Re: [Bacula-users] Overland-Tandberg TD-LTO9xSA LTO9HH SAS External Tape Drive Kit

[Cejka Rudolf via Bacula-users]

Rob Gerber wrote (2024/02/29):
> For a changer, which I think you will not have with your system, it mainly
> needs to work with the mtx command (I think same command is available on
> bsd?).

Hello, native is chio. There are some examples in examples/autochangers
in sources, where I do use chio-changer-freebsd.

-- 
Rudolf Cejka  https://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 1/2, 612 00  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO tape performances, again...

[Cejka Rudolf via Bacula-users]

Hello, you need all of these:

In job director section:

- Spool Data = yes, attribute spooling is automatically enabled too

In device storage section:

- Maximum File Size = 32 GB, for LTO-6 and older you can think about 16 GB

- Sufficient Maximum Job Spool Size or Maximum Spool Size, so overall write 
speed
  could be effective, I use 200 GB and 1000 GB

- Maximum Block Size = 262144 or 131072, for LTO-6 and older 65536 (this is 
tested
  on my FreeBSD system, Linux or another systems may need higher values - but 
do not
  use too big, because probability of write errors increases)

- SSD based disk spooling, M2 or SATA RAID

I finished write speed tests with different file sizes, so here they are
(+ is my recommended file size, - is alternative and I expect that LTO-9
is still similar to LTO-8 and LTO-7 in this view):

LTO-5, block size 65536:

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file
 (at 140 MB/s)
  1 GB  6:44:1861.8 MB/s   1498688192512  149900:00:07
  2 GB  4:51:3085.7 MB/s   1498737278976   75000:00:14
  4 GB  3:54:57   106.3 MB/s   1498762641408   37500:00:29
  8 GB  3:24:22   122.2 MB/s   1499224932352   18800:00:57
 16 GB +3:11:09   130.7 MB/s   14994564055049400:01:54
 32 GB -3:04:35   135.4 MB/s   14996903034884700:03:49
 64 GB  3:01:17   137.8 MB/s   14995749601282400:07:37
128 GB  2:59:37   139.1 MB/s   14996036648961200:15:14

LTO-6, block size 65536:

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file
 (at 160 MB/s)
  1 GB  9:51:3870.4 MB/s   2500322394112  250100:00:06
  2 GB  7:05:4597.9 MB/s   2500404314112  125100:00:13
  4 GB  5:42:29   121.6 MB/s   2500445274112   62600:00:25
  8 GB  4:56:39   140.5 MB/s   2501218009088   31300:00:50
 16 GB +4:37:28   150.2 MB/s   2501603885056   15700:01:40
 32 GB -4:27:55   155.6 MB/s   25016084725767900:03:20
 64 GB  4:24:19   157.7 MB/s   25018018037764000:06:40
128 GB  4:20:58   159.7 MB/s   25019952660482000:13:20
256 GB  4:19:43   160.5 MB/s   25017795215361000:26:40

LTO-7, block size 262144 (65536 started to be limiting):

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file
 (at 300 MB/s)
  2 GB  9:17:20   179.8 MB/s   6013258301440  300700:00:07
  4 GB  7:25:39   224.8 MB/s   6013256204288  150400:00:13
  8 GB  6:30:19   256.9 MB/s   6017229258752   75300:00:27
 16 GB  6:02:51   276.3 MB/s   6015444844544   37600:00:53
 32 GB +5:48:56   287.4 MB/s   6017326514176   18900:01:47
 64 GB -5:42:07   293.1 MB/s   60173259898889500:03:33
128 GB  5:38:45   296.0 MB/s   60171076239364800:07:07
256 GB  5:36:56   297.6 MB/s   60174560133122400:14:13
512 GB  5:36:12   298.3 MB/s   60174557511681200:28:27

LTO-8, block size 262144:

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file
 (at 300 MB/s)
  2 GB 17:59:35   185.3 MB/s  12005379407872  600400:00:07
  4 GB 14:34:50   228.7 MB/s  12005942231040  300200:00:13
  8 GB 12:52:26   259.2 MB/s  12013881786368  150200:00:27
 16 GB 12:01:17   277.5 MB/s  12010304307200   75100:00:53
 32 GB +   11:35:37   287.8 MB/s  12012189122560   37600:01:47
 64 GB -   11:26:04   291.8 MB/s  12014075510784   18800:03:33
128 GB 11:16:23   296.0 MB/s  120145743708169400:07:07
256 GB 11:13:14   297.4 MB/s  120143392276484700:14:13
512 GB 11:11:39   298.1 MB/s  120131572203522400:28:27

