help getting started

[Max Waterman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I am located in China and, as far as I can tell, the amanda web site
(http://www.amanda.org/) is not accessible from here.
I am trying to get amanda running on a rh9 host, primarily using
http://www.harkness.co.uk/services/backup.html as my guide.
I seem to be running into some trouble though.
When I get to this part :
su amanda -c "/usr/sbin/amdump DailySet1"
it seems to return quite quickly without any messages at all.
I have tried these lines in my /etc/amanda/DailySet1/disklist file :
#localhost /etc comp-root-tar
#localhost /home comp-root-tar
localhost /dev/ida/c0d0p5 comp-root-tar
but it doesn't seem to make any difference; it still returns really
quickly :
# time su amanda -c "/usr/sbin/amdump DailySet1"
0.110u 0.070s 0:06.86 2.6%  0+0k 0+0io 5962pf+0w
#
Also, I've tried one or two of the other amanda commands, and I get
some errors about missing files. The first was a
/usr/local/lib/amanda/exclude.gtar file, and another is :
# time su amanda -c "/usr/sbin/amstatus DailySet1"
/var/lib/amanda/DailySet1/amflush: No such file or directory at
/usr/sbin/amstatus line 106.
0.160u 0.010s 0:00.42 40.4% 0+0k 0+0io 1245pf+0w
I just manually created the exclude file, but I am starting to think
that I didn't install correctly. This is what rpm says I have :
# rpm -qa | fgrep amanda
amanda-server-2.4.3-4
amanda-client-2.4.3-4
amanda-2.4.3-4
Can someone help, or point me to somewhere that helps (not on
www.amanda.org)?
Thanks.
Max.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFA2SVv4frZ3W7OHZMRArW8AJ0QX+uCPjLX2NPvUQk/BtDiwM9OmQCdFmeF
wMQixn9Kxi6mxTk3TUrvMCg=
=4g4+
-END PGP SIGNATURE-

Re: help getting started

[Paul Bijnens]

Max Waterman wrote:
When I get to this part :
su amanda -c "/usr/sbin/amdump DailySet1"
it seems to return quite quickly without any messages at all.

The normal way is to run "amcheck" first, fix everything
it complains about, before proceeding with a real amdump.
The error messages of amcheck are usually much less cryptic than
the other utilities.
If you got that far, and still have trouble, you can have a look
in /tmp/amanda, where each client has his logfiles.
And there is this mailing list, where you can post the exact
error messages you find.  Of course.
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: help getting started

[Max Waterman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Bijnens wrote:
| Max Waterman wrote:
|
|> When I get to this part :
|>
|> su amanda -c "/usr/sbin/amdump DailySet1"
|>
|> it seems to return quite quickly without any messages at all.
|
|
|
| The normal way is to run "amcheck" first, fix everything
| it complains about, before proceeding with a real amdump.
Yeah, that was the previous step on the step-by-step page I was using :
su amanda -c "/usr/sbin/amcheck DailySet1"
First time it ran, it complained about the
/usr/local/lib/amanda/exclude.gtar - there was no /usr/local/lib/amanda
directory, so I created it and the file.
After that, it ran fine; well, I think so - here's what it does now (but
I've done some later steps now) :
| # su amanda -c "/usr/sbin/amcheck DailySet1"
| Amanda Tape Server Host Check
| -
| Holding disk /var/tmp: 1122360 KB disk space available, that's plenty
| ERROR: cannot overwrite active tape DailySet101
|(expecting a new tape)
| NOTE: skipping tape-writable test
| NOTE: info dir /var/lib/amanda/DailySet1/curinfo/localhost/_home: does
not exist
| NOTE: index dir /var/lib/amanda/DailySet1/index/localhost/_home: does
not exist
| Server check took 4.220 seconds
|
| Amanda Backup Client Hosts Check
| 
| Client check: 1 host checked in 0.120 seconds, 0 problems found
|
| (brought to you by Amanda 2.4.3)
Spot anything?
|
| If you got that far, and still have trouble, you can have a look
| in /tmp/amanda, where each client has his logfiles.
Ooh. Many files. I guess I'll look in reverse chronological order.
| And there is this mailing list, where you can post the exact
| error messages you find.  Of course.
Hrm. Looking at those files, I don't see anything which looks like an
error. I have a few 'runtar' files which, I assume are an indication
that it is trying to make the backup, and there's nothing in them that
indicates an error.
Could it be that it's actually working, but doesn't give an output that
I am expecting - ie a list of files, or something that says 'I backed up
so many files' or even "I'm done".
Perhaps I am not specifying a valid place to backup. I am trying to
backup '/home' which is a mount point - though I did specify the
/dev/ida/c0d0p5 at one point, and /etc, but with the same effect.
How can I check that a backup completed successfully?
BTW, email isn't configured on the host machine - perhaps it is telling
me it is working, but I can't read what it says.
Max.
|
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFA2Tgx4frZ3W7OHZMRAjNAAJ9M6CjsRiFhZL0HVmZaPUiDun2SqgCfWhPm
5wGsSHbdUZ0P8dkevAsytEc=
=PQ/b
-END PGP SIGNATURE-

Re: help getting started

[Phil Homewood]

Max Waterman wrote:
> Could it be that it's actually working, but doesn't give an output that
> I am expecting - ie a list of files, or something that says 'I backed up
> so many files' or even "I'm done".

amdump doesn't generally (ever?) give output on stdout -- it
emails it.

> BTW, email isn't configured on the host machine - perhaps it is telling
> me it is working, but I can't read what it says.

