- Storage Software Engineer http://www.zmanda.com
Re: changer problems with 2.6.1
On Tue, Feb 17, 2009 at 11:33:12PM -0500, Dustin J. Mitchell wrote: > On Tue, Feb 17, 2009 at 11:06 PM, stan wrote: > > When we upgraded from 2.5.1p2 to 2.5.2p1, we had to copy the chg-multi > > script > > from the 2.5.1p1 release into the new tree, as the newer script did not > > work correctly. > > Interesting -- did you submit patches? No, we never had the time to figure out what problem had been introduced. But, I suspose we must do this now, as we cannot make 2/`.6.1 work at all, the old script does not work with it. > > > 2.5.1p2 chg-multi does not work with 2.6.1. We have observed a couple of > > things. First it seems that the tape handler that we were using, ammt from > > the earlier amanda release is depricated. Since this mahcine is a Linux > > machine we are back to the bad old days of trying to figure out wheter to > > use mt-st or > > gnu-mt, and I honestly cannot remember which one of those used to work. > > I'm curious - what do you need mt for? I thought that chg-multi used it, no? > > > In addition, we seem to be having problems with some syntax in chg-multi. > > There are lines that look like this: > > > > echo `_ '$var'` >> $logfile > > > > I (nor my sh interperter) understand these. Looks to me like we are trying > > to execute something called + which does not exist. > > It's a rather ill-advised attempt at internationalization. It's not > pretty, especially in shell. It's a syntax erro in any Bourne relate shell I am aware of the back ticks say run this command, and substiture the output of it into the coammand line. Thus we are trying to excute a command called _ which does not exist. I have checked my bash, and korn shell books, and niether has any reference to a construct like this. What is it "supposed" to do? > > > So since we are no longer going to be able to use ammt, what commands can > > we expect mt to be involed with & what are the expected returns. Cause both > > versions of the Linux mt commands (gnu-mt & mt-st) available for Ubuntu > > return something different for an, mt -f /dev/nst0 status than ammt -f > > /dev/nst0 status (none of the three match results). > > Amanda does not use mt at all at this point. I do notice that > chg-multi still begins with: > > 51 if ! error=try_find_mt; then > 52 echo $error > 53 exit 2 > 54 fi > > but the result ($MT) is never used. I'd rip that out if I weren't > about to delete the whole file :) I had looked at that. > > > We got past the syntax > > problem after getting rid of all the '_ ' in chg-multi & then amcheck which > > calls the changer script fails to be able to find a valid slot. We suspect > > that this is because chg-multi expects different behavior from mt than > > it used to get from ammt. We tried forcing the existing, working 2.5.2p1, > > that normally uses chg-multi from 2.5.1p2 and ammt to not find ammt & so > > use the system mt. It reproduces the same error just described (confirmed > > suspicion that Ubuntu gnu-mt & mt-st are incompatible with the changer > > code) > > OK, a few pieces of Amanda history may be relevant to your investigations > here. > I'll comment on this ina seperate message. -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems with 2.6.1
On Tue, Feb 17, 2009 at 11:06 PM, stan wrote: > When we upgraded from 2.5.1p2 to 2.5.2p1, we had to copy the chg-multi script > from the 2.5.1p1 release into the new tree, as the newer script did not > work correctly. Interesting -- did you submit patches? > 2.5.1p2 chg-multi does not work with 2.6.1. We have observed a couple of > things. First it seems that the tape handler that we were using, ammt from > the earlier amanda release is depricated. Since this mahcine is a Linux > machine we are back to the bad old days of trying to figure out wheter to use > mt-st or > gnu-mt, and I honestly cannot remember which one of those used to work. I'm curious - what do you need mt for? > In addition, we seem to be having problems with some syntax in chg-multi. > There are lines that look like this: > > echo `_ '$var'` >> $logfile > > I (nor my sh interperter) understand these. Looks to me like we are trying > to execute something called + which does not exist. It's a rather ill-advised attempt at internationalization. It's not pretty, especially in shell. > So since we are no longer going to be able to use ammt, what commands can > we expect mt to be involed with & what are the expected returns. Cause both > versions of the Linux mt commands (gnu-mt & mt-st) available for Ubuntu > return something different for an, mt -f /dev/nst0 status than ammt -f > /dev/nst0 status (none of the three match results). Amanda does not use mt at all at this point. I do notice that chg-multi still begins with: 51 if ! error=try_find_mt; then 52 echo $error 53 exit 2 54 fi but the result ($MT) is never used. I'd rip that out if I weren't about to delete the whole file :) > We got past the syntax > problem after getting rid of all the '_ ' in chg-multi & then amcheck which > calls the changer script fails to be able to find a valid slot. We suspect > that this is because chg-multi expects different behavior from mt than > it used to get from ammt. We tried forcing the existing, working 2.5.2p1, > that normally uses chg-multi from 2.5.1p2 and ammt to not find ammt & so > use the system mt. It reproduces the same error just described (confirmed > suspicion that Ubuntu gnu-mt & mt-st are incompatible with the changer > code) OK, a few pieces of Amanda history may be relevant to your investigations here. First, Amanda no longer assumes it is using a tape drive (that is, something which 'mt' might be able to operate). This is a part of the Device API which was introduced in 2.6.0. The only point at which changer scripts need to interact with a device is to determine whether it's ready or not -- in chg-zd-mtx and chg-manual. For this, these scripts now use amdevcheck. This has brought its own set of issues, but is fully functional and does not involve 'mt' in any way. Second, there are a number of other changes to the changer API in 2.6.1 -- see amanda-changers(7) and http://wiki.zmanda.com/index.php/Changer_API for a taste. Among lots of other interesting action, we're deprecating and removing all of the old shell scripts. At the moment, the new Changer API is *always* used, via a C-to-perl (and old-to-new) wrapper script (changer-src/chg-glue.pl[1]) and a perl-to-shell (and new-to-old) wrapper script (perl/Amanda/Changer/compat.pm). There's room for problems at several points along this chain. I'm busily at work rewriting changers into the new API, and rewriting the Amanda taper in Perl so that it can utilize the new API directly. You can see the new Amanda::Changer::rait at http://github.com/djmitche/amanda/tree/chg-rait-pl/perl/Amanda/Changer/rait.pm (and, if you're willing to build Amanda from source, you may be able to use this to fix your chg-multi problems). I woul suggest looking into the source of chg-multi to see where it puts its debugging information, and potentially adding more debugging echo's as necessary to see what it's doing. Also look at the taper debug log to see what commands it's sending to the changer. You may be able to run simpler operations using 'amtape', rather than running an entire amdump just to replicate the failure. Good luck, happy hunting, and, if you've got some Perl-fu, please consider coding up a new-style Amanda::Changer::multi! Dustin [1] You can find the source at http://github.com/nikolasco/amanda/tree/ -- Storage Software Engineer http://www.zmanda.com
changer problems with 2.6.1
We are trying to upgrade all of our machines to 2.6.1. Presently we are struggling with tape changer problems on the server, an Ubuntu 8.04 machine. A bit of history here. We have a somewaht different arrangement. We have a RAIT, one side of which consists of 25 vtapes, the other side of which consists of a single LTO drive, not a changer, just a plain single tape drive, not a changer. We have had this working for several years now. When we upgraded from 2.5.1p2 to 2.5.2p1, we had to copy the chg-multi script from the 2.5.1p1 release into the new tree, as the newer script did not work correctly. 2.5.1p2 chg-multi does not work with 2.6.1. We have observed a couple of things. First it seems that the tape handler that we were using, ammt from the earlier amanda release is depricated. Since this mahcine is a Linux machine we are back to the bad old days of trying to figure out wheter to use mt-st or gnu-mt, and I honestly cannot remember which one of those used to work. In addition, we seem to be having problems with some syntax in chg-multi. There are lines that look like this: echo `_ '$var'` >> $logfile I (nor my sh interperter) understand these. Looks to me like we are trying to execute something called + which does not exist. So since we are no longer going to be able to use ammt, what commands can we expect mt to be involed with & what are the expected returns. Cause both versions of the Linux mt commands (gnu-mt & mt-st) available for Ubuntu return something different for an, mt -f /dev/nst0 status than ammt -f /dev/nst0 status (none of the three match results). We got past the syntax problem after getting rid of all the '_ ' in chg-multi & then amcheck which calls the changer script fails to be able to find a valid slot. We suspect that this is because chg-multi expects different behavior from mt than it used to get from ammt. We tried forcing the existing, working 2.5.2p1, that normally uses chg-multi from 2.5.1p2 and ammt to not find ammt & so use the system mt. It reproduces the same error just described (confirmed suspicion that Ubuntu gnu-mt & mt-st are incompatible with the changer code) -- One of the main causes of the fall of the roman empire was that, lacking zero, they had no way to indicate successful termination of their C programs.
Re: changer problems when running from cron
Hi,
John R. Jackson schrieb:
Let me ask again ...
What does your version of the request() function look like (the actual
code, not a general description)?
Here's the code, but I found that this function will never invoked,
chg-manual ends before invoking the request() function:
The first function is info(), I added here a sleep(2) and then this part
works.
The next function is loadslot(), the script exits in this function
around line 312:
if [ ! -z "$used" -a $load = $slot ];then
# already loaded
answer="$slot $tape"
echo "Exit -> $answer" >> $logfile
echo "$answer"
exit 0
fi
#!/bin/sh
export home=/
#export DD="/bin/dd bs=32k"
export ONLINEREGEX="61000|ONLINE|READY|sense[_ ]key[(]0x0[)]|sense key
error = 0|^er=0$"
request() {
echo -n "Request start:" >> $logfile
echo `date` >> $logfile
timeout=0
while true; do
label=`grep -v no-reuse
/usr/local/etc/amanda/gamed/tapelist | tail -1 | \
while read line
do
set $line
echo $2
done`
if [ $timeout -le 0 ]; then
echo "Bitte Band $label einlegen" \
| mail -s "$ORG AMANDA TAPE MOUNT REQUEST FOR
SLOT $1" \
$REPORTTO 2>/dev/null
timeout=`expr 360 \* 60`
fi
echo " -> rewind $tape" >> $logfile
$MT $MTF $tape rewind >> $logfile 2>&1
sleep 5
echo " -> status $tape" >> $logfile
used=`$MT $MTF $tape status 2>&1 | tee -a $logfile |
$EGREP "$ONLINEREGEX"`
echo " -> loaded <$used>" >> $logfile
echo "SEPP3" >>/tmp/changer.debug
if [ ! -z "$used" ];then
break
fi
sleep 60
timeout=`expr $timeout - 60`
done
}
What is in /tmp/amanda/changer.debug?