Pierre Bernhardt wrote (2024/01/25):
> Am 25.01.24 um 10:06 schrieb Marco Gaiarin:
> > 
> > > 2) checked disk performance (data came only from local disk); i've 
> > > currently
> > >   3 servers, some perform better, some worster, but the best one have a 
> > > read
> > > disk performance pretty decent, at least 200MB/s on random access (1500 
> > > MB/s
> > > on sequential one).
> > 
> > Jim Pollard on private email ask me about controllers: i've not specified,
> > sorry, but LTO units are 

Re: [Bacula-users] Tape Libraries

[Cejka Rudolf via Bacula-users]

IT Manager via Bacula-users wrote (2023/06/26):
> I'm looking at replacing 4 different backup systems with a single solution 
> that can backup
> Windows (server and desktop), Linux and Unix. Bacula is currently very high 
> up the list but
> there's one make or break requirement, which is tape backups.
> Due to the number of systems we need to backup, we need a decent sized tape 
> library (40+ tapes)
> that can handle 3 or 4 tape drives and LTO 6 tapes.
> 
> Can anyone recommend suitable hardware that will work with Bacula without too 
> much swearing and
> cursing being required?

Hello,
  new tape library with LTO-6 can be hard to buy today. Maybe LTO-7 drives would
be the way to go, or look for something used, or rather switch to LTO-9 or 
LTO-8,
because raw data capacity is pretty cheap and it is nice to have 5 to 7 times
bigger data capacity using the same number of slots and 2 times better speed. 
Further,
3 or 4 tape drives needs something from midrange rather than entry level. There 
are
many products. I expect that all of them or atleast most of them would be just 
to
buy, install, connect, use and pay support ;o) You can look at HP StoreEver 
MSL3040,
Quantum Scalar i3, Overland Tandberg NEOxl 40, Qualstar Q48 Autoloader or Q40 
Library,
StorageTek SL150 (my case), and certainly more.

-- 
Rudolf Cejka  https://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 1/2, 612 00  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Slow Spooling / Despooling - Solved

[Cejka Rudolf via Bacula-users]

Dr. Thorsten Brandau wrote (2023/05/26):
> Understanding the parameter helps. Thanks to UDO I could dramatically
> increase the despooling speed.
> 
> MaximumFileSize MUST be increased at last with tapes like an LTO. So I have
> set
> 
> MaximumFileSize = 54GB
> 
> An went from 100-120 MB/s to 260 MB/s.
> 
> The default of MaximumFileSize = 1GB, which is too small for LTO drives as
> it seems to get their potential.

Hello, I have noticed this discussion just now.

I did some tests in the past for LTO-3 and LTO-5 and I have also
prepared LTO-6, LTO-7, LTO-M8 and LTO-8 (LTO-7 in LTO-8 drive), but
forgot to finish them :o( Below are results atleast for LTO-5. I think
that you can multiply them proportionally for newer LTO generations.

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file
 (at 140 MB/s)
  1 GB  6:44:1861.8 MB/s   1498688192512  149900:00:07
  2 GB  4:51:3085.7 MB/s   1498737278976   75000:00:14
  4 GB  3:54:57   106.3 MB/s   1498762641408   37500:00:29
  8 GB  3:24:22   122.2 MB/s   1499224932352   18800:00:57
 16 GB  3:11:09   130.7 MB/s   14994564055049400:01:54
 32 GB  3:04:35   135.4 MB/s   14996903034884700:03:49
 64 GB  3:01:17   137.8 MB/s   14995749601282400:07:37
128 GB  2:59:37   139.1 MB/s   14996036648961200:15:14

And I'm going to return to these tests right now...

-- 
Rudolf Cejka  https://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 1/2, 612 00  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Windows backup slow

[Cejka Rudolf via Bacula-users]

Hello,
  just small FYI, after switching from 9.x.x to 11.x.x it seems that
speed problem has been fixed.