That could well be the case. :-) Though it should spend
a noticeable amount of time telling you nothing before
it emails you.

Does "amadmin config info" tell you anything useful?
-- 
Phil Homewood, Systems Janitor, http://www.SnapGear.com
[EMAIL PROTECTED] Ph: +61 7 3435 2810 Fx: +61 7 3891 3630
SnapGear - A CyberGuard Company

Re: help getting started

[Max Waterman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Phil Homewood wrote:
| amdump doesn't generally (ever?) give output on stdout -- it
| emails it.
|
|
|>BTW, email isn't configured on the host machine - perhaps it is telling
|>me it is working, but I can't read what it says.
|
|
| That could well be the case. :-) Though it should spend
| a noticeable amount of time telling you nothing before
| it emails you.
Hrm - it does return pretty quickly :(
In any case, I should put more effort into getting mail working then;
thanks :)
|
| Does "amadmin config info" tell you anything useful?
# su amanda -c "/usr/sbin/amadmin DailySet1 info"
Current info for localhost /home:
~  Stats: dump rates (kps), Full:   -1.0,  -1.0,  -1.0
~Incremental:   -1.0,  -1.0,  -1.0
~  compressed size, Full: -100.0%,-100.0%,-100.0%
~Incremental: -100.0%,-100.0%,-100.0%
~  Dumps: lev datestmp  tape file   origK   compK secs
Make any sense?
Max.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFA2Xxr4frZ3W7OHZMRApAmAJ46iwAH35SR4PSPlKNatE1ePBdlOACbB0H8
jTI1rIV7bSA2LkttkOInc1s=
=MQZo
-END PGP SIGNATURE-

Re: help getting started

[Paul Bijnens]

Max Waterman wrote:
Phil Homewood wrote:
| amdump doesn't generally (ever?) give output on stdout -- it
| emails it.
You can generate the email by hand and safe it in a file instead:
  cd ~amanda/DailySet1
  amreport DailySet1 -l log.X.X -f save-mail-file
At least if you have file named log.datestamp.lvl.
There should also be a file named amdump.1 with a detailed
log of the serverside amdump program.
Any useful error message in that file?

# su amanda -c "/usr/sbin/amadmin DailySet1 info"
Current info for localhost /home:
~  Stats: dump rates (kps), Full:   -1.0,  -1.0,  -1.0
~Incremental:   -1.0,  -1.0,  -1.0
~  compressed size, Full: -100.0%,-100.0%,-100.0%
~Incremental: -100.0%,-100.0%,-100.0%
~  Dumps: lev datestmp  tape file   origK   compK secs
Make any sense?
Seems to me like amdump did nothing at all.
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: help getting started

[Max Waterman]

Paul Bijnens wrote:
You can generate the email by hand and safe it in a file instead:
  cd ~amanda/DailySet1
  amreport DailySet1 -l log.X.X -f save-mail-file

Ah, more interesting, but still difficult to figure out if anything has 
gone wrong.

At least if you have file named log.datestamp.lvl.
There should also be a file named amdump.1 with a detailed
log of the serverside amdump program.
Any useful error message in that file?
I am not really sure. Maybe this :
driver: result time 6.776 from taper: TAPE-ERROR [cannot overwrite 
active tape DailySet101]

I have put the other files here, if you care to have a look (I've made 
'.mail' files using the above 'amreport' command) :

http://reality.sgiweb.org/maxw/tmp/amanda/
I'm guessing I should concentrate on making email work...
Max.

# su amanda -c "/usr/sbin/amadmin DailySet1 info"
Current info for localhost /home:
~  Stats: dump rates (kps), Full:   -1.0,  -1.0,  -1.0
~Incremental:   -1.0,  -1.0,  -1.0
~  compressed size, Full: -100.0%,-100.0%,-100.0%
~Incremental: -100.0%,-100.0%,-100.0%
~  Dumps: lev datestmp  tape file   origK   compK secs
Make any sense?

Seems to me like amdump did nothing at all.

Re: help getting started

[Paul Bijnens]

Max Waterman wrote:
I am not really sure. Maybe this :
driver: result time 6.776 from taper: TAPE-ERROR [cannot overwrite 
active tape DailySet101]
It seems that the very first time you ran the backup for '/etc' it
worked fine. See:
  http://reality.sgiweb.org/maxw/tmp/amanda/log.20040623.0.mail )
This is how a normal run looks like.
For the next test, you seem to be using the same tape.
But you instructed to amanda ("tapecycle XXX") that she will not
overwrite an active tape.  So she cannot write to that tape.
She falls back to so called "degraded mode" backups dumping to
the holdingdisk.  You also have specified "reserve 100" (that's the
default if you didn't change it), meaning that in degraded mode,
100% of the holdingdisk is reserved for incremental backups.
But the only disklistentry was a new one, and amanda cannot
make an incremental backup when there is no full backup to base
it on.  Then she gives up.
Solution:  while testing, you can "amrmtape" the testing tape,
or you need to give more tapes (until you have "tapecycle" number
of tapes).
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: help getting started

[Max Waterman]

Thanks everyone for helping.
I got email working (albeit only locally since my isp is blocking my 
[chinese] IP address), so now I can get the reports.