MT -> /usr/local/sbin/ammt -f
DD -> /usr/local/sbin/amdd
Args -> -info
-> rewind /dev/nst0
-> status /dev/nst0
/dev/nst0 status: ONLINE BOT
-> loaded
Exit -> 42 99 1
MT -> /usr/local/sbin/ammt -f
DD -> /usr/local/sbin/amdd
Args -> -slot current
-> rewind /dev/nst0
-> status /dev/nst0
/dev/nst0 status: ONLINE BOT
-> loaded
Exit -> 42 /dev/nst0
While this is the most important thing :-), it would be good if we
could figure out why the script is failing. And Christoph brings up a
very good point about the cron PATH possibly being different than your
interactive PATH.
this was also my first guess, but all the environment vars are correct set.
bye,
--
--- --
Michael Bretterklieber - [EMAIL PROTECTED]
JAWA Management Software GmbH - http://www.jawa.at
Liebenauer Hauptstr. 200-- privat
A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712
Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED]
Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com
--- --
"...the number of UNIX installations has grown to 10, with more
expected..." - Dennis Ritchie and Ken Thompson, June 1972
Re: changer problems when running from cron
>>>we wrote our own request() function. >> >> >> And what's in your version of the function? My guess is it is not >> waiting long enough for the tape to go ready. >> >> What's in /tmp/amanda/changer.debug? >AFAIK, if I'm invoking amcheck, then this programm starts the >changer-script and waits with waitpid() for the end of the changer >script, but the changer-script ends to quick and then waitpid() returns >an error. ... I agree. Which is why I asked two specific questions to help diagnose exactly what the script is doing wrong. Let me ask again ... What does your version of the request() function look like (the actual code, not a general description)? What is in /tmp/amanda/changer.debug? >however, I disabled the changer-mechanism and now the backup works. While this is the most important thing :-), it would be good if we could figure out why the script is failing. And Christoph brings up a very good point about the cron PATH possibly being different than your interactive PATH. >Michael Bretterklieber John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
Re: changer problems when running from cron
Hi, First guess: double check your path statements in cron. If programs/scripts work from commandline but not from cron most times a needed part of the $PATH env-variable is missing, so a part of your script/command can't be executed. And this is not necesarryly reported back to you. Christoph Michael Bretterklieber schrieb: Hi, John R. Jackson schrieb: we wrote our own request() function. And what's in your version of the function? My guess is it is not waiting long enough for the tape to go ready. What's in /tmp/amanda/changer.debug? AFAIK, if I'm invoking amcheck, then this programm starts the changer-script and waits with waitpid() for the end of the changer script, but the changer-script ends to quick and then waitpid() returns an error. This must be a timing problem, as a said if I add a sleep i the changer-script then it works better, but not completely. The weird thing was that when I'm invoking the amcheck (or amdump) from the commandline, then it works perfect, only when starting from cron this doesen't work. We have the same scripts on other machines running for more then a year with no problems, but on this machine it doesen't work. We use the changer scripts for avoiding that amanda puts the dump on the holding disk if no tape was inserted, instead using our scripts amanda, waits for the right tape and after the right tape was inserted amanda starts with the backup. however, I disabled the changer-mechanism and now the backup works. thanx, bye,
Re: changer problems when running from cron
Hi, John R. Jackson schrieb: we wrote our own request() function. And what's in your version of the function? My guess is it is not waiting long enough for the tape to go ready. What's in /tmp/amanda/changer.debug? AFAIK, if I'm invoking amcheck, then this programm starts the changer-script and waits with waitpid() for the end of the changer script, but the changer-script ends to quick and then waitpid() returns an error. This must be a timing problem, as a said if I add a sleep i the changer-script then it works better, but not completely. The weird thing was that when I'm invoking the amcheck (or amdump) from the commandline, then it works perfect, only when starting from cron this doesen't work. We have the same scripts on other machines running for more then a year with no problems, but on this machine it doesen't work. We use the changer scripts for avoiding that amanda puts the dump on the holding disk if no tape was inserted, instead using our scripts amanda, waits for the right tape and after the right tape was inserted amanda starts with the backup. however, I disabled the changer-mechanism and now the backup works. thanx, bye, -- --- -- Michael Bretterklieber - [EMAIL PROTECTED] JAWA Management Software GmbH - http://www.jawa.at Liebenauer Hauptstr. 200-- privat A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712 Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED] Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com --- -- "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972
Re: changer problems when running from cron
>we wrote our own request() function. And what's in your version of the function? My guess is it is not waiting long enough for the tape to go ready. What's in /tmp/amanda/changer.debug? And I (very respectfully) disagree with Joshua about the need for a changer with runtapes == 1. Our changer is integrated with our Operations systems so they get the tape mount request in a more "pro-active" manner than E-mail (etc). This is completely independent of the number of tapes involved. >Michael Bretterklieber John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
Re: changer problems when running from cron
Hi, Joshua Baker-LePain schrieb: On Fri, 7 Feb 2003 at 2:09pm, Michael Bretterklieber wrote If you have runtapes = 1, then why do you need a changer? for verifying that the right tape with the right label is in the tapedrive and to be able to send an email with the name of the requested tape to the user who is responsible for inserting the tapes. The next tape requested is at the top of the amreport email. Also, amcheck can do this for you. Before the end of the workday, have cron run 'amcheck -m '. If the correct tape isn't inserted (or there are any other problems), amcheck will email the addresses specified in amanda.conf, letting them know this and letting them know which tape it wants. aha. makes sense. I changed my config, lets see if the backup works this night. thanx, bye, -- --- -- Michael Bretterklieber - [EMAIL PROTECTED] JAWA Management Software GmbH - http://www.jawa.at Liebenauer Hauptstr. 200-- privat A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712 Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED] Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com --- -- "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972
Re: changer problems when running from cron
On Fri, 7 Feb 2003 at 2:09pm, Michael Bretterklieber wrote > > If you have runtapes = 1, then why do you need a changer? > > for verifying that the right tape with the right label is in the > tapedrive and to be able to send an email with the name of the requested > tape > to the user who is responsible for inserting the tapes. The next tape requested is at the top of the amreport email. Also, amcheck can do this for you. Before the end of the workday, have cron run 'amcheck -m '. If the correct tape isn't inserted (or there are any other problems), amcheck will email the addresses specified in amanda.conf, letting them know this and letting them know which tape it wants. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: changer problems when running from cron
Hi, Joshua Baker-LePain schrieb: On Fri, 7 Feb 2003 at 1:58pm, Michael Bretterklieber wrote Do you really need runtapes > 1 No. I have runtapes = 1 in my amanda.conf: runtapes 1 # number of tapes to be used in a single run of amdump tpchanger "chg-manual" # the tape-changer glue script tapedev "/dev/nst0" # the no-rewind tape device to be used rawtapedev "/dev/nst0" # the raw device to be used (ftape only) changerfile "/usr/local/etc/amanda/gamed/changer" changerdev "/dev/null" If you have runtapes = 1, then why do you need a changer? for verifying that the right tape with the right label is in the tapedrive and to be able to send an email with the name of the requested tape to the user who is responsible for inserting the tapes. bye, -- --- -- Michael Bretterklieber - [EMAIL PROTECTED] JAWA Management Software GmbH - http://www.jawa.at Liebenauer Hauptstr. 200-- privat A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712 Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED] Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com --- -- "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972
Re: changer problems when running from cron
On Fri, 7 Feb 2003 at 1:58pm, Michael Bretterklieber wrote > > Do you really need runtapes > 1 > > No. I have runtapes = 1 in my amanda.conf: > > runtapes 1 # number of tapes to be used in a single run of > amdump > tpchanger "chg-manual" # the tape-changer glue script > tapedev "/dev/nst0" # the no-rewind tape device to be used > rawtapedev "/dev/nst0" # the raw device to be used (ftape only) > changerfile "/usr/local/etc/amanda/gamed/changer" > changerdev "/dev/null" If you have runtapes = 1, then why do you need a changer? -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: changer problems when running from cron
Hi, Joshua Baker-LePain schrieb: On Fri, 7 Feb 2003 at 11:31am, Michael Bretterklieber wrote we have problems with amanda when running from cron, amanda always means that the right tape is not or no tape is inserted but it is. I tried to track down the problem and I can reproduce this problem with amcheck, when running amcheck from cron: amcheck-server: could not get changer info: wait for "/usr/local/libexec/chg-manual" failed: No child processes chg-manual is meant to be used when you run amdump interactively. It waits for you to put the right tape in and then let it know. This won't work when running from cron. we wrote our own request() function. The weird thing is that the same configuration runs on other systems (FreeBSD-systems) without problems. Do you really need runtapes > 1 No. I have runtapes = 1 in my amanda.conf: runtapes 1 # number of tapes to be used in a single run of amdump tpchanger "chg-manual" # the tape-changer glue script tapedev "/dev/nst0" # the no-rewind tape device to be used rawtapedev "/dev/nst0" # the raw device to be used (ftape only) changerfile "/usr/local/etc/amanda/gamed/changer" changerdev "/dev/null" bye, -- --- -- Michael Bretterklieber - [EMAIL PROTECTED] JAWA Management Software GmbH - http://www.jawa.at Liebenauer Hauptstr. 200-- privat A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712 Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED] Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com --- -- "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972
Re: changer problems when running from cron
On Fri, 7 Feb 2003 at 11:31am, Michael Bretterklieber wrote > we have problems with amanda when running from cron, amanda always means > that the right tape is not or no tape is inserted but it is. > > I tried to track down the problem and I can reproduce this problem with > amcheck, when running amcheck from cron: > > amcheck-server: could not get changer info: wait for > "/usr/local/libexec/chg-manual" failed: No child processes chg-manual is meant to be used when you run amdump interactively. It waits for you to put the right tape in and then let it know. This won't work when running from cron. Do you really need runtapes > 1 -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
changer problems when running from cron
Hi, we have problems with amanda when running from cron, amanda always means that the right tape is not or no tape is inserted but it is. I tried to track down the problem and I can reproduce this problem with amcheck, when running amcheck from cron: amcheck-server: could not get changer info: wait for "/usr/local/libexec/chg-manual" failed: No child processes Amanda Tape Server Host Check - (brought to you by Amanda 2.4.3) It seems that the chg-manual script ends to quick, when adding a "sleep 1" into the info() function of the chg-manual script then the error above doesen't occur but this error occurs. Amanda Tape Server Host Check - amcheck-server: fatal slot : wait for "/usr/local/libexec/chg-manual" failed: No child processes ERROR: label gamed02003 or new tape not found in rack (expecting tape gamed02003 or a new tape) NOTE: skipping tape-writable test Server check took 2.659 seconds (brought to you by Amanda 2.4.3) Doing the same sleep in loadslot() has no effect. NOTE: everything works OK when invoking from the shell. We are running Caldera Openlinux, Kernel 2.4.13, with amanda 2.4.3. It is a real problem for us, because we must invoke amflush every day by hand. bye, -- --- -- Michael Bretterklieber - [EMAIL PROTECTED] JAWA Management Software GmbH - http://www.jawa.at Liebenauer Hauptstr. 200-- privat A-8041 GRAZ GSM: ++43-(0)676-84 03 15 712 Tel: ++43-(0)316-403274-12 E-mail: [EMAIL PROTECTED] Fax: ++43-(0)316-403274-10 http://www.bretterklieber.com --- -- "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972
Amanda 2.4.3b4 chio changer problems
Since upgrading from 2.4.2p1 to 2.4.3b3 and 2.4.3b4 I get this problem. The changer does do the change, but I guess the script doesn't wait to check. So as long as I know that I'm not on the last tape, I ignore it. But I'd rather it work. You know, I'm wondering if maybe this wasn't an OS upgrade. I'm running FreeBSD 4.7-Stable. But this was happening in 4.6-Stable for a while too. Any ideas? Paul. Amanda Tape Server Host Check - Holding disk /amanda/hold: 4659412 KB disk space available, that's plenty amcheck-server: slot 1: date 20021029 label IACES22 (active tape) amcheck-server: fatal slot chio:: /dev/ch0: CHIOMOVE: Invalid argument ERROR: label IACES23 or new tape not found in rack (expecting tape IACES23 or a new tape) NOTE: skipping tape-writable test Server check took 72.793 seconds Amanda Backup Client Hosts Check Client check: 4 hosts checked in 1.587 seconds, 0 problems found (brought to you by Amanda 2.4.3b4)
RE: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
> Message: 16
>Date: Sun, 24 Mar 2002 02:22:47 -0500
>From: Derrick Miller <[EMAIL PROTECTED]>
> Subject: ADIC / Dell Powervault 120T changer problems under
> Linux (Pt 2)
>
> 1) Does anyone have this changer working with Amanda? I
> would give just
> about anything to get my hands on copies of working config
> and chg-zd-mtx
> files.