Cejka Rudolf wrote (2020/04/30):
> Josh Fisher wrote (2020/04/23):
> > On 4/22/2020 12:23 PM, Cejka Rudolf wrote:
> > > Hello,
> > >I have exactly the same problem: Too slow filesystem traversal by
> > > Bacula Windows Client. I think that it has to be some problem with
> > > Bacula Client cross compilation (low level of compiler optimalization?),
> > > some setting or some other little thing, because if I switch the
> > > client to Bareos, incremental backup is usually two to three times
> > > faster than with Bacula client in my environment. I tried many
> > > versions of clients including the oldest clients from times, when
> > > projects "splitted", but behavior of all versions has been the
> > > same: All Bacula clients were two to three times slower than all
> > > Bareos clients for incremental backups in my environment. Very
> > > disappointing and I did not find anything yet, what could change that.
> > 
> > Are you using the same Maximum Network Buffer size on both? People have
> > reported issues in the past with the default 64k Maximum Network Buffer Size
> > on Windows. Try reducing that to 32k.
> 
> Yes, I tried both. Default and 32768. With no differences for Bacula Windows
> Client for incremental backup with little data.

-- 
Rudolf Cejka  https://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 1/2, 612 00  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Windows backup slow

[Cejka Rudolf]

Josh Fisher wrote (2020/04/23):
> On 4/22/2020 12:23 PM, Cejka Rudolf wrote:
> > Hello,
> >I have exactly the same problem: Too slow filesystem traversal by
> > Bacula Windows Client. I think that it has to be some problem with
> > Bacula Client cross compilation (low level of compiler optimalization?),
> > some setting or some other little thing, because if I switch the
> > client to Bareos, incremental backup is usually two to three times
> > faster than with Bacula client in my environment. I tried many
> > versions of clients including the oldest clients from times, when
> > projects "splitted", but behavior of all versions has been the
> > same: All Bacula clients were two to three times slower than all
> > Bareos clients for incremental backups in my environment. Very
> > disappointing and I did not find anything yet, what could change that.
> 
> Are you using the same Maximum Network Buffer size on both? People have
> reported issues in the past with the default 64k Maximum Network Buffer Size
> on Windows. Try reducing that to 32k.

Yes, I tried both. Default and 32768. With no differences for Bacula Windows
Client for incremental backup with little data.

-- 
Rudolf Cejka  http://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Windows backup slow

[Cejka Rudolf]

Hello,
  I have exactly the same problem: Too slow filesystem traversal by
Bacula Windows Client. I think that it has to be some problem with
Bacula Client cross compilation (low level of compiler optimalization?),
some setting or some other little thing, because if I switch the
client to Bareos, incremental backup is usually two to three times
faster than with Bacula client in my environment. I tried many
versions of clients including the oldest clients from times, when
projects "splitted", but behavior of all versions has been the
same: All Bacula clients were two to three times slower than all
Bareos clients for incremental backups in my environment. Very
disappointing and I did not find anything yet, what could change that.

I have two Windows servers. The first one is doable using Bacula Windows
Client, but on the second one, I had to switch to Bareos Windows Client,
because Bacula Windows Client did not want to finish under 20 hours,
while Bareos Windows Client needs just 8 to 9 hours.

And I really do not know, what to do after a year, two or three...

backup-dir Version: 9.4.4 (28 May 2019) x86_64-unknown-freebsd11.3 freebsd 
11.3-STABLE

bacula-fd Version: 9.4.4 (28 May 2019)  VSS Linux Cross-compile Win64

bareos-fd Version: 17.2.4 (21 Sep 2017)  VSS Linux Cross-compile Win64
(maybe the last working client with Bacula server?)

Best regards.