I bought some more tapes, and have used the amrmtape command to start 
from scratch. Seems to be working like a charm. Next step is 'day 4' :)

http://www.harkness.co.uk/services/backup.html
Thanks again...until next time.
Max.
Paul Bijnens wrote:
Max Waterman wrote:
I am not really sure. Maybe this :
driver: result time 6.776 from taper: TAPE-ERROR [cannot overwrite 
active tape DailySet101]

It seems that the very first time you ran the backup for '/etc' it
worked fine. See:
  http://reality.sgiweb.org/maxw/tmp/amanda/log.20040623.0.mail )
This is how a normal run looks like.
For the next test, you seem to be using the same tape.
But you instructed to amanda ("tapecycle XXX") that she will not
overwrite an active tape.  So she cannot write to that tape.
She falls back to so called "degraded mode" backups dumping to
the holdingdisk.  You also have specified "reserve 100" (that's the
default if you didn't change it), meaning that in degraded mode,
100% of the holdingdisk is reserved for incremental backups.
But the only disklistentry was a new one, and amanda cannot
make an incremental backup when there is no full backup to base
it on.  Then she gives up.
Solution:  while testing, you can "amrmtape" the testing tape,
or you need to give more tapes (until you have "tapecycle" number
of tapes).

Re: help getting started

[Max Waterman]

Hi,
I'm getting some errors in my amstatus output - can anyone enlighten me?
# su amanda -c "/usr/sbin/amstatus DailySet1"
Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16 CST 2004
localhost:/etc   2  24k finished (16:46:53)
Use of uninitialized value in printf at /usr/sbin/amstatus line 760.
Use of uninitialized value in numeric ne (!=) at /usr/sbin/amstatus line 761.
localhost:/home  017720729k dumping0k (16:49:27)
Use of uninitialized value in addition (+) at /usr/sbin/amstatus line 770.
localhost:/usr/local 0   22701k finished (16:47:32)
localhost:/var   0  232660k wait for dumping driver: (aborted)
SUMMARY  part real estimated
  size  size
partition   :   4
estimated   :   4   17976128k
flush   :   00k
failed  :   0  0k   (  0.00%)
wait for dumping:   1 232660k   (  1.29%)
dumping to tape :   0  0k   (  0.00%)
dumping :   10k 17720729k (  0.00%) (  0.00%)
dumped  :   222725k22739k ( 99.94%) (  0.13%)
wait for writing:   00k0k (  0.00%) (  0.00%)
wait to flush   :   00k0k (100.00%) (  0.00%)
writing to tape :   00k0k (  0.00%) (  0.00%)
failed to tape  :   00k0k (  0.00%) (  0.00%)
taped   :   222725k22739k ( 99.94%) (  0.13%)
3 dumpers idle  : not-idle
taper writing, tapeq: 0
network free kps:   67
holding space   :   296960k (100.00%)
 dumper0 busy   :  0:02:17  ( 88.75%)
   taper busy   :  0:00:10  (  6.90%)
 0 dumpers busy :  0:00:17  ( 10.98%)  start-wait:  0:00:17  (100.00%)
 1 dumper busy  :  0:02:17  ( 89.02%)no-bandwidth:  0:01:49  ( 79.46%)
   start-wait:  0:00:28  ( 20.54%)
It seems to be taking ages to complete this backup...is something wrong?
Max.
Max Waterman wrote:
Thanks everyone for helping.
I got email working (albeit only locally since my isp is blocking my 
[chinese] IP address), so now I can get the reports.

I bought some more tapes, and have used the amrmtape command to start 
from scratch. Seems to be working like a charm. Next step is 'day 4' :)

http://www.harkness.co.uk/services/backup.html
Thanks again...until next time.
Max.
Paul Bijnens wrote:
Max Waterman wrote:
I am not really sure. Maybe this :
driver: result time 6.776 from taper: TAPE-ERROR [cannot overwrite 
active tape DailySet101]

It seems that the very first time you ran the backup for '/etc' it
worked fine. See:
  http://reality.sgiweb.org/maxw/tmp/amanda/log.20040623.0.mail )
This is how a normal run looks like.
For the next test, you seem to be using the same tape.
But you instructed to amanda ("tapecycle XXX") that she will not
overwrite an active tape.  So she cannot write to that tape.
She falls back to so called "degraded mode" backups dumping to
the holdingdisk.  You also have specified "reserve 100" (that's the
default if you didn't change it), meaning that in degraded mode,
100% of the holdingdisk is reserved for incremental backups.
But the only disklistentry was a new one, and amanda cannot
make an incremental backup when there is no full backup to base
it on.  Then she gives up.
Solution:  while testing, you can "amrmtape" the testing tape,
or you need to give more tapes (until you have "tapecycle" number
of tapes).

Re: help getting started

[Stefan G. Weichinger]

Hi, Max,

on Mittwoch, 30. Juni 2004 at 13:06 you wrote to amanda-users:

MW> Hi,

MW> I'm getting some errors in my amstatus output - can anyone enlighten me?

>> # su amanda -c "/usr/sbin/amstatus DailySet1"
>> Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16 CST 2004
>> 
>> localhost:/etc   2  24k finished (16:46:53)
>> Use of uninitialized value in printf at /usr/sbin/amstatus line 760.
>> Use of uninitialized value in numeric ne (!=) at /usr/sbin/amstatus line 761.
>> localhost:/home  017720729k dumping 0k (16:49:27)
>> Use of uninitialized value in addition (+) at /usr/sbin/amstatus line 770.
>> localhost:/usr/local 0   22701k finished (16:47:32)
>> localhost:/var   0  232660k wait for dumping driver: 
>> (aborted)

MW> It seems to be taking ages to complete this backup...is something wrong?

yes, it is. Those uninitialized values should not be there. A quick
look into amstatus seems to lead us to the estimates  

But the estimates seem to succeed.