I am using a different changer script supplied by Chris Pascoe and once
upon a time posted to this list (in a slightly earlier version). I have
attached my current copy (I hope you don't mind, Chris). Beware of line
endings - I read email on a windows box, even though most of our servers
are unix/linux.
With this tape drive running alone on the SCSI bus from an adaptec 2944,
amverify gets occasional "Unexpected busfree in Data-in phase". It would
be interesting to know if anyone else with this hardware combination
gets this error when running amverify or not. If you have not bothered
with amverify, I would urge you to try it, lest your tapes are
unreadable when you need them most.
#!/bin/sh
#
# Exit Status:
# 0 Alles Ok
# 1 Illegal Request
# 2 Fatal Error
#
# Contributed by Eric DOUTRELEAU <[EMAIL PROTECTED]>
# This is supposed to work with Zubkoff/Dandelion version of mtx
#
# Modified by Joe Rhett <[EMAIL PROTECTED]>
# to work with MTX 1.2.9 by Eric Lee Green http://mtx.sourceforge.net
#
# Modified by Chris Pascoe <[EMAIL PROTECTED]> to work with a
# Dell PowerVault 120T, MTX 1.2.10 and add some intelligence to
# exit when all slots are empty. (Jan 2001). Fixed introduced
# bugs when going backwards and potential problem with changers having
# more than 9 slots (July 2001).
#
# You may need to customize these things
MT=/bin/mt
MTF=-f
MTX=/sbin/mtx
DD=/bin/dd
cleanslot=-1
accessbeforeclean=119
ONLINEREGEX="ONLINE|READY|sense[_ ]key[(]0x0[)]|sense key error = 0|^er=0$"
# No user-level customized required beyond this point.
# try to hit all the possibilities here
prefix=/usr/local/amanda/2.4.2p1
exec_prefix=${prefix}
sbindir=${exec_prefix}/sbin
libexecdir=${exec_prefix}/libexec
PATH=$sbindir:$libexecdir:/usr/bin:/bin:/usr/sbin:/sbin:/opt/local/bin
export PATH
if [ -d "/tmp/amanda" ]; then
DBGFILE=/tmp/amanda/changer.debug
else
DBGFILE=/dev/null
fi
if [ ! -f "amanda.conf" ]; then
echo "No amanda config in current directory"
exit 2;
fi
USE_VERSION_SUFFIXES="no"
if test "$USE_VERSION_SUFFIXES" = "yes"; then
SUF="-2.4.2p1"
else
SUF=
fi
myname=$0
tape=`amgetconf$SUF tapedev`
TAPE=`amgetconf$SUF changerdev`; export TAPE # for mtx command
if [ "$tape" = "/dev/null" -o "$TAPE" = "/dev/null" ]; then
echo "Both tapedev and changerdev must be specified in config file";
exit 2;
fi
changerfile=`amgetconf$SUF changerfile`
if [ -z "$changerfile" ];
then
echo "changerfile not specified in amanda config?"
exit 2
fi
cleanfile=$changerfile-clean
accessfile=$changerfile-access
slotfile=$changerfile-slot
[ ! -f $cleanfile ] && echo 0 > $cleanfile
[ ! -f $accessfile ] && echo 0 > $accessfile
[ ! -f $slotfile ] && echo 0 > $slotfile
cleancount=`cat $cleanfile`
accesscount=`cat $accessfile`
# Routines start here
readstatus() {
usedslot=`$MTX status |
sed -n 's/Data Transfer Element 0:Empty/-1/p;s/Data Transfer Element
0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p'`
if [ "$usedslot" -eq "-1" ]; then
echo '-1' > $slotfile
fi
echo "STATUS -> currently loaded slot = $usedslot" >> $DBGFILE
}
getslots() {
slotlist=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):\(Empty\|Full\)/\1/p'`
fullslots=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):Full/\1/p;s/Data Transfer
Element 0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p'`
dataslots=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):Full/\1/p;s/Data Transfer
Element 0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p' |
fgrep -v "^${cleanslot}$" | sort -n`
numdataslots=`echo $dataslots | wc -w | awk '{print $1}'`
if [ -z "$dataslots" ]; then
dataslots="-1"
fi
firstslot=`echo $slotlist | awk '{print $1}'`
lastslot=`echo $slotlist | awk '{print $NF}'`
firstdataslot=`echo $dataslots | awk '{print $1}'`
lastdataslot=`echo $dataslots | awk '{print $NF}'`
nextdataslot=`ech
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
On Sun, 24 Mar 2002, Derrick Miller wrote: > So, my questions: > > 1) Does anyone have this changer working with Amanda? I would give just > about anything to get my hands on copies of working config and chg-zd-mtx > files. For what it may be worth, Derrick, I have a Sun StorEdge jukebox attached to an Enterprise class Solaris system. While mtx itself worked, I never got the chg-zd-mtx script to work. I had to do things like hard-code variables just to even get it to run but I never got it to perform any significant functions. I inhereted the installation and found out that chg-scsi was never tried, but lo and behold, I spent a little time tweaking it, it worked well, and have used it ever since. It really is a great little program. The only functionality that does not work for me right now is the barcode stuff but Thomas Hepper has been working with me to make that work. BTW, I use 2.4.3b2. The StorEdge is a 30-slot jukebox with two DLT IV tape drives. > Any pointers whatsoever would be greatly appreciated, as I've been beating my > head against this for over a month and have made very little progress. Good luck. If I hadn't discovered chg-scsi would do the job, I would have tried writing my own perl script. Lots of people use chg-zd-mtx, though, which always made me feel worse! :^P - - - -- Eric Trager CMAG, DCB, CIT, NIH [EMAIL PROTECTED]
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
I'm using the RedHat RPMS without any problems so far. Jim On Fri, Mar 29, 2002 at 03:31:50PM -0500, Derrick Miller wrote: ... > Are you using the Amanda RPMs supplied with Red Hat 7.2, or did you compile > 2.4.2p2 from source?
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
* Derrick Miller ([EMAIL PROTECTED]) [020329 12:49]: > Jim, > > > I have this changer working with RedHat 7.2 using the 2.4.2p2 chg-zd-mtx > > script. The trick to getting it working is getting the configuration file > > for chg-zd-mtx right. Make sure all the variables mentioned in the script > > are defined. In my case I have (in /etc/amanda/CONFIG/chg-zd-mtx.conf, > > replace CONFIG with your configuration name): > > Are you using the Amanda RPMs supplied with Red Hat 7.2, or did you compile > 2.4.2p2 from source? Since I posted my previous message asking for help, I > have managed to get Amanda working well with the changer, by compiling from > source. I would like to switch back to the RPM version if it will work, > though, since I find it's much easier to manage the system using RPM and > up2date than it is with source tarballs. As a datapoint, I am happily using the RedHat 7.2 RPMs with an HP DLT changer. Best, -- Patrick Michael Kane <[EMAIL PROTECTED]>
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
Jim, > I have this changer working with RedHat 7.2 using the 2.4.2p2 chg-zd-mtx > script. The trick to getting it working is getting the configuration file > for chg-zd-mtx right. Make sure all the variables mentioned in the script > are defined. In my case I have (in /etc/amanda/CONFIG/chg-zd-mtx.conf, > replace CONFIG with your configuration name): Are you using the Amanda RPMs supplied with Red Hat 7.2, or did you compile 2.4.2p2 from source? Since I posted my previous message asking for help, I have managed to get Amanda working well with the changer, by compiling from source. I would like to switch back to the RPM version if it will work, though, since I find it's much easier to manage the system using RPM and up2date than it is with source tarballs. Thanks for your help, Derrick Miller
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
Derrick Miller writes: > > 1) Does anyone have this changer working with Amanda? I > > would give just about anything to get my hands on copies > > of working config and chg-zd-mtx files. > I am using a different changer script supplied by Chris Pascoe and once > upon a time posted to this list (in a slightly earlier version). I strongly recommend using the chg-zd-mtx from the current CVS now as recommended by John R. Jackson. I dropped it in place on Feb 17, and haven't had any problems since on my amanda-2.4.2p2 installation. In my amanda.conf, I have: tpchanger "/opt/local/stow/amanda-2.4.3-20020217/libexec/chg-zd-mtx" tapedev "/dev/nst0" changerfile "changer" changerdev "/dev/changer" (where /dev/changer is a symlink to the changer's sg device) In my amanda config directory, I have a changer.conf with the following values: firstslot=1 lastslot=7 cleanslot=-1 autoclean=0 autocleancount=99 driveslot=0 offline_before_unload=1 unloadpause=3 poll_drive_ready=5 max_drive_wait=300 > With this tape drive running alone on the SCSI bus from an adaptec 2944, > amverify gets occasional "Unexpected busfree in Data-in phase". I'm not seeing these errors (and never have) on my verify runs, with the configuration you describe. I suggest checking all your cabling and the HVD terminator.. Chris
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
>... While mtx itself worked, I never >got the chg-zd-mtx script to work. ... >BTW, I use 2.4.3b2. ... Was that the 2.4.3 chg-zd-mtx that you had trouble with? I did a pretty significant overhaul for this release and if it doesn't work, now's the time to tell me. If it was 2.4.2, that's a different story. >Eric Trager John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
On Sun, 24 Mar 2002, Derrick Miller wrote: > So, my questions: > > 1) Does anyone have this changer working with Amanda? I would give just > about anything to get my hands on copies of working config and chg-zd-mtx > files. For what it may be worth, Derrick, I have a Sun StorEdge jukebox attached to an Enterprise class Solaris system. While mtx itself worked, I never got the chg-zd-mtx script to work. I had to do things like hard-code variables just to even get it to run but I never got it to perform any significant functions. I inhereted the installation and found out that chg-scsi was never tried, but lo and behold, I spent a little time tweaking it, it worked well, and have used it ever since. It really is a great little program. The only functionality that does not work for me right now is the barcode stuff but Thomas Hepper has been working with me to make that work. BTW, I use 2.4.3b2. The StorEdge is a 30-slot jukebox with two DLT IV tape drives. > Any pointers whatsoever would be greatly appreciated, as I've been beating my > head against this for over a month and have made very little progress. Good luck. If I hadn't discovered chg-scsi would do the job, I would have tried writing my own perl script. Lots of people use chg-zd-mtx, though, which always made me feel worse! :^P - - - -- Eric Trager CMAG, DCB, CIT, NIH [EMAIL PROTECTED]
RE: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
> Message: 16
>Date: Sun, 24 Mar 2002 02:22:47 -0500
>From: Derrick Miller <[EMAIL PROTECTED]>
> Subject: ADIC / Dell Powervault 120T changer problems under
> Linux (Pt 2)
>
> 1) Does anyone have this changer working with Amanda? I
> would give just
> about anything to get my hands on copies of working config
> and chg-zd-mtx
> files.