Andrew Watkins wrote (2020/04/22):
> 
> Thanks folks,
> 
> I have a feeling it's nothing to do with the network, but the speed of
> bacula searching the filesystem.
> 
> In 1 minute it only searched 11k (Examined=67,235 & 60seconds later
> Examined=72,873)
> 
> Any ideas except get a better server?
> and the following seems to be showing that there is only 1 process scanning
> the files?
> 
> *stat client=winfs-fd
> Connecting to Client winfs-fd at winfs.dcs.bbk.ac.uk:9102
> winfs-fd Version: 9.6.3 (09 March 2020)  VSS Linux Cross-compile Win64
> Daemon started 22-Apr-20 15:02. Jobs: run=0 running=1.
> Microsoft Standard Edition (build 9200), 64-bit
> Priv 0x22f
> Memory: WorkingSetSize: 23,273,472 QuotaPagedPoolUsage: 259,456
> QuotaNonPagedPoolUsage: 19,896 PagefileUsage: 12,668,928
> APIs=OPT,ATP,LPV,CFA,CFW,
>  WUL,WMKD,GFAA,GFAW,GFAEA,GFAEW,SFAA,SFAW,BR,BW,SPSP,
>  WC2MB,MB2WC,FFFA,FFFW,FNFA,FNFW,SCDA,SCDW,
>  GCDA,GCDW,GVPNW,GVNFVMPW,LZO,!EFS
>  Heap: heap=23,273,472 smbytes=399,505 max_bytes=425,467 bufs=296
> max_bufs=353
>  Sizes: boffset_t=8 size_t=8 debug=100 trace=1 mode=0,0 bwlimit=0kB/s
> 
> Running Jobs:
> JobId 155 Job BackupWINFS.2020-04-22_14.10.06_10 is running.
>     VSS Incremental Backup Job started: 22-Apr-20 15:10
>     Files=2 Bytes=1,829 AveBytes/sec=2 LastBytes/sec=0 Errors=0
>     Bwlimit=0 ReadBytes=1,829
>     Files: Examined=67,235 Backed up=2
>     Processing file: G:/home/abella05/Desktop
>     SDReadSeqNo=6 fd=1300 SDtls=0
> Director connected at: 22-Apr-20 15:23
> 

-- 
Rudolf Cejka  http://www.fit.vut.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Problems getting a tape library to work under FreeBSD

[Cejka Rudolf]

Kenneth Garges wrote (2018/02/01):
> I tried chio but gave up on it because anything I did gave an
> "Inappropriate ioctl for device" error. Now that a reboot seems to
> have gotten mtx working I found that chio works too. 

Ok, mtx works just once after reboot. Some delay helps, or not?
And chio, is it the same and after reboot it works just once too?
It seems that it is in a command, which your library does not like
very much. Could you search if there is any library firmware update?

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Problems getting a tape library to work under FreeBSD

[Cejka Rudolf]

Kenneth Garges wrote (2018/01/30):
> I???m having trouble getting a tape library to work.

Hello, try chio status -v or mtx status. Both should return
status of your autochanger including tape labels. If they both
do return an error, do you have correct permissons on /dev/ch*?
I have the following (please note the operator group):

bacula> ls -l /dev/ch*
crw-rw  1 root  operator  0x9e 11 led 11:31 /dev/ch0
bacula> id bacula
uid=366(bacula) gid=366(bacula) groups=366(bacula),0(wheel),5(operator)

If mtx returns an error but chio does work, please try chio-changer-freebsd,
which uses FreeBSD's native chio instead of mtx from ports.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bytes/block=64,512 for an LT04

[Cejka Rudolf]

Dan Langille wrote (2017/12/21):
> Hello,
> 
> I'm using Bacula 7.4.7 on FreeBSD 11
> 
> What Bytes/block are you using for your LTO4 tapes?

Oops, I waited for the others and now I see that I'm waiting too much :o)

I still do use Maximum Block Size = 65536 and I never had the problem.
The big win is that 64 KB (yes, somewhat smaller would be even safer...)
is portable and you will not have problem read it anywhere. And FreeBSD does
not have a problem with this. I can feed LTO-5 with >= 250 MB/s. It just works
and currently I do not have any reason to increase it. I just seen problems
with such small blocks on Linux until now. But try it yourself, somebody else
can have different experiences.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Configuring autochanger with SAS LTO-5 drives

[Cejka Rudolf]