What release of AMANDA is this, btw?

At first edit your disklist.

Don't use "localhost", use the full hostname instead.
Read docs/TOP-TEN-QUESTIONS for the explanation, if you want.

Stefan

Re: help getting started

[Paul Bijnens]

Max Waterman wrote:
I'm getting some errors in my amstatus output - can anyone enlighten me?
# su amanda -c "/usr/sbin/amstatus DailySet1"
Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16 CST 2004
localhost:/etc   2  24k finished (16:46:53)
Use of uninitialized value in printf at /usr/sbin/amstatus line 760.
This seems to be produced by amanda 2.4.3.  Not sure of the cause,
but just that part in the amstatus code (among others) got updated
in 2.4.4.  The most recent stable version is 2.4.4p3.
I guess upgrading will at least gid rid of that error message.
...
localhost:/home   0 17720729k dumping 0k (16:49:27)
(reformatted the above line, and added a space)
So this seems to be a 17 Gbyte partition.  And it seems to be
dumping directly to tape, bypassing the holdingdisk.
Am I correct?

It seems to be taking ages to complete this backup...is something wrong?
I don't know. How long is it running?  How long would it take
to write 17 Gbyte to tape at full speed?
Maybe your tapedrive is not streaming anymore, because the bypassing
of the holdingdisk?  If the datastream cannot follow the speed
of the tapedrive, the drive has to stop/backup a little/restart
and this has tremendous influence on the throughput (not to mention
the wear and tear on the drive and the cartridge!!!).
If this is indeed the case, possible solutions are:
- use a larger holdingdisk
- divide /home in smaller chunks, each fitting into the holdingdisk
  (best is fitting at least two of them: one taping and rest dumping)
  See docs/disklist in the sources for an example how to do this.
- disable software compression if cpu is the bottleneck
(amstatus in amanda 2.4.4p3 also has better indications of dumps
that bypass holdingdisk.  Another reason to upgrade.)
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: help getting started

[Max Waterman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Stefan G. Weichinger wrote:
| Hi, Max,
|
| on Mittwoch, 30. Juni 2004 at 13:06 you wrote to amanda-users:
|
| MW> Hi,
|
| MW> I'm getting some errors in my amstatus output - can anyone
enlighten me?
|
|
|>># su amanda -c "/usr/sbin/amstatus DailySet1"
|>>Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16 CST 2004
|>>
|>>localhost:/etc   2  24k finished (16:46:53)
|>>Use of uninitialized value in printf at /usr/sbin/amstatus line 760.
|>>Use of uninitialized value in numeric ne (!=) at /usr/sbin/amstatus
line 761.
|>>localhost:/home  017720729k dumping 0k
(16:49:27)
|>>Use of uninitialized value in addition (+) at /usr/sbin/amstatus line
770.
|>>localhost:/usr/local 0   22701k finished (16:47:32)
|>>localhost:/var   0  232660k wait for dumping
driver: (aborted)
|
|
| MW> It seems to be taking ages to complete this backup...is something
wrong?
|
| yes, it is. Those uninitialized values should not be there. A quick
| look into amstatus seems to lead us to the estimates  
|
| But the estimates seem to succeed.
|
| What release of AMANDA is this, btw?
2.4.3-4, it seems.
|
| At first edit your disklist.
|
| Don't use "localhost", use the full hostname instead.
OK, that's easy :)
| Read docs/TOP-TEN-QUESTIONS for the explanation, if you want.
Unfortunately, www.amanda.org is still not accessible from China :(
I guess I can try one of the anonymous proxys, but they also,
unsuprisingly, tend not to work very well from here either.
Is there a mirror somewhere, is the relevant info installed with amanda?
I see there's a fair bit of info. I'll dig some...esp. FAQ that's there.
Thanks.
Max.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFA4qur4frZ3W7OHZMRAvaCAJ91f8tCvD5nPYi7IQR9QTbi9VbbhACeOTvU
B3ncR50BIUykoH30KQS/faQ=
=btYK
-END PGP SIGNATURE-

Re: help getting started

[Stefan G. Weichinger]

Hi, Max,

on Mittwoch, 30. Juni 2004 at 14:01 you wrote to amanda-users:

MW> | Read docs/TOP-TEN-QUESTIONS for the explanation, if you want.

MW> Unfortunately, www.amanda.org is still not accessible from China :(

MW> I guess I can try one of the anonymous proxys, but they also,
MW> unsuprisingly, tend not to work very well from here either.

MW> Is there a mirror somewhere, is the relevant info installed with amanda?
MW> I see there's a fair bit of info. I'll dig some...esp. FAQ that's there.

This is ALPHA-stage when it comes to formatting, but the content is
the same as in the AMANDA-tarball:

http://www.oops.co.at/AMANDA-docs/index.html

There you can also find the top-ten ...
-- 
best regards,
Stefan

Stefan G. Weichinger
mailto:[EMAIL PROTECTED]

Re: help getting started

[Gene Heskett]

On Wednesday 30 June 2004 07:06, Max Waterman wrote:
>Hi,
>
>I'm getting some errors in my amstatus output - can anyone enlighten
> me?
>
>> # su amanda -c "/usr/sbin/amstatus DailySet1"
>> Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16
>> CST 2004
>>
>> localhost:/etc   2  24k finished

Please re-read the fAQ and the TOP-TEN-QUESTIONS in the docs dir, and 
then give your machine a unique name.  localhost is not unique.