I am using a different changer script supplied by Chris Pascoe and once
upon a time posted to this list (in a slightly earlier version). I have
attached my current copy (I hope you don't mind, Chris). Beware of line
endings - I read email on a windows box, even though most of our servers
are unix/linux.
With this tape drive running alone on the SCSI bus from an adaptec 2944,
amverify gets occasional "Unexpected busfree in Data-in phase". It would
be interesting to know if anyone else with this hardware combination
gets this error when running amverify or not. If you have not bothered
with amverify, I would urge you to try it, lest your tapes are
unreadable when you need them most.
#!/bin/sh
#
# Exit Status:
# 0 Alles Ok
# 1 Illegal Request
# 2 Fatal Error
#
# Contributed by Eric DOUTRELEAU <[EMAIL PROTECTED]>
# This is supposed to work with Zubkoff/Dandelion version of mtx
#
# Modified by Joe Rhett <[EMAIL PROTECTED]>
# to work with MTX 1.2.9 by Eric Lee Green http://mtx.sourceforge.net
#
# Modified by Chris Pascoe <[EMAIL PROTECTED]> to work with a
# Dell PowerVault 120T, MTX 1.2.10 and add some intelligence to
# exit when all slots are empty. (Jan 2001). Fixed introduced
# bugs when going backwards and potential problem with changers having
# more than 9 slots (July 2001).
#
# You may need to customize these things
MT=/bin/mt
MTF=-f
MTX=/sbin/mtx
DD=/bin/dd
cleanslot=-1
accessbeforeclean=119
ONLINEREGEX="ONLINE|READY|sense[_ ]key[(]0x0[)]|sense key error = 0|^er=0$"
# No user-level customized required beyond this point.
# try to hit all the possibilities here
prefix=/usr/local/amanda/2.4.2p1
exec_prefix=${prefix}
sbindir=${exec_prefix}/sbin
libexecdir=${exec_prefix}/libexec
PATH=$sbindir:$libexecdir:/usr/bin:/bin:/usr/sbin:/sbin:/opt/local/bin
export PATH
if [ -d "/tmp/amanda" ]; then
DBGFILE=/tmp/amanda/changer.debug
else
DBGFILE=/dev/null
fi
if [ ! -f "amanda.conf" ]; then
echo "No amanda config in current directory"
exit 2;
fi
USE_VERSION_SUFFIXES="no"
if test "$USE_VERSION_SUFFIXES" = "yes"; then
SUF="-2.4.2p1"
else
SUF=
fi
myname=$0
tape=`amgetconf$SUF tapedev`
TAPE=`amgetconf$SUF changerdev`; export TAPE # for mtx command
if [ "$tape" = "/dev/null" -o "$TAPE" = "/dev/null" ]; then
echo "Both tapedev and changerdev must be specified in config file";
exit 2;
fi
changerfile=`amgetconf$SUF changerfile`
if [ -z "$changerfile" ];
then
echo "changerfile not specified in amanda config?"
exit 2
fi
cleanfile=$changerfile-clean
accessfile=$changerfile-access
slotfile=$changerfile-slot
[ ! -f $cleanfile ] && echo 0 > $cleanfile
[ ! -f $accessfile ] && echo 0 > $accessfile
[ ! -f $slotfile ] && echo 0 > $slotfile
cleancount=`cat $cleanfile`
accesscount=`cat $accessfile`
# Routines start here
readstatus() {
usedslot=`$MTX status |
sed -n 's/Data Transfer Element 0:Empty/-1/p;s/Data Transfer Element
0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p'`
if [ "$usedslot" -eq "-1" ]; then
echo '-1' > $slotfile
fi
echo "STATUS -> currently loaded slot = $usedslot" >> $DBGFILE
}
getslots() {
slotlist=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):\(Empty\|Full\)/\1/p'`
fullslots=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):Full/\1/p;s/Data Transfer
Element 0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p'`
dataslots=`$MTX status |
sed -n 's/Storage Element \([0-9]\+\):Full/\1/p;s/Data Transfer
Element 0:Full (Storage Element \([0-9]\+\) Loaded)/\1/p' |
fgrep -v "^${cleanslot}$" | sort -n`
numdataslots=`echo $dataslots | wc -w | awk '{print $1}'`
if [ -z "$dataslots" ]; then
dataslots="-1"
fi
firstslot=`echo $slotlist | awk '{print $1}'`
lastslot=`echo $slotlist | awk '{print $NF}'`
firstdataslot=`echo $dataslots | awk '{print $1}'`
lastdataslot=`echo $dataslots | awk '{print $NF}'`
nextdataslot=`ech
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
I have this changer working with RedHat 7.2 using the 2.4.2p2 chg-zd-mtx script. The trick to getting it working is getting the configuration file for chg-zd-mtx right. Make sure all the variables mentioned in the script are defined. In my case I have (in /etc/amanda/CONFIG/chg-zd-mtx.conf, replace CONFIG with your configuration name): firstslot=1 1st tape slot lastslot=7 Last tape slot cleanslot=7 Slot with cleaner tape AUTOCLEAN=0 Set to '1' or greater to enable autocleancount=99 Number of access before a clean. havereader=0 If you have a barcode reader, set to 1. offlinestatus=1 Set to 0 if 'mt status' gives an OFFLINE_BEFORE_UNLOAD=1 Does your tape driver require a In /etc/amanda/CONFIG/amanda.conf I have: tpchanger "chg-zd-mtx" tapedev "/dev/nst0" changerfile "/etc/amanda/CONFIG/chg-zd-mtx" changerdev "/dev/sg5" Of course, you're tapedev and changerdev may be different. Also, if you've been having problems getting it working you may have stuff in chg-zd-mtx-* that is wrong -- I'd delete them before trying again. Jim On Sun, Mar 24, 2002 at 02:22:47AM -0500, Derrick Miller wrote: > Hi all, > > A couple weeks ago I posted a message asking for help configuring Amanda > 2.4.2p2 to work under Red Hat Linux 7.2 (i386) with an ADIC FastStor DLT > changer. (It's actually a Dell PowerVault 120T DLT1, but the changer > component is manufactured by ADIC) > > To recap, everything works fine from the command line - mtx and tar work fine > for loading/unloading tapes and doing backups/restores. I just can't get > Amanda to even begin to work with the changer. I've tried both the chg-mtx > and chg-zd-mtx scripts. > > John Jackson pointed me to a new chg-zd-mtx script available at > ftp://gandalf.cc.purdue.edu/pub/amanda/chg-zd-mtx.sh.in-243 (thanks John) but > it also barfs, complaining about "/usr/lib/amanda/chg-zd-mtx: > @AMANDA_TMPDIR@/chg-zd-mtx.status.30913: No such file or directory" I could > be wrong, but from digging through the script, I'm guessing that it requires > a later version of Amanda than 2.4.2p2. > > So, my questions: > > 1) Does anyone have this changer working with Amanda? I would give just > about anything to get my hands on copies of working config and chg-zd-mtx > files. > > 2) If it's not possible to get it working under 2.4.2p2, should I try the > 2.4.3b3 release? Is that version useable in a production environment, and/or > is it any more likely to work for me than 2.4.2p2? > > Any pointers whatsoever would be greatly appreciated, as I've been beating my > head against this for over a month and have made very little progress. > > Thanks, > Derrick Miller
Re: ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
>John Jackson pointed me to a new chg-zd-mtx script ... but >it also barfs, complaining about "/usr/lib/amanda/chg-zd-mtx: >@AMANDA_TMPDIR@/chg-zd-mtx.status.30913: No such file or directory" ... I also said: Save your current chg-zd-mtx.sh.in and replace it with this one, then rerun ./configure (or config.status) and re-install. Did you do this? If the installed chg-zd-mtx still has a bunch of @XXX@ things in it, it implies you didn't rerun ./configure (which substitutes them while converting chg-zd-mtx.sh.in to chg-zd-mtx.sh). Then you need to rerun "make" and "make install" to finish the creation and installation of the script. You cannot just take the script as I presented it and drop it into your executable area. >Derrick Miller John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
ADIC / Dell Powervault 120T changer problems under Linux (Pt 2)
Hi all, A couple weeks ago I posted a message asking for help configuring Amanda 2.4.2p2 to work under Red Hat Linux 7.2 (i386) with an ADIC FastStor DLT changer. (It's actually a Dell PowerVault 120T DLT1, but the changer component is manufactured by ADIC) To recap, everything works fine from the command line - mtx and tar work fine for loading/unloading tapes and doing backups/restores. I just can't get Amanda to even begin to work with the changer. I've tried both the chg-mtx and chg-zd-mtx scripts. John Jackson pointed me to a new chg-zd-mtx script available at ftp://gandalf.cc.purdue.edu/pub/amanda/chg-zd-mtx.sh.in-243 (thanks John) but it also barfs, complaining about "/usr/lib/amanda/chg-zd-mtx: @AMANDA_TMPDIR@/chg-zd-mtx.status.30913: No such file or directory" I could be wrong, but from digging through the script, I'm guessing that it requires a later version of Amanda than 2.4.2p2. So, my questions: 1) Does anyone have this changer working with Amanda? I would give just about anything to get my hands on copies of working config and chg-zd-mtx files. 2) If it's not possible to get it working under 2.4.2p2, should I try the 2.4.3b3 release? Is that version useable in a production environment, and/or is it any more likely to work for me than 2.4.2p2? Any pointers whatsoever would be greatly appreciated, as I've been beating my head against this for over a month and have made very little progress. Thanks, Derrick Miller
Re: ADIC / Dell Powervault 120T changer problems under Linux
>The problem is that I cannot get Amanda to play nice with the changer. ... Ah, yes. Changer configuration. A true art form :-). >... and am using the 'chg-mtx' changer script. ... As I recall, that one is *only* for use with an old HP version of mtx. If you're using any recent version of mtx, you should use chg-zd-mtx. You might try this version of the changer: ftp://gandalf.cc.purdue.edu/pub/amanda/chg-zd-mtx.sh.in-243 Save your current chg-zd-mtx.sh.in and replace it with this one, then rerun ./configure (or config.status) and re-install. It's what will be released with 2.4.3, but I'm pretty sure it works with older versions. There is a lot of cleanup in the script, including a whole bunch of comments at the front about how to set it up. It also does a lot better job of logging (to /tmp/amanda). >Derrick John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: ADIC / Dell Powervault 120T changer problems under Linux
Hi, there are some files amanda needs to store the status of your changer, but doesn't create by it's own. yust create the files it complains about by "touch /path/to/file" as amanda-user and all should work. AFAIK the names of these files are stored in the amanda config-file for your changer-script. Have a nice day Christoph Derrick Miller schrieb: > > Hi all: > > I have been working for the past several weeks to get Amanda 2.4.2p2 working > under Red Hat Linux 7.2 (i386) with an ADIC FastStor DLT changer. (It's > actually a Dell PowerVault 120T DLT1, but the changer component is > manufactured by ADIC) > > Anyway, I have everything working well enough outside of Amanda, from the > command line. mtx works very well - I can load and unload tapes at will, > check status, etc. And, I can do tape backups of local filesystems just fine > with tar. > > The problem is that I cannot get Amanda to play nice with the changer. I've > configured everything as best as I can, and am using the 'chg-mtx' changer > script. When I try to label my tapes with the following command... > > su amanda -c "amlabel DailySet1 DailySet101 slot 1" > > ...I get this error message: > > amlabel: could not load slot "1": /usr/adm/amanda/changer-status-clean: No > such file or directory > > Does anyone know what this "changer-status-clean" file is? Am I using the > wrong changer script for this model? > > Any pointers at all would be greatly appreciated, as I am clueless as to what > to try next. > > Thanks, > Derrick
ADIC / Dell Powervault 120T changer problems under Linux
Hi all: I have been working for the past several weeks to get Amanda 2.4.2p2 working under Red Hat Linux 7.2 (i386) with an ADIC FastStor DLT changer. (It's actually a Dell PowerVault 120T DLT1, but the changer component is manufactured by ADIC) Anyway, I have everything working well enough outside of Amanda, from the command line. mtx works very well - I can load and unload tapes at will, check status, etc. And, I can do tape backups of local filesystems just fine with tar. The problem is that I cannot get Amanda to play nice with the changer. I've configured everything as best as I can, and am using the 'chg-mtx' changer script. When I try to label my tapes with the following command... su amanda -c "amlabel DailySet1 DailySet101 slot 1" ...I get this error message: amlabel: could not load slot "1": /usr/adm/amanda/changer-status-clean: No such file or directory Does anyone know what this "changer-status-clean" file is? Am I using the wrong changer script for this model? Any pointers at all would be greatly appreciated, as I am clueless as to what to try next. Thanks, Derrick
Amanda tape changer problems
Hi,
I am having some problems when i try to use multi-tape deices. The problem is
that i cannot dump to the tape the backup. this is the message i receive
after amdump:
These dumps were to tape avati01.