Cejka Rudolf wrote (2017/06/05):
> Cejka Rudolf wrote (2017/06/02):
> > Ivan Adzhubey wrote (2017/06/01):
> > > b) What is the effect of MaximumFileSize option and what would be its 
> > > optimal 
> > > value for my IBM LTO-5 SAS drives? I have used 8GB value found in one of 
> > > the 
> > > list posts, while the documentation suggests 2GB for LTO-4. But even set 
> > > at 
> > > 8GB this would create lots of EOF marks on a 1.5TB tape, do we really 
> > > need so 
> > > many?
> > 
> > Hi, I do use 16 GB. Every EOF mark means around 3 seconds delay. So if you 
> > have
> > over 200 files on the tape using 8 GB, it is around 10 minutes extra per 
> > tape.
> 
> Hi, small fix. It seems that it is even around 5-6 seconds delay resulting in
> extra 20 minutes per tape. And worse, I'm afraid that the tape drive do not
> stop nor slowdown tape movement, which would mean over 10 % of the tape lost
> because of filemarks. Hope that I'm wrong.

Hi, I finally have done and finished real testing with Maximum File Size option.
Tested on LTO-6 drive with LTO-5 tape, so the results are relevant for LTO-5,
used btape utility and fill s (simplified) command.

Fortunately, I has negligible impact on tape capacity, but unfortunately, I has
really dramatical impact on overall write speed and you have to count not just 
5-6
seconds, but better 8-10 seconds per file. I also added time to read one file at
140 MB/s, which is important for seeking during restores.

  File   Simplified   Simplified   Bytes Files   Rough time to
  sizefill timefill rate written   written   read one file

  1 GB  6:44:1861.8 MB/s   1498688192512  1499   00:00:07
  2 GB  4:51:3085.7 MB/s   1498737278976   750   00:00:14
  4 GB  3:54:57   106.3 MB/s   1498762641408   375   00:00:29
  8 GB  3:24:22   122.2 MB/s   1499224932352   188   00:00:57
 16 GB  3:11:09   130.7 MB/s   149945640550494   00:01:54
 32 GB  3:04:35   135.4 MB/s   149969030348847   00:03:49
 64 GB  3:01:17   137.8 MB/s   149957496012824   00:07:37
128 GB  2:59:37   139.1 MB/s   149960366489612   00:15:14

And it seems that the results are perfectly reproducible. Three successive
tests with 16 GB file size have had exactly the same fill time 3:11:09!
Tests were done by mistake, when I wondered how it is possible that the
time stays exactly the same :o)

Regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Configuring autochanger with SAS LTO-5 drives

[Cejka Rudolf]

Alan Brown wrote (2017/06/08):
> LTO5 drives contain at least 1GB of ram which is used to buffer what's 
> coming from the computer and feed the internal compression/encryption 
> engines, then buffer the output of those sections to the actual tape itself.

Hello, please, which ones? I just found LTO5 drives with 256 MB from HP
and 512 MB from IBM, but no one with 1 GB.

with 1 GB, I have found LTO6 from IBM or LTO7 from both HP and IBM.

Regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Configuring autochanger with SAS LTO-5 drives

[Cejka Rudolf]

Cejka Rudolf wrote (2017/06/02):
> Ivan Adzhubey wrote (2017/06/01):
> > b) What is the effect of MaximumFileSize option and what would be its 
> > optimal 
> > value for my IBM LTO-5 SAS drives? I have used 8GB value found in one of 
> > the 
> > list posts, while the documentation suggests 2GB for LTO-4. But even set at 
> > 8GB this would create lots of EOF marks on a 1.5TB tape, do we really need 
> > so 
> > many?
> 
> Hi, I do use 16 GB. Every EOF mark means around 3 seconds delay. So if you 
> have
> over 200 files on the tape using 8 GB, it is around 10 minutes extra per tape.

Hi, small fix. It seems that it is even around 5-6 seconds delay resulting in
extra 20 minutes per tape. And worse, I'm afraid that the tape drive do not
stop nor slowdown tape movement, which would mean over 10 % of the tape lost
because of filemarks. Hope that I'm wrong.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Configuring autochanger with SAS LTO-5 drives

[Cejka Rudolf]

Ivan Adzhubey wrote (2017/06/01):
> b) What is the effect of MaximumFileSize option and what would be its optimal 
> value for my IBM LTO-5 SAS drives? I have used 8GB value found in one of the 
> list posts, while the documentation suggests 2GB for LTO-4. But even set at 
> 8GB this would create lots of EOF marks on a 1.5TB tape, do we really need so 
> many?