>> (16:46:53) Use of uninitialized value in printf at
>> /usr/sbin/amstatus line 760. Use of uninitialized value in numeric
>> ne (!=) at /usr/sbin/amstatus line 761. localhost:/home   
>>   017720729k dumping0k (16:49:27) Use of
>> uninitialized value in addition (+) at /usr/sbin/amstatus line
>> 770. localhost:/usr/local 0   22701k finished
>> (16:47:32) localhost:/var   0  232660k
>> wait for dumping driver: (aborted)
>>
>> SUMMARY  part real estimated
>>   size  size
>> partition   :   4
>> estimated   :   4   17976128k
>> flush   :   00k
>> failed  :   0  0k   (  0.00%)
>> wait for dumping:   1 232660k   (  1.29%)
>> dumping to tape :   0  0k   (  0.00%)
>> dumping :   10k 17720729k (  0.00%) (  0.00%)
>> dumped  :   222725k22739k ( 99.94%) (  0.13%)
>> wait for writing:   00k0k (  0.00%) (  0.00%)
>> wait to flush   :   00k0k (100.00%) (  0.00%)
>> writing to tape :   00k0k (  0.00%) (  0.00%)
>> failed to tape  :   00k0k (  0.00%) (  0.00%)
>> taped   :   222725k22739k ( 99.94%) (  0.13%)
>> 3 dumpers idle  : not-idle
>> taper writing, tapeq: 0
>> network free kps:   67
>> holding space   :   296960k (100.00%)
>>  dumper0 busy   :  0:02:17  ( 88.75%)
>>taper busy   :  0:00:10  (  6.90%)
>>  0 dumpers busy :  0:00:17  ( 10.98%)  start-wait: 
>> 0:00:17  (100.00%) 1 dumper busy  :  0:02:17  ( 89.02%)   
>> no-bandwidth:  0:01:49  ( 79.46%) start-wait:  0:00:28  ( 20.54%)
>
>It seems to be taking ages to complete this backup...is something
> wrong?

When it is done, what speed is reported in the email the dump sends 
you?  By that, I mean does it match within say 10% of the drives 
rated speed of data transfer?  Also, note the holding disk area of 
only 296,960k, (297 megs) which is not nearly enough for amdump to 
stuff a 17,720,729k (18Gb? break that into subdirs and use tar, not 
dump if thats what you are now using) dump, and this is forceing 
amdump to write directly to the tape, possibly with lots of 
under-runs that will result in the drive stopping, backing up and 
restarting (shoe shining) which takes time out of the write time, so 
I'd expect the write speeds reported to be a bit slow.

So in addition to giving the machine a unique name that can at least 
be resolved in the /etc/hosts file, find it some more holding disk 
area.  The localhost issue probably won't effect the backups, but it 
sure will mung any chances of a clean recovery should you need it.
>
>Max.
>
>Max Waterman wrote:
>> Thanks everyone for helping.
>>
>> I got email working (albeit only locally since my isp is blocking
>> my [chinese] IP address), so now I can get the reports.
>>
>> I bought some more tapes, and have used the amrmtape command to
>> start from scratch. Seems to be working like a charm. Next step is
>> 'day 4' :)
>>
>> http://www.harkness.co.uk/services/backup.html
>>
>> Thanks again...until next time.
>>
>> Max.
>>
>> Paul Bijnens wrote:
>>> Max Waterman wrote:
 I am not really sure. Maybe this :

 driver: result time 6.776 from taper: TAPE-ERROR [cannot
 overwrite active tape DailySet101]
>>>
>>> It seems that the very first time you ran the backup for '/etc'
>>> it worked fine. See:
>>>   http://reality.sgiweb.org/maxw/tmp/amanda/log.20040623.0.mail )
>>>
>>> This is how a normal run looks like.
>>>
>>> For the next test, you seem to be using the same tape.
>>> But you instructed to amanda ("tapecycle XXX") that she will not
>>> overwrite an active tape.  So she cannot write to that tape.
>>>
>>> She falls back to so called "degraded mode" backups dumping to
>>> the holdingdisk.  You also have specified "reserve 100" (that's
>>> the default if you didn't change it), meaning that in degraded
>>> mode, 100% of the holdingdisk is reserved for incremental
>>> backups.
>>>
>>> But the only disklistentry was a new one, and amanda cannot
>>> make an incremental backup when there is no full backup to base
>>> it on.  Then she gives up.
>>>
>>> Solution:  while testing, you can "amrmtape" the testing tape,
>>> or you need to give more tapes (until you have "tapecycle" number
>>> of tapes).

-- 
Cheers, Gene
There are 4 boxes to be used in defense of liberty. 
Soap, ballot, jury, and ammo.
Ple

Re: help getting started

[Gene Heskett]