The next tape Amanda expects to use is: a new tape.
FAILURE AND STRANGE DUMP SUMMARY:
localhost /teste lev 0 FAILED [dump larger than tape, but cannot
incremental dump new disk]
planner: FATAL cannot fit anything on tape, bailing out
I am sending my amanda.conf file with some other informations ( disklist,
change.conf , amtape show, amcheck) so that someone can help me.
I am using a Adaptec scsi device AHA-2940 and my tape devices are :
nst0 - SONY SDT S7000
nst1 - SEAGATE std2800-N
both devices with 90M tapes
hope anyone can help me
this are the fles:
amanda.conf
org "avati"
mailto "[EMAIL PROTECTED]"
dumpuser "amanda"
inparallel 4
netusage 6000 Kbps
dumpcycle 2 day
runspercycle 2
tapecycle 3 tapes
bumpsize 10 Mb
bumpdays 2
bumpmult 1.5
etimeout 300
dtimeout 1800
ctimeout 30
tapebufs 20
runtapes 1
tpchanger "chg-multi"
tapedev "0"
changerfile "/etc/amanda/avati/changer.conf"
tapetype HP-DAT
labelstr "^avati[0-9][0-9]*$"
infofile "/var/lib/amanda/DailySet1/curinfo"
logdir "/var/lib/amanda/DailySet1"
indexdir "/var/lib/amanda/DailySet1/index"
define tapetype HP-DAT {
comment "DAT tape drives"
length 1930 mbytes
filemark 111 kbytes
speed 468 kbytes
}
define dumptype global {
comment "Global definitions"
}
define dumptype root-tar {
global
program "GNUTAR"
comment "root partitions dumped with tar"
compress none
index
exclude list "/usr/local/lib/amanda/exclude.gtar"
priority low
}
define interface local {
comment "a local disk"
use 1000 kbps
}
define interface eth0 {
comment "100 Mbps ethernet"
use 1 kbps
}
=//=
disklist
localhost /teste root-tar
=//=
changer.conf
multieject 0
gravity 0
needeject 0
ejectdelay 0
statefile /var/lib/amanda/changer-status
firstslot 1
lastslot 2
slot 1 /dev/nst0
slot 2 /dev/nst1
=//=
amtape avati show
amtape: scanning all 2 slots in tape-changer rack:
slot 1: date Xlabel avati01
slot 2: date Xlabel avati02
=//=
amcheck avati
lot 1 /dev/nst0
slot 2 /dev/nst1
=//=
amtape avati show
Amanda Tape Server Host Check
-
amcheck-server: slot 1: date Xlabel avati01 (first labelstr match)
NOTE: skipping tape-writable test
Tape avati01 label ok
NOTE: info dir /var/lib/amanda/DailySet1/curinfo: does not exist
NOTE: it will be created on the next run
NOTE: index dir /var/lib/amanda/DailySet1/index: does not exist
Server check took 5.488 seconds
Amanda Backup Client Hosts Check
Client check: 1 host checked in 0.041 seconds, 0 problems found
(brought to you by Amanda 2.4.2p2)
[]s
Túlio Machado de Faria
Coordenador Técnico
Avati Segurança Digital Avançada
Av. Raja Gabaglia, 2708 sala 312
Estoril - CEP 30350-540
Belo Horizonte - Minas Gerais - Brasil
http://www.avati.com.br
[EMAIL PROTECTED]
Re: changer problems
Hi, On Tue, Jul 10, 2001 at 04:41:12AM -, [EMAIL PROTECTED] wrote: > > The chg-scsi in 2.4.2p2 and 2.5.0 cvs does not work right with > this changer - after loading 10th tape won't do anything else Huh, if have an EXB library here as my development equipment and it works great. > (yes, I have gravity set to 0) - Hmm there is no option gravity for chg-scsi, could it be that your setup ist wrong. Can you send me your amanda.conf and changer config file to see whats wrong, and if there is something wrong in chg-scsi i will fix it :-)... Thomas -- --- | Thomas Hepper[EMAIL PROTECTED] | | ( If the above address fail try ) | | ( [EMAIL PROTECTED])| ---
Re: changer problems
exact same problem: [amanda@bserver fontenot]$ mtx --version mtx version 1.2.13 [amanda@bserver fontenot]$ mtx status Storage Changer /dev/changer:2 Drives, 11 Slots ( 0 Import/Export ) Data Transfer Element 0:Empty Data Transfer Element 1:Empty Storage Element 1:Full Storage Element 2:Full :VolumeTag= Storage Element 3:Full Storage Element 4:Full Storage Element 5:Full :VolumeTag= Storage Element 6:Full Storage Element 7:Full Storage Element 8:Full Storage Element 9:Full Storage Element 10:Full :VolumeTag= Storage Element 11:Full This is an ExaByte EXB-210 library with one 8mm tape drive, no barcode reader. The chg-scsi in 2.4.2p2 and 2.5.0 cvs does not work right with this changer - after loading 10th tape won't do anything else (yes, I have gravity set to 0) - so I switched to mtx which controls the changer great. Unfortunately, of the three(!) chg-mtx type scripts now at my disposal - two come with amanda, one with mtx - NONE of them work right. Between chg-scsi not working and this problem, amanda is taking me a lot longer to get going than it should. I also lost a day because the cvs version of amlabel does not report errors like the 2.4.2p2 version does - it did not give me the sense errors so I had no way of knowing that it was failing due to bad tapes. Once I reverted to 2.4.2p2 that problem became obvious. Dwayne ., "John R. Jackson" <[EMAIL PROTECTED]> wrote: > >Still I have problems with amanda 2.4.2p2 and my changer. I got the > >error "amcheck-server: could not get changer info: badly formed result > >from changer: "/usr/local/libexec/chg-zd-mtx: [: -eq: unary operator > >expected" > > What version of mtx are you using? What does "mtx -f /dev/sga status" > have to say (in particular, the "Data Transfer Element" lines)? > > >Sandra > > John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: changer problems
>Still I have problems with amanda 2.4.2p2 and my changer. I got the >error "amcheck-server: could not get changer info: badly formed result >from changer: "/usr/local/libexec/chg-zd-mtx: [: -eq: unary operator >expected" What version of mtx are you using? What does "mtx -f /dev/sga status" have to say (in particular, the "Data Transfer Element" lines)? >Sandra John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
changer problems
Hi : Still I have problems with amanda 2.4.2p2 and my changer. I got the error "amcheck-server: could not get changer info: badly formed result from changer: "/usr/local/libexec/chg-zd-mtx: [: -eq: unary operator expected" Amanda Backup Client Hosts Check Client check: 8 hosts checked in 0.445 seconds, 0 problems found (brought to you by Amanda 2.4.2p2) this my amanda.conf that is /usr/local/etc/amanda/miro_daily runtapes 6 # number of tapes to be used in a single run of amdump # we have two to enable writing to holding disk for morn ing amflushes. tpchanger "chg-zd-mtx" # the tape-changer glue script # Even though tapedev is ignored since we use # tpchanger, we keep this in case we need to revert # to no changing. tapedev "/dev/nst0" # the no-rewind tape device to be used rawtapedev "/dev/null" # the raw device to be used (ftape only) #changerstatus "/usr/local/etc/amanda/miro_daily/changer-status" changerfile "/usr/local/etc/amanda/miro_daily/changer" changerdev "/dev/sga" tapetype DDS3 -- my changer.conf file is on /usr/local/etc/amanda/miro_daily number_configs 1 eject 0 # Tapedrives need an eject command number_configs 1 sleep 90 # Seconds to wait until the tape gets ready cleanmax100 # How many times could a cleaning tape get used changerdev /dev/sga # # Next comes the data for drive 0 # config 0 drivenum0 dev /dev/nst0 scsitapedev /dev/sga startuse1 # The slots associated with the drive 0 enduse 6 # statfile/usr/local/etc/amanda/miro_daily/tape0-slot # The file where th e actual slot is stored #cleancart 6 # the slot where the cleaningcartridge for drive 0 is lo cated #cleanfile /usr/local/etc/amanda/miro_daily/cleanfile # The file where the cleanings are recorded usagecount /usr/local/etc/amanda/miro_daily/totaltime tapestatus /usr/local/etc/amanda/miro_daily/changer-status # here will some status infos be stored [root@miro miro_daily]# vi changer.conf number_configs 1 eject 0 # Tapedrives need an eject command sleep 90 # Seconds to wait until the tape gets ready cleanmax100 # How many times could a cleaning tape get used changerdev /dev/sga # # Next comes the data for drive 0 # config 0 drivenum0 dev /dev/nst0 scsitapedev /dev/sga startuse1 # The slots associated with the drive 0 enduse 6 # statfile/usr/local/etc/amanda/miro_daily/tape0-slot # The file where th e actual slot is stored cleancart 6 # the slot where the cleaningcartridge for drive 0 is lo cated cleanfile /usr/local/etc/amanda/miro_daily/cleanfile # The file where the cleanings are recorded usagecount /usr/local/etc/amanda/miro_daily/totaltime tapestatus /usr/local/etc/amanda/miro_daily/changer-status # here will some status infos be stored this my chg-zd-mtx file in /usr/local/libexec firstslot=1 1st tape slot lastslot=5 Last tape slot cleanslot=6 Slot with cleaner tape # # # Do you want to clean the drive after a certain number of accesses? # # NOTE - This is unreliable, since 'accesses' aren't 'uses', and we # #have no reliable way to count this. A single amcheck could # #generate as many accesses as slots you have, plus 1. # # ALSO NOTE - many modern tape loaders handle this automatically. # AUTOCLEAN=0 Set to '1' or greater to enable # autocleancount=99 Number of access before a clean. # havereader=1 If you have a barcode reader, set to 1. # offlinestatus=0 Set to 0 if 'mt status' gives an OFFLINE_BEFORE_UNLOAD=0 Does your tape driver require a # 'mt offline' before mtx unload? # You may need to customize these things MT=/bin/mt MTF=-f MTX=/usr/local/bin/mtx So anything is wrong with that? I hope you guys can help me. Thanks, Sandra
Re: Changer problems
Hi, On Sun, Jun 03, 2001 at 03:41:15PM -0500, Gene Heskett wrote: > Hi all; > > I've been lurking for about a day, but haven't seen any messages that > relate to my problem. > > Specifically, even though I've recompiled amanda with the changer device > (/dev/sg/2) defined, no amanda tape utils can find the changer. Hmm, can you give some more Infos, whcih OS, which amanda-version > > I can run it just fine with mtx. > > Also, it seem every version of the scripts in changer-src has their own > idea of where the config files should be, can't this be consolidated, > I've now got amanda stuffs in 8 or 9 different directories! Ups, the syntax might be different, but you can keep all config files in one directory .. > > The changer is a Seagate 4586NP, brand new, and I sure could use some > hand-holding till I get the feel of how and what this program is doing. > OK lets try it. If you have a linux/bsd system you can try to get the development version of amanda 2.5.0 and use the chg-scsi prgramm from there. (this version is available through CVS, see www.amanda.org) After compiling but not installaing this version you can try to run cd to_your_amanda_conifg_directory /path_to_the_source/changer-src/chg-scsi -genconf If this works you can copy the chg-scsi version to your libexec directory, save the ouput from the above in a file chg-scsi.conf, and add the following to amanda.conf tpchanger "chg-scsi"# the tape-changer glue script tapedev "0" changerfile "/path_to_your_config/chg-scsi.conf" remove everything like changerdev, tapedev from the amanda.conf No you can try to get a known state for the robot amtape reset and write a label amlabel conf xxx slot current (xxx must be a valid label name) Hope this helps a little bit Thomas -- --- | Thomas Hepper[EMAIL PROTECTED] | | ( If the above address fail try ) | | ( [EMAIL PROTECTED])| ---