Hi, I do use 16 GB. Every EOF mark means around 3 seconds delay. So if you have
over 200 files on the tape using 8 GB, it is around 10 minutes extra per tape.

> c) Should I try increasing the tape block size? Set it to fixed? I know this 

I think that there is no reason to use fixed tape block size. On the other side,
increasing tape block size would help, there are many discussions about that.

> topic has been rather controversial, so any recent experience from similar 
> drives/system would be appreciated. I plan to use spooling to a dedicated 
> RAID 
> volume, so hopefully hard drives should not be a bottleneck.

Hard drives are potential bottleneck, they can not handle several write
streams and at the same time one read stream going 150-300 MB/s. Rather
use SSD drives.

Regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula LTO-5

[Cejka Rudolf]

Cejka Rudolf wrote (2017/06/01):
> It is too low for LTO-5. It is at the lower limit before tape
> shoe shining, problably with some tape stops already here.
> The real limit is 47 MB/s for HP and 40 MB/s for IBM, but no
> guarantee it is correct.

Small clarification about this: Speeds 47 MB/s or 40 MB/s are
physical writing speeds after compression, so uncompressed datas
from server to tape drive have to be faster. And because of
variable compression ratio over time, typically between 1:1 and
2:1, then you are safe just with much bigger writing speeds.
>From the other side, if you have writing speed under 100 MB/s,
it is sometimes recommended to switch off compression, so that
even speeds between 50-100 MB/s are safe.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula LTO-5

[Cejka Rudolf]

Richard Fox wrote (2017/06/01):
> Otherwise, this advice is a little contradictory to the documentation which 
> states "On most modern tape drives, you will not need to specify this 
> directive.". 

Given that Linux with LTO-X tape drive is probably a majority system here (not 
counting
configurations without tape drives), the statement is slightly misleading. I'm 
convinced
that it is really not needed because of tape drive, server nor HBA, but it 
seems that it
is really needed because of Linux. However it is not a real problem, because 
Linux allows
to increase the block size "naturally", with the exception that you have 
limiting HBA.

> More importantly however, the documentation (for Bacula 7.2 anyways) says: 
> "The maximum size-in-bytes possible is 2,000,000." which contradicts the 
> assertion that these can be specified as 2MB which is not the same thing. Is 
> the documentation inaccurate on this subject?

I wrote 2 MB as a general recommendation over various manufacturers and software
developers, with non-written suggested value 256 KB or 512 KB as max., so please
take my 2 MB limit just loosely :o)

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula LTO-5

[Cejka Rudolf]

Steven Hammond wrote (2017/05/31):
> 1. Is it safe (Bacula 7.XX) to set the block size to something other 
> than 64K?

You have to try it, but it should be safe in Linux. Write data with
bigger blocks and read them back with source comparison. btape should
do the job too.

> 2. Does increasing the block size increase the throughput?  (2 x 3.6Ghz 
> PROCS - 24 cores, 32GB RAM, 6x600GB 15K RPM SAS, external SAS  / HP LTO-5)

If the disks are not a bottleneck, then in Linux it should increase the
throughput very much. Yes, you have very good disks, but experiences say,
that you really need SSD disks to take full advantage of LTO-5.

> 3. Are there any down sides to increasing the block size?  (e.g., 
> existing tapes created)

Lower portability. If you want to read the tape back, you need a system,
which does support atleast the same block size, which depends on used
operating system, host adapter and its driver. It also depends on tape
drive, but just do not use block bigger than 8 MB and you are safe.
However, it is recommended do not use blocks bigger than 2 MB and please
do not exceed it if you do not know what you are doing.

> Despooling speeds avg 60+ MB/s.  I also have a quad nic bonded 

It is too low for LTO-5. It is at the lower limit before tape
shoe shining, problably with some tape stops already here.
The real limit is 47 MB/s for HP and 40 MB/s for IBM, but no
guarantee it is correct.

> O/S: Ubuntu 17.04 (so we could get a later version of bacula without 
> compiling from source)
> 
> I'm not overly concerned about the speed (everything backs up over night 

You are concerned, because you want to protect your tape drive from
a malfunction.