On Wednesday 30 June 2004 08:01, Max Waterman wrote:
>-BEGIN PGP SIGNED MESSAGE-
>Hash: SHA1
>
>Stefan G. Weichinger wrote:
>| Hi, Max,
>|
>| on Mittwoch, 30. Juni 2004 at 13:06 you wrote to amanda-users:
>|
>| MW> Hi,
>|
>| MW> I'm getting some errors in my amstatus output - can anyone
>
>enlighten me?
>
>|>># su amanda -c "/usr/sbin/amstatus DailySet1"
>|>>Using /var/lib/amanda/DailySet1/amdump from Wed Jun 30 16:46:16
>|>> CST 2004
>|>>
>|>>localhost:/etc   2  24k finished
>|>> (16:46:53) Use of uninitialized value in printf at
>|>> /usr/sbin/amstatus line 760. Use of uninitialized value in
>|>> numeric ne (!=) at /usr/sbin/amstatus
>
>line 761.
>
>|>>localhost:/home  017720729k dumping
>|>> 0k
>
>(16:49:27)
>
>|>>Use of uninitialized value in addition (+) at /usr/sbin/amstatus
>|>> line
>
>770.
>
>|>>localhost:/usr/local 0   22701k finished
>|>> (16:47:32) localhost:/var   0  232660k
>|>> wait for dumping
>
>driver: (aborted)
>
>| MW> It seems to be taking ages to complete this backup...is
>| something
>
>wrong?
>
>| yes, it is. Those uninitialized values should not be there. A
>| quick look into amstatus seems to lead us to the estimates  
>|
>| But the estimates seem to succeed.
>|
>| What release of AMANDA is this, btw?
>
>2.4.3-4, it seems.

Humm, would it be possible for one of us to email you a more recent 
tar.gz of say 2.4.5b1?  The last one I have on site is dated 
20040622.

I'd go on an education campaign to get access to www.amanda.org 
enabled, along with access to the site Jean-Louis has for new 
snapshots at .

Only by severe miss-information on the part of the chinese filtering 
folks would they set a block on those two sites.

>| At first edit your disklist.
>|
>| Don't use "localhost", use the full hostname instead.
>
>OK, that's easy :)
>
>| Read docs/TOP-TEN-QUESTIONS for the explanation, if you want.
>
>Unfortunately, www.amanda.org is still not accessible from China :(
>
>I guess I can try one of the anonymous proxys, but they also,
>unsuprisingly, tend not to work very well from here either.
>
>Is there a mirror somewhere, is the relevant info installed with
> amanda? I see there's a fair bit of info. I'll dig some...esp. FAQ
> that's there.
>
>Thanks.
>
>Max.
>-BEGIN PGP SIGNATURE-
>Version: GnuPG v1.2.4 (Darwin)
>Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
>
>iD8DBQFA4qur4frZ3W7OHZMRAvaCAJ91f8tCvD5nPYi7IQR9QTbi9VbbhACeOTvU
>B3ncR50BIUykoH30KQS/faQ=
>=btYK
>-END PGP SIGNATURE-

-- 
Cheers, Gene
There are 4 boxes to be used in defense of liberty. 
Soap, ballot, jury, and ammo.
Please use in that order, starting now.  -Ed Howdershelt, Author
Additions to this message made by Gene Heskett are Copyright 2004, 
Maurice E. Heskett, all rights reserved.

Need help getting started.

[Eric Webb]

Howdy folks.  To keep it short:

1) Can someone point me to online docs that I have seen allusions to in the 
archives but haven't found online yet?

2) amcheck reports the messages below.  Is there a requirement to label tapes 
prior to usage?  The tapes in my changer are used:

   amcheck-server: slot 0: not an amanda tape
   amcheck-server: slot 1: not an amanda tape
   amcheck-server: slot 2: not an amanda tape
   amcheck-server: slot 3: not an amanda tape
   amcheck-server: slot 4: not an amanda tape
   ERROR: new tape not found in rack
  (expecting a new tape)

3) If I attempt what I believe to be labelling a tape, I get:

   labeling tape in slot 0 (/dev/nst0):
   rewinding, reading label, not an amanda tape
   rewinding, writing label test00
   amlabel: writing label: tape is write-protected

4) My planner is failing and giving up, supposedly because it isn't receiving 
and responses on the estimates.  Shouldn't it calculate these on its own?  
I'm confused about how the estimate system works.

Amanda 2.4.3
Linux 2.4.18 / GCC 2.95.3
Adaptec 2940UW and HP 24/48 (DDS-2, 4GB 6-slot internal)
Using chg-scsi 


Thanks,
Eric

Re: Need help getting started.

[Frank Smith]

--On Saturday, November 09, 2002 23:20:07 -0500 Eric Webb <[EMAIL PROTECTED]> wrote:

> 
> Howdy folks.  To keep it short:
> 
> 1) Can someone point me to online docs that I have seen allusions to in the 
> archives but haven't found online yet?

www.amanda.org  Useful links near the bottom include the FAQ-O-Matic
(occasionally referred to as the FOM in this group) and "the chapter"
from Unix Backup & Recovery.  Start with "the chapter", it has both
a good overview and a lot of detail.

> 2) amcheck reports the messages below.  Is there a requirement to label tapes 
> prior to usage?  The tapes in my changer are used:
> 
>amcheck-server: slot 0: not an amanda tape
>amcheck-server: slot 1: not an amanda tape
>amcheck-server: slot 2: not an amanda tape
>amcheck-server: slot 3: not an amanda tape
>amcheck-server: slot 4: not an amanda tape
>ERROR: new tape not found in rack
>   (expecting a new tape)
 
Yes, you need to label the tapes first before Amanda will
use them (it prevents you from accidently overwriting some
other tape left in the drive).

> 3) If I attempt what I believe to be labelling a tape, I get:
> 
>labeling tape in slot 0 (/dev/nst0):
>rewinding, reading label, not an amanda tape
>rewinding, writing label test00
>amlabel: writing label: tape is write-protected
 
Tape is write-protected.  Check the write-protect tab on the tape.
An outside possibility is that the Amanda user doesn't have write
permissions on the device, but I think that would give you a
different error. 