RE: chg-manual "poor man's" changer problems
On Thu, 19 Apr 2001, Bort, Paul wrote: Hello, here I'm again. Well, the solution for my problem is really simple, but I have had to read the script first as Paul Bort implied. FreeBSD's 'mt' command reports in a different way than other OS' mt and for that, the variable $ONLINEREGEX in the script chg-manual contains no identifying expression to detect that a tape is 'online'. I added there a string which is only present if a tape has been sucessfully loaded into the drive and now, voila, it works fine also for FreeBSD driven UNIX boxes ;-) Thanks a lot ... :>(Warning: Stream-Of-Trace) :> :>OK, I'm getting closer. The reason that it keeps asking for a tape in the :>same slot is because it doesn't know what slot it's on. The message: :> :> -> loaded <> :> :>Should really have a slot number in the <>, but doesn't. The variable $used :>in chg-manual is empty for unknown reasons. :> :>I wish I knew more shell script. :> :>OK, it looks like the line before the "loaded" message is the relevant one: :> :>used=`$MT $MTF $tape status 2>&1 | tee -a $dbglog | $EGREP "$ONLINEREGEX"` :> :>This is placing the status message in the log, but it looks like no status :>message I've ever seen. :> :>My status messages look like this: :> :>[amanda@tape configdir]$ mt -f /dev/nst0 status :>SCSI 2 tape drive: :>File number=0, block number=1, partition=0. :>Tape block size 0 bytes. Density code 0x0 (default). :>Soft error count since last status=0 :>General status bits on (101): :> ONLINE IM_REP_EN :> :>I think that the grep is looking for 'ONLINE' and not finding it in your :>status message, so it doens't think the tape drive is ready, and asks you to :>insert a tape. :> :>Next steps would be to find out what AMANDA is invoking with $MT, and find :>out if you have an mt command like mine, and how to bring the two closer :>together. :> :>Sorry this took so long. I'm still learning shell script. :> :> :>-Original Message- :>From: Hartmann, O. [mailto:[EMAIL PROTECTED]] :>Sent: Thursday, April 19, 2001 8:55 AM :>To: Bort, Paul :>Cc: [EMAIL PROTECTED] :>Subject: RE: chg-manual "poor man's" changer problems :> :> :>On Wed, 18 Apr 2001, Bort, Paul wrote: :> :> :>Of course, you're right, but this problem is reported by AMANDA. This error :>seems nasty. :> :>All right, I'll try to explain again the configuration and you'll see that :>many things are clear as a canonical declaration- there must be an small :>error or a microscopic misconfiguration ... :-( :> :>Operating system: FreeBSD 4.3-RC :>AMANDA: Amanda 2.4.2p1 :> :>Tapedrive:HP C1533A 9503 (4GB DDS-2 DAT drive) :> The DAT drive is online and accessible! :> AMANDA user 'operator' has full access to :> the device. :> Actually I can read and write normal cpio/pax/tar :> from and to this drive as 'operator'. :> :>Tape cardridges: HP and TDK DDS-2 DAT tapes, NOT labeled! :> :>Configuration:/etc/inetd.conf is sdet up already, also :>/etc/services. :> Within "operator's" home .amandahosts exists, but it :> is not used usually for local administrative :>operations. :> As defined at compiling time /etc/amandates exists :>and :> ist operator-writeable. :> A amanda.conf file exists and is clearly read out by :> AMANDA. The important configuration tags are: :> :> tpchanger "chg-manual" :> tapedev "/dev/nsa0" :> #rawtapedev "/dev/nrsa0" :> changerfile :>"/usr/local/etc/amanda/klima/changer.conf" :> :> The file 'changer.conf' exists, but it is empty. for :>any tag I put :> in there, I get an error by running amcheck so it :>seems to be :> right to leave this file clear, but it must exists :>due the fact :> the driver script 'chg-manual' needs it as the base :>former for the :> following also existent files: :> :> changer.confempty :> changer.conf-access 22 (times of access) :> changer.conf-clean 1 :>
RE: chg-manual "poor man's" changer problems
If you have time to submit a diff or patch to amanda-hackers that would be very helpful. -Original Message- From: Hartmann, O. [mailto:[EMAIL PROTECTED]] Sent: Friday, April 20, 2001 2:09 PM To: Bort, Paul Cc: [EMAIL PROTECTED] Subject: RE: chg-manual "poor man's" changer problems On Thu, 19 Apr 2001, Bort, Paul wrote: Hello, here I'm again. Well, the solution for my problem is really simple, but I have had to read the script first as Paul Bort implied. FreeBSD's 'mt' command reports in a different way than other OS' mt and for that, the variable $ONLINEREGEX in the script chg-manual contains no identifying expression to detect that a tape is 'online'. I added there a string which is only present if a tape has been sucessfully loaded into the drive and now, voila, it works fine also for FreeBSD driven UNIX boxes ;-) Thanks a lot ... :>(Warning: Stream-Of-Trace) :> :>OK, I'm getting closer. The reason that it keeps asking for a tape in the :>same slot is because it doesn't know what slot it's on. The message: :> :> -> loaded <> :> :>Should really have a slot number in the <>, but doesn't. The variable $used :>in chg-manual is empty for unknown reasons. :> :>I wish I knew more shell script. :> :>OK, it looks like the line before the "loaded" message is the relevant one: :> :>used=`$MT $MTF $tape status 2>&1 | tee -a $dbglog | $EGREP "$ONLINEREGEX"` :> :>This is placing the status message in the log, but it looks like no status :>message I've ever seen. :> :>My status messages look like this: :> :>[amanda@tape configdir]$ mt -f /dev/nst0 status :>SCSI 2 tape drive: :>File number=0, block number=1, partition=0. :>Tape block size 0 bytes. Density code 0x0 (default). :>Soft error count since last status=0 :>General status bits on (101): :> ONLINE IM_REP_EN :> :>I think that the grep is looking for 'ONLINE' and not finding it in your :>status message, so it doens't think the tape drive is ready, and asks you to :>insert a tape. :> :>Next steps would be to find out what AMANDA is invoking with $MT, and find :>out if you have an mt command like mine, and how to bring the two closer :>together. :> :>Sorry this took so long. I'm still learning shell script. :> :> :>-Original Message- :>From: Hartmann, O. [mailto:[EMAIL PROTECTED]] :>Sent: Thursday, April 19, 2001 8:55 AM :>To: Bort, Paul :>Cc: [EMAIL PROTECTED] :>Subject: RE: chg-manual "poor man's" changer problems :> :> :>On Wed, 18 Apr 2001, Bort, Paul wrote: :> :> :>Of course, you're right, but this problem is reported by AMANDA. This error :>seems nasty. :> :>All right, I'll try to explain again the configuration and you'll see that :>many things are clear as a canonical declaration- there must be an small :>error or a microscopic misconfiguration ... :-( :> :>Operating system: FreeBSD 4.3-RC :>AMANDA: Amanda 2.4.2p1 :> :>Tapedrive:HP C1533A 9503 (4GB DDS-2 DAT drive) :> The DAT drive is online and accessible! :> AMANDA user 'operator' has full access to :> the device. :> Actually I can read and write normal cpio/pax/tar :> from and to this drive as 'operator'. :> :>Tape cardridges: HP and TDK DDS-2 DAT tapes, NOT labeled! :> :>Configuration:/etc/inetd.conf is sdet up already, also :>/etc/services. :> Within "operator's" home .amandahosts exists, but it :> is not used usually for local administrative :>operations. :> As defined at compiling time /etc/amandates exists :>and :> ist operator-writeable. :> A amanda.conf file exists and is clearly read out by :> AMANDA. The important configuration tags are: :> :> tpchanger "chg-manual" :> tapedev "/dev/nsa0" :> #rawtapedev "/dev/nrsa0" :> changerfile :>"/usr/local/etc/amanda/klima/changer.conf" :> :> The file 'changer.conf' exists, but it is empty. for :>any tag I put :> in there, I get an error by running amcheck so it :>seems to be :> right to leave this file clear, but it must exists :>due the fact :> the driver script 'chg-manual' needs it as the b
RE: chg-manual "poor man's" changer problems
(Warning: Stream-Of-Trace) OK, I'm getting closer. The reason that it keeps asking for a tape in the same slot is because it doesn't know what slot it's on. The message: -> loaded <> Should really have a slot number in the <>, but doesn't. The variable $used in chg-manual is empty for unknown reasons. I wish I knew more shell script. OK, it looks like the line before the "loaded" message is the relevant one: used=`$MT $MTF $tape status 2>&1 | tee -a $dbglog | $EGREP "$ONLINEREGEX"` This is placing the status message in the log, but it looks like no status message I've ever seen. My status messages look like this: [amanda@tape configdir]$ mt -f /dev/nst0 status SCSI 2 tape drive: File number=0, block number=1, partition=0. Tape block size 0 bytes. Density code 0x0 (default). Soft error count since last status=0 General status bits on (101): ONLINE IM_REP_EN I think that the grep is looking for 'ONLINE' and not finding it in your status message, so it doens't think the tape drive is ready, and asks you to insert a tape. Next steps would be to find out what AMANDA is invoking with $MT, and find out if you have an mt command like mine, and how to bring the two closer together. Sorry this took so long. I'm still learning shell script. -Original Message- From: Hartmann, O. [mailto:[EMAIL PROTECTED]] Sent: Thursday, April 19, 2001 8:55 AM To: Bort, Paul Cc: [EMAIL PROTECTED] Subject: RE: chg-manual "poor man's" changer problems On Wed, 18 Apr 2001, Bort, Paul wrote: Of course, you're right, but this problem is reported by AMANDA. This error seems nasty. All right, I'll try to explain again the configuration and you'll see that many things are clear as a canonical declaration- there must be an small error or a microscopic misconfiguration ... :-( Operating system: FreeBSD 4.3-RC AMANDA: Amanda 2.4.2p1 Tapedrive: HP C1533A 9503 (4GB DDS-2 DAT drive) The DAT drive is online and accessible! AMANDA user 'operator' has full access to the device. Actually I can read and write normal cpio/pax/tar from and to this drive as 'operator'. Tape cardridges:HP and TDK DDS-2 DAT tapes, NOT labeled! Configuration: /etc/inetd.conf is sdet up already, also /etc/services. Within "operator's" home .amandahosts exists, but it is not used usually for local administrative operations. As defined at compiling time /etc/amandates exists and ist operator-writeable. A amanda.conf file exists and is clearly read out by AMANDA. The important configuration tags are: tpchanger "chg-manual" tapedev "/dev/nsa0" #rawtapedev "/dev/nrsa0" changerfile "/usr/local/etc/amanda/klima/changer.conf" The file 'changer.conf' exists, but it is empty. for any tag I put in there, I get an error by running amcheck so it seems to be right to leave this file clear, but it must exists due the fact the driver script 'chg-manual' needs it as the base former for the following also existent files: changer.confempty changer.conf-access 22 (times of access) changer.conf-clean 1 changer.conf-slot 1 (reflects the slot to use), changed this temporarily to 0, doesn't matter ... What I did, do and what the logs are: With the above mentioned configuration I tried to label or check tapes. For instance of labeling, I type exactly amlabel -f CONFIG_NAME CONFIG_TAG and gets: insert tape into slot 0 and press return I hit return, but nothing happened and the message occurs again, and again, and again ... I removed manually the tape, reinserted it and hit again return. The same. I ejected and reinserted again, but then I did a 'mt rewind' and in one case 'mt retention', with the effect of nothing in the first case and a complete forward and rewind of the tape. then I hit return again, but I got the same message again to insert a tape in slot 0 (or one, depends on what is in 'changer.conf-slot'. The content of of /tmp/amanda/amlabel.debug