> 4) My planner is failing and giving up, supposedly because it isn't receiving 
> and responses on the estimates.  Shouldn't it calculate these on its own?  
> I'm confused about how the estimate system works.

Planner runs on the Amanda server and determines what needs
to be backed up at what level in order to use approximately
the same amount of tape each day while still honoring your
dumpcycle, runspercycle, runtapes, etc.  In order to do that
it collects estimates of how much space would be needed for
different levels of dumps from each client.  Basically, it
does a dump to /dev/null and counts the bytes.  If you have
large filesystems you may need to increase etimeout in your
config.  If there are other problems, post the exact error
messages so the group will be better able to help pinpoint
the problem(s).

Frank

> Amanda 2.4.3
> Linux 2.4.18 / GCC 2.95.3
> Adaptec 2940UW and HP 24/48 (DDS-2, 4GB 6-slot internal)
> Using chg-scsi 
> 
> 
> Thanks,
> Eric



--
Frank Smith[EMAIL PROTECTED]
Systems Administrator Voice: 512-374-4673
Hoover's Online Fax: 512-374-4501

Re: Need help getting started.

[Galen Johnson]

Frank Smith wrote:


--On Saturday, November 09, 2002 23:20:07 -0500 Eric Webb <[EMAIL PROTECTED]> wrote:


3) If I attempt what I believe to be labelling a tape, I get:

  labeling tape in slot 0 (/dev/nst0):
  rewinding, reading label, not an amanda tape
  rewinding, writing label test00
  amlabel: writing label: tape is write-protected
   


Tape is write-protected.  Check the write-protect tab on the tape.
An outside possibility is that the Amanda user doesn't have write
permissions on the device, but I think that would give you a
different error. 

Actually, it gives the same error.  When I first set up I didn't put 
amanda in the correct group and got that error on a regular, now-wrote 
protected tape.

=G=

Re: Need help getting started.

[Eric Webb]

Okay, tape labelling issues were related to device node permissions.  Silly 
me.

Still having the issue with the planner.  amcheck claims it can't find the 
DUMP program, despite it being installed already and detected by amanda's 
configure (/usr/local/sbin/dump).

amdump logs indicate problem getting results back on the estimates.  I presume 
this is because DUMP isn't being found, though I've tried it as GNUTAR with 
no luck either.

--- log --
   FAIL planner alexandria hda3 20021110 0 [missing result for hda3 in 
alexandria response]
-- amdump --
   error result for host alexandria disk hda3: missing estimate
--

Another question:  How do I specify device nodes for filesystems running under 
LVM in linux?  These are mounted from /dev/vgname/lvname... do I just strip 
off the /dev and give "vgname/lvname" to Amanda?

> > Amanda 2.4.3
> > Linux 2.4.18 / GCC 2.95.3
> > Adaptec 2940UW and HP 24/48 (DDS-2, 4GB 6-slot internal)
> > Using chg-scsi

Thanks,
Eric

Re: Need help getting started.

[Galen Johnson]

Eric Webb wrote:


Okay, tape labelling issues were related to device node permissions.  Silly 
me.

Still having the issue with the planner.  amcheck claims it can't find the 
DUMP program, despite it being installed already and detected by amanda's 
configure (/usr/local/sbin/dump).

amdump logs indicate problem getting results back on the estimates.  I presume 
this is because DUMP isn't being found, though I've tried it as GNUTAR with 
no luck either.

--- log --
  FAIL planner alexandria hda3 20021110 0 [missing result for hda3 in 
alexandria response]
-- amdump --
  error result for host alexandria disk hda3: missing estimate
--
 

Make sure your path is correct.  If you run as root and su to amanda to 
run the checks, you're only getting root's path. 'su - amanda' and 
'export $PATH' to verify that you should be seeing the correct path.  

Another question:  How do I specify device nodes for filesystems running under 
LVM in linux?  These are mounted from /dev/vgname/lvname... do I just strip 
off the /dev and give "vgname/lvname" to Amanda?
 

what are they mounted to?  Use the mount point itself .

=G=

Re: Need help getting started.

[Eric Webb]

On Sunday 10 November 2002 03:30 pm, Galen Johnson wrote:
> >
> >Still having the issue with the planner.  amcheck claims it can't find the
> >DUMP program, despite it being installed already and detected by amanda's
> >configure (/usr/local/sbin/dump).
> >
> >amdump logs indicate problem getting results back on the estimates.  I
> > presume this is because DUMP isn't being found, though I've tried it as
> > GNUTAR with no luck either.
> >
> >--- log --
> >   FAIL planner alexandria hda3 20021110 0 [missing result for hda3 in
> >alexandria response]
> >-- amdump --
> >   error result for host alexandria disk hda3: missing estimate
> >--
>
> Make sure your path is correct.  If you run as root and su to amanda to
> run the checks, you're only getting root's path. 'su - amanda' and
> 'export $PATH' to verify that you should be seeing the correct path.

I'm running this logged into user "amanda" directly.  The only time root was 
involved was at compile/install time.  User "amanda" has /usr/local/bin and 
/usr/local/sbin in path.  What specifically is it looking for?

> >Another question:  How do I specify device nodes for filesystems running
> > under LVM in linux?  These are mounted from /dev/vgname/lvname... do I
> > just strip off the /dev and give "vgname/lvname" to Amanda?
>
> what are they mounted to?  Use the mount point itself .

Well, hang on -- I thought you were supposed to supply device nodes, not 
filesystem mount points.  All the examples were things like "sda1" which is 
the device node minus the "/dev" pathing.