RE: chg-manual "poor man's" changer problems
er State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> -> rewind /dev/nsa0 -> status /dev/nsa0 Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> END-- That seems really normal to me, the tape is inserted, loaded and accessible and recognized by the tape subdevice and amanda recognizes it. That seems more than courios to me :-( :>Usually if AMANDA kicks out a tape during amcheck it is either because the :>tape isn't labeled or isn't the labeled tape she was expecting (for example, :>if you forgot to change the tape.) :> :>With no offense intended, have you already amlabel'ed the tape you are :>inserting during amcheck? :> :>Could you post everything that amcheck returns up (and including) the insert :>message? This might give a clue about why it is deciding to reject that tape :>and ask for another one. :> :> :> :>-Original Message- :>From: Hartmann, O. [mailto:[EMAIL PROTECTED]] :>Sent: Wednesday, April 18, 2001 4:50 PM :>To: Bort, Paul :>Cc: [EMAIL PROTECTED] :>Subject: RE: chg-manual "poor man's" changer problems :> :> :>On Wed, 18 Apr 2001, Bort, Paul wrote: :> :>Hello. :>Thanks for responding. :> :>Our systems are all based on FreeBSD 4.3-RC3, I forgot to mention that. :>well, the backup operator is 'operator' and this user has full read/write :>permissions to the tape device (/dev/nsa0, /dev/nrsa0). :> :>whenever I put any line into "changer.conf" as configured with "changerfile" :>in amanda.conf I receive errors about that line - so I omit all lines, I :>simple :>"touch" a changerfile 'changer.conf' to get the control files as expected :>from :>"chg-manual". :> :>When trying 'amcheck config' I get some infos about the holding disk and I :>was :>offered to insert a tape in slot 0 - and I should hit return. either I do :>immediately :>while the tape is already loaded into the HP DAT (4 GB DDS-2 drive) or I :>eject and :>reload it again, wait for the tape to be initialised and the hit return. :>Always :>with the same effect: I was offered to hit return again and load a tape into :>slot 0 - but nothing happened to the tape inserted into the tapedrive. :> :>When checking /tmp/amanda/changer.debug then I get a big file with this: :> :>Mode Density Blocksize bpi Compression :>Current: 0x24:DDS-2 variable 61000DCLZ :>-available modes- :>0:0x24:DDS-2 variable 61000DCLZ :>1:0x24:DDS-2 variable 61000DCLZ :>2:0x24:DDS-2 variable 61000DCLZ :>3:0x24:DDS-2 variable 61000DCLZ :>- :>Current Driver State: at rest. :>- :>File Number: 0 Record Number: 0Residual Count 0 :> -> loaded <> :> -> rewind /dev/nsa0 :> -> status /dev/nsa0 :>Mode Density Blocksize bpi Compression :>Current: 0x24:DDS-2 variable 61000DCLZ :>-available modes- :>0:0x24:DDS-2 variable 61000DCLZ :>1:0x24:DDS-2 variable 61000DCLZ :>2:0x24:DDS-2 variable 61000DCLZ :>3:0x24:DDS-2 variable 61000DCLZ :>- :>Current Driver State: at rest. :>- :>File Number: 0 Record Number: 0Residual Count 0 :> -> loaded <> :> -> rewind /dev/nsa0 :> -> status /dev/nsa0 :>Mode Density Blocksize bpi Compression :>Current: 0x24:DDS-2 variable 61000DCLZ :>-available modes- :>0:0x24:DDS-2 variable 61000DCLZ :>1:0x24:DDS-2 variable 61000DCLZ :>2:0x24:DDS-2 variable 61000DCLZ :>3:0x24:DDS-2 variable 61000DCLZ :>- :>Current Drive
RE: chg-manual "poor man's" changer problems
Usually if AMANDA kicks out a tape during amcheck it is either because the tape isn't labeled or isn't the labeled tape she was expecting (for example, if you forgot to change the tape.) With no offense intended, have you already amlabel'ed the tape you are inserting during amcheck? Could you post everything that amcheck returns up (and including) the insert message? This might give a clue about why it is deciding to reject that tape and ask for another one. -Original Message- From: Hartmann, O. [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 18, 2001 4:50 PM To: Bort, Paul Cc: [EMAIL PROTECTED] Subject: RE: chg-manual "poor man's" changer problems On Wed, 18 Apr 2001, Bort, Paul wrote: Hello. Thanks for responding. Our systems are all based on FreeBSD 4.3-RC3, I forgot to mention that. well, the backup operator is 'operator' and this user has full read/write permissions to the tape device (/dev/nsa0, /dev/nrsa0). whenever I put any line into "changer.conf" as configured with "changerfile" in amanda.conf I receive errors about that line - so I omit all lines, I simple "touch" a changerfile 'changer.conf' to get the control files as expected from "chg-manual". When trying 'amcheck config' I get some infos about the holding disk and I was offered to insert a tape in slot 0 - and I should hit return. either I do immediately while the tape is already loaded into the HP DAT (4 GB DDS-2 drive) or I eject and reload it again, wait for the tape to be initialised and the hit return. Always with the same effect: I was offered to hit return again and load a tape into slot 0 - but nothing happened to the tape inserted into the tapedrive. When checking /tmp/amanda/changer.debug then I get a big file with this: Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> -> rewind /dev/nsa0 -> status /dev/nsa0 Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> -> rewind /dev/nsa0 -> status /dev/nsa0 Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> ->END<- Well, it seems to be all right, tape is loaded, clear to be written on - but whats going on with the changer? It seems not to be willing to write anything :-( A normal tar, mt operations and so on work well with the tape and as user 'operator', so that is not a kind of a permission issue. :>The config stuff looks OK to me, I would suggest leaving off the slot part :>of the amlabel command. When I label tapes through chg-manual, I always just :>`amlabel Config2 NextTapeName`, and if it asks me to put a tape in a :>specific slot, I just re-insert the tape in the drive, wait for it to settle :>down (it's an EXB-8505, they take a while) and hit enter. Solid as a rock. :> :>The only other thing that comes to mind is to make sure that the amanda user :>has read and write permissions on the changerfile. :> :>Good Luck. :> :> :>-Original Message- :>From: Hartmann, O. [mailto:[EMAIL PROTECTED]] :>Sent: Wednesday, April 18, 2001 4:03 PM :>To: [EMAIL PROTECTED] :>Subject: chg-manual "poor man's" changer problems :> :> :>Dear Sirs. :> :>I already set up a HP SureStore 6x40i changer based tape device :>backing up our main server, now I tried to
RE: chg-manual "poor man's" changer problems
On Wed, 18 Apr 2001, Bort, Paul wrote: Hello. Thanks for responding. Our systems are all based on FreeBSD 4.3-RC3, I forgot to mention that. well, the backup operator is 'operator' and this user has full read/write permissions to the tape device (/dev/nsa0, /dev/nrsa0). whenever I put any line into "changer.conf" as configured with "changerfile" in amanda.conf I receive errors about that line - so I omit all lines, I simple "touch" a changerfile 'changer.conf' to get the control files as expected from "chg-manual". When trying 'amcheck config' I get some infos about the holding disk and I was offered to insert a tape in slot 0 - and I should hit return. either I do immediately while the tape is already loaded into the HP DAT (4 GB DDS-2 drive) or I eject and reload it again, wait for the tape to be initialised and the hit return. Always with the same effect: I was offered to hit return again and load a tape into slot 0 - but nothing happened to the tape inserted into the tapedrive. When checking /tmp/amanda/changer.debug then I get a big file with this: Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> -> rewind /dev/nsa0 -> status /dev/nsa0 Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> -> rewind /dev/nsa0 -> status /dev/nsa0 Mode Density Blocksize bpi Compression Current: 0x24:DDS-2 variable 61000DCLZ -available modes- 0:0x24:DDS-2 variable 61000DCLZ 1:0x24:DDS-2 variable 61000DCLZ 2:0x24:DDS-2 variable 61000DCLZ 3:0x24:DDS-2 variable 61000DCLZ - Current Driver State: at rest. - File Number: 0 Record Number: 0Residual Count 0 -> loaded <> ->END<- Well, it seems to be all right, tape is loaded, clear to be written on - but whats going on with the changer? It seems not to be willing to write anything :-( A normal tar, mt operations and so on work well with the tape and as user 'operator', so that is not a kind of a permission issue. :>The config stuff looks OK to me, I would suggest leaving off the slot part :>of the amlabel command. When I label tapes through chg-manual, I always just :>`amlabel Config2 NextTapeName`, and if it asks me to put a tape in a :>specific slot, I just re-insert the tape in the drive, wait for it to settle :>down (it's an EXB-8505, they take a while) and hit enter. Solid as a rock. :> :>The only other thing that comes to mind is to make sure that the amanda user :>has read and write permissions on the changerfile. :> :>Good Luck. :> :> :>-Original Message----- :>From: Hartmann, O. [mailto:[EMAIL PROTECTED]] :>Sent: Wednesday, April 18, 2001 4:03 PM :>To: [EMAIL PROTECTED] :>Subject: chg-manual "poor man's" changer problems :> :> :>Dear Sirs. :> :>I already set up a HP SureStore 6x40i changer based tape device :>backing up our main server, now I tried to install on a secondary :>machine with a simple HP DAT drive a second piece of amanda, but :>without success. :> :>I followed the steps described in the really confusing manuals and :>other FAQ-based stuff. :> :>If I'm right, I have to setup the following in amanda.conf: :> :>(here are only the relevant entries) :> :>tpchanger "chg-manual" :>tapedev "/dev/nsa0" :>changerfile :>"/usr/local/perfector/amanda/services-backup/changer.conf" :> :>The entry "changerfile" should be set due the fact, that chg-manual driver :>script :>searches for several status-semaphores created with t
RE: chg-manual "poor man's" changer problems
The config stuff looks OK to me, I would suggest leaving off the slot part of the amlabel command. When I label tapes through chg-manual, I always just `amlabel Config2 NextTapeName`, and if it asks me to put a tape in a specific slot, I just re-insert the tape in the drive, wait for it to settle down (it's an EXB-8505, they take a while) and hit enter. Solid as a rock. The only other thing that comes to mind is to make sure that the amanda user has read and write permissions on the changerfile. Good Luck. -Original Message- From: Hartmann, O. [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 18, 2001 4:03 PM To: [EMAIL PROTECTED] Subject: chg-manual "poor man's" changer problems Dear Sirs. I already set up a HP SureStore 6x40i changer based tape device backing up our main server, now I tried to install on a secondary machine with a simple HP DAT drive a second piece of amanda, but without success. I followed the steps described in the really confusing manuals and other FAQ-based stuff. If I'm right, I have to setup the following in amanda.conf: (here are only the relevant entries) tpchanger "chg-manual" tapedev "/dev/nsa0" changerfile "/usr/local/perfector/amanda/services-backup/changer.conf" The entry "changerfile" should be set due the fact, that chg-manual driver script searches for several status-semaphores created with the name of the base changerfile. If I omit this, I receive an error. The targeted changerfile is either empty or I configure some parameters like slot /dev/nsa0 etc. But I receive allways the same result. After typing "amlabel -f CONFIG CONFIGLABEL slot [0-9] (I tried slot 0 to 9, but configured either slot 0 or 1, for testing purposes to see whether the result changes) I get the message that I'm to be to insert a cartridge into drive 0 or drive 1 (depends on what I configured or what slot I gave amlabel). I inserted already a cartridge, but I always get the same message after hitting RETURN. I checked /tmp/amanda/ amlabel.debug and there is a mt output I receive by 'mt status' - and this seems absolutely normal as it shows some informations about the used tape (DDS-2 and so on). Can anybody help? What to be aware of when configuring a chg-manual device? -- MfG O. Hartmann [EMAIL PROTECTED] IT-Administration des Institut fuer Physik der Atmosphaere (IPA) Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinensaal) Tel: +496131/3924144 FAX: +496131/3923532
chg-manual "poor man's" changer problems
Dear Sirs. I already set up a HP SureStore 6x40i changer based tape device backing up our main server, now I tried to install on a secondary machine with a simple HP DAT drive a second piece of amanda, but without success. I followed the steps described in the really confusing manuals and other FAQ-based stuff. If I'm right, I have to setup the following in amanda.conf: (here are only the relevant entries) tpchanger "chg-manual" tapedev "/dev/nsa0" changerfile "/usr/local/perfector/amanda/services-backup/changer.conf" The entry "changerfile" should be set due the fact, that chg-manual driver script searches for several status-semaphores created with the name of the base changerfile. If I omit this, I receive an error. The targeted changerfile is either empty or I configure some parameters like slot /dev/nsa0 etc. But I receive allways the same result. After typing "amlabel -f CONFIG CONFIGLABEL slot [0-9] (I tried slot 0 to 9, but configured either slot 0 or 1, for testing purposes to see whether the result changes) I get the message that I'm to be to insert a cartridge into drive 0 or drive 1 (depends on what I configured or what slot I gave amlabel). I inserted already a cartridge, but I always get the same message after hitting RETURN. I checked /tmp/amanda/ amlabel.debug and there is a mt output I receive by 'mt status' - and this seems absolutely normal as it shows some informations about the used tape (DDS-2 and so on). Can anybody help? What to be aware of when configuring a chg-manual device? -- MfG O. Hartmann [EMAIL PROTECTED] IT-Administration des Institut fuer Physik der Atmosphaere (IPA) Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinensaal) Tel: +496131/3924144 FAX: +496131/3923532
Re: GTAR Tape Changer problems
>I configured AMANDA that way to use the first time 8 magazines, each >holding 5 tapes and one cleaning tape. For all filesystems GNU tar >(GTAR) is used so I do not fall into problems when dumping images >bigger than one tape. I think you're confused about some things, so let me explain a bit. Using GNU tar does not guarantee you avoid the problem of a dump bigger than a tape. That's a problem no matter what program you use with the current version of Amanda. Using GNU tar, however, allows you to break up on large dump into several smaller ones by listing each top level directory independently. But if you have a 2 GByte tape (just to pick a number) and tell GNU tar to dump something that is 5 GBytes, it's still not going to work. >The first time this runs well, AMANDA saved all the images out to >two or three tapes (like the multi volume mechanism when using a >simple tar -M backing up a big filesystem ...). ... But what really happened here was that when Amanda ran into EOT while writing one image, it started that image over again **from the beginning** on the next tape. Also note that tar had no idea this was going on. In fact, tar may not even have been running any more if the dump was put in the holding disk and Amanda is in the process of writing the image from there to tape. >amlabel labels each used cartridge used by AMANDA and in the manner >of using dump as the backup/save tool this means that each run is >made to tapes with increasing numbers. I didn't quite understand this, but I think I understand that you want to label each tape with an increasing number (which is normal). >At this moment, our magazines are filled up with cartridges each >labeled with the same label, means mag one is completely labeled >throughout each cartridge with LABEL00, the second mag's tapes >LABEL01 ... That was a mistake. You need to put a unique label on each tape. When Amanda wants a new tape, it looks in the "tapelist" file (probably in the same directory as amanda.conf) and it looks at your tapecycle value. If tapelist does not have as many tapes as tapecycle, Amanda will insist on a new tape (one not listed in tapelist). If there are enough tapes in tapelist, Amanda will ask for the oldest one but will also accept a new one. I'm not quite sure I understand what you want to do with your cartridges. If one amdump run uses three of the five tapes, what does the next run use? Does it use the next two tapes in the same cartridge, or do you want to put in the next cartridge? >How can I delete each label a fast way? ERASE is not a good idea >(using mt) due the fact it takes a long time for each tape to be >cleaned and delabeled ... You can use the -f option on amlabel. Or you can "mt rewind ; mt eof", but then you'll have to run amlabel again anyway. >O. Hartmann John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
GTAR Tape Changer problems
Dear Sirs. For our 250 GBytes RAID array backup on a FBSD 4.3 Server we use AMANDA 2.4.2p1. The tape changer is a HP SureStore 40x6i, which has a 6 cartridges magazine. I configured AMANDA that way to use the first time 8 magazines, each holding 5 tapes and one cleaning tape. For all filesystems GNU tar (GTAR) is used so I do not fall into problems when dumping images bigger than one tape. The first time this runs well, AMANDA saved all the images out to two or three tapes (like the multi volume mechanism when using a simple tar -M backing up a big filesystem ...). Now we changed several parameters (due to the lacking docimentation how to use GTAR we have had to recalibrate all parameters to the right size and want to start from the begining). amlabel labels each used cartridge used by AMANDA and in the manner of using dump as the backup/save tool this means that each run is made to tapes with increasing numbers. At this moment, our magazines are filled up with cartridges each labeled with the same label, means mag one is completely labeled throughout each cartridge with LABEL00, the second mag's tapes LABEL01 ... that means now, whenever GNU tar/AMANDA reaches the end of the first tape to continiue on the next one, it finds the same labeld tape in each slot, greps through the whole magazine and tells that it is out of tape. Well, I'm not very familiar how AMANDA/GTAR signs up each tape to keep in track with each one, but I want it to have 'clean' tapes because I think gtar is not continuing a LABEL00-tape on a second, like LABEL00:1, instead it needs LABEL01 - tape, or I'm wrong? How can I delete each label a fast way? ERASE is not a good idea (using mt) due the fact it takes a long time for each tape to be cleaned and delabeled ... -- MfG O. Hartmann [EMAIL PROTECTED] IT-Administration des Institut fuer Physik der Atmosphaere (IPA) Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinensaal) Tel: +496131/3924144 FAX: +496131/3923532
Re: amanda 2.4.2 and Changer problems
>Tue Jan 30 0:04:03 2001: enter: getTapeParams >Tue Jan 30 0:04:03 2001: running: /bin/chio -f /dev/null params I don't know much about chio, but passing /dev/null to it seems like a bad idea. That comes from "changerdev" in amanda.conf. Are you sure that parameter didn't get dinged in the upgrade or testing? > Clem Kumah John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
amanda 2.4.2 and Changer problems
Hi, I recently upgraded my version of amanda to 2.4.2 from 2.4.2-beta19 release. Since then I have problems getting the changer to work properly. For example if I look in the changer.debug file in /tmp/amanda, Tue Jan 30 0:04:03 2001: enter: getTapeParams Tue Jan 30 0:04:03 2001: running: /bin/chio -f /dev/null params If I issue the following command , I get the following error: chio: /dev/ch0: CHIOMOVE: Invalid argument /bin/chio -f /dev/ch0 move slot 0 drive 0 I know that the system (FreeBSD 4.2-release) is seeing the changer device fine. The only change has been made is an upgrade of amanda. -- Clem Kumah [EMAIL PROTECTED] System Administrator / Teamleader www.worldonline.co.uk World Online UK+44 (0) 870 748