-E.

Re: Need help getting started.

[Galen Johnson]

Eric Webb wrote:


On Sunday 10 November 2002 03:30 pm, Galen Johnson wrote:
 

Still having the issue with the planner.  amcheck claims it can't find the
DUMP program, despite it being installed already and detected by amanda's
configure (/usr/local/sbin/dump).

amdump logs indicate problem getting results back on the estimates.  I
presume this is because DUMP isn't being found, though I've tried it as
GNUTAR with no luck either.

--- log --
 FAIL planner alexandria hda3 20021110 0 [missing result for hda3 in
alexandria response]
-- amdump --
 error result for host alexandria disk hda3: missing estimate
--
 

Make sure your path is correct.  If you run as root and su to amanda to
run the checks, you're only getting root's path. 'su - amanda' and
'export $PATH' to verify that you should be seeing the correct path.
   


I'm running this logged into user "amanda" directly.  The only time root was 
involved was at compile/install time.  User "amanda" has /usr/local/bin and 
/usr/local/sbin in path.  What specifically is it looking for?

What do your debug logs have to say?


Another question:  How do I specify device nodes for filesystems running
under LVM in linux?  These are mounted from /dev/vgname/lvname... do I
just strip off the /dev and give "vgname/lvname" to Amanda?
 

what are they mounted to?  Use the mount point itself .
   


Well, hang on -- I thought you were supposed to supply device nodes, not 
filesystem mount points.  All the examples were things like "sda1" which is 
the device node minus the "/dev" pathing.

-E.
 

hI think that the example disklist is misleading...however, if 
you'll look further down into it you'll see that it is looking for the 
_partitions_ themselves not the device nodes.I believe the Maryland 
folks have a strange naming scheme if this is truly what they name their 
partitions.

If you are indeed point to something other than the partitions then this 
could explain why you're timing out on the planner.  Try pointing to the 
partition not the device node and see if it starts being happy for you.

=G=

Fwd: Re: Need help getting started.

[Eric Webb]

On Sunday 10 November 2002 07:32 pm, Galen Johnson wrote:
> Eric Webb wrote:
> >On Sunday 10 November 2002 03:30 pm, Galen Johnson wrote:
> >>>Still having the issue with the planner.  amcheck claims it can't find
> >>> the DUMP program, despite it being installed already and detected by
> >>> amanda's configure (/usr/local/sbin/dump).

Yeah, I'm about 90% sure that my problems are with the "dump" program.
amcheck consistently tells me "DUMP program not available", which is echoed
by the "sendsize" program's debug log ("sendsize[15363]: time 0.027: no dump
program available").

I did a make distclean, configure found the path to the dump binary.  "dump"
that I used was found via search on sourceforge -- dump-0.4b31, which
compiled and installed into /usr/local/sbin.  I've been browsing sendsize.c
and it seems that the message I'm getting may be because of the DUMP define
not beind set (but the #if nesting is losing me at this hour).

configure's output shows "checking for dump... /usr/local/sbin/dump", which
makes me believe it finds it and sets it, but the line "#define DUMP $dump"
is never added to confdefs.h during the configure process.

I'm confused.  I must be missing something obvious.

-E.

Re: Fwd: Re: Need help getting started.

[Jon LaBadie]

On Wed, Nov 13, 2002 at 04:17:13AM +, Eric Webb wrote:
> On Sunday 10 November 2002 07:32 pm, Galen Johnson wrote:
> > Eric Webb wrote:
> > >On Sunday 10 November 2002 03:30 pm, Galen Johnson wrote:
> > >>>Still having the issue with the planner.  amcheck claims it can't find
> > >>> the DUMP program, despite it being installed already and detected by
> > >>> amanda's configure (/usr/local/sbin/dump).
> 
> Yeah, I'm about 90% sure that my problems are with the "dump" program.
> amcheck consistently tells me "DUMP program not available", which is echoed
> by the "sendsize" program's debug log ("sendsize[15363]: time 0.027: no dump
> program available").
> 
> I did a make distclean, configure found the path to the dump binary.  "dump"
> that I used was found via search on sourceforge -- dump-0.4b31, which
> compiled and installed into /usr/local/sbin.  I've been browsing sendsize.c
> and it seems that the message I'm getting may be because of the DUMP define
> not beind set (but the #if nesting is losing me at this hour).
> 
> configure's output shows "checking for dump... /usr/local/sbin/dump", which
> makes me believe it finds it and sets it, but the line "#define DUMP $dump"
> is never added to confdefs.h during the configure process.

I don't see a confdefs.h on my system, neither 2.4.2 nor 2.4.3b4.  May have
missed it, where is it located.  In both my release trees the "#define DUMP"
line gets added to /config/config.h.  Do you have that file
and definition?

One way to look at the code with all the #defines made is to compile with the
"stop after the preprocessor" options.  I just tried the following on my system
in the client-src directory and created the intermediate file "sendsize.i".

  rm sendsize.o
  AM_CFLAGS='-E -C' make sendsize.o > sendsize.i

This will have lots of *.h stuff at the start (17000 lines on my system),
but the last 1500 lines or so will be from sendsize.c with #if stuff removed
(left as blank lines) and #define'ed macros substituted by their definitions.

-- 
Jon H. LaBadie  [EMAIL PROTECTED]
 JG Computing
 4455 Province Line Road(609) 252-0159
 Princeton, NJ  08540-4322  (609) 683-7220 (fax)