> fine) but I was hoping to get over 100 MB/s seeing that the LTO-5 
> theoretically can go up to 144 MB/s.

Yes, my typical speeds with LTO-5 are 140-280 MB/s on FreeBSD with
64 KB blocks, but FreeBSD is not so sensitive to block sizes as is Linux.

> Any tips would be greatly appreciated.

Yes, really try to do something with increasing tape write speed and do
not go under 100 MB/s even in low peaks.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] SCSI Errors in LTO External Drive

[Cejka Rudolf]

Francisco Javier Funes Nieto wrote (2016/11/07):
> Bacula Server 7.4.4 in FreeBSD 10.3 with SCSI 29320ALP and LTO 3 External
> SCSI Drive ...
> 
> Doing a btape fill, same thing with btape test, in dmesg:
> 
> (sa0:ahd0:0:3:0): CAM status: SCSI Status Error
> (sa0:ahd0:0:3:0): SCSI status: Check Condition
> (sa0:ahd0:0:3:0): SCSI sense: UNIT ATTENTION asc:29,2 (SCSI bus reset
> occurred)
> (sa0:ahd0:0:3:0): Retrying command (per sense data)
> (sa0:ahd0:0:3:0): tape is now frozen- use an OFFLINE, REWIND or MTEOM
> command to clear this state.
> 
> I don't know what's the error means but I think it's a Cable or Unit error.
> 
> Someone  see this error before ? Any suggestion to debug the error ?
> Discard the cable or unit ?

Hi,
  I have had many problems with SCSI LTO-3. It helped me to connect just
one device per one SCSI bus and to lower speed from 320 MB/s to 160 MB/s.
My position is that External SCSI never worked with 320 MB/s. Maybe you
could have the same problem.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO-7 tape drive settings in bacula 7.4.0

[Cejka Rudolf]

Rao, Uthra R. (GSFC-672.0)[ADNET SYSTEMS INC] wrote (2016/05/25):
> I have a Qualstar Tape library (RLS-87120) with three LTO7 tape drives.
> It is connected to the backup server directly through Fiber Channel.
> I am  running bacula 7.4.  I ran a test backup which completed
> successfully but I found that the LTO-7 Tape was not used up to
> its full capacity of 6TB. It wrote 2.6TB and marked the tape "FULL"
> then moved on to the next tape in the pool. I expected the backup to
> write 6TB to the tape before marking it "FULL"? Here is the information
> from my bacula-sd.conf:

Do you really have LTO-7 tape media in LTO-7 tape drive? It looks like
only LTO-6 tape media in LTO-7 tape drive. Or, you have to have sustained
data rate atleast 100 MB/s, so tape drive can avoid backhitching. In
another case, when you are not unable to feed your tape drive atleast
100 MB/s, then backhitching does occur and full tape capacity could
not be used.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Copy disk to tape is 4x slower than tar

[Cejka Rudolf]

Alan Brown wrote (2016/03/17):
> Caveat: BDXL is up to 120GB per disc (quad layer) and  It _may_ be worth 
> investigating this format for backups, but bacula doesn't play nicely 
> with optical media.
> 
> HVD development (6TB per disc) was abandoned in 2008. Ritek demonstrated 
> 250GB BDXL discs (10 layer) 8 years ago but they've never been marketed. 
> Ditto with Pioneer's 400GB BDXL format and the "1TB Blueray" disk is now 
> 4 years past proposed launch date. What's killed all these "smaller" 
> formats is cheap(ish) HDD/SSDs, cloud storage and the likes of Netflix. 
> That's despite even BDXL 120GB not being large enough capacity to hold a 
> complete 4k video title.

I think that two main killers for all these optical things for data
storage are unbelief in reliability (or simply unreliability) and
slooowness.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Copy disk to tape is 4x slower than tar

[Cejka Rudolf]

Dan Langille wrote (2016/03/13):
> > As well as increasing max file size you need to boost the tape buffer size 
> > from the 64kB default. I use 2MB
> 
> This is a hardware setting?

Hello, it is software settings:

* Increasing max file size is "Maximum File Size". I use 16 GB for
  LTO-5 and for LTO-4 it should be similar. Every file mark means delay
  of several seconds (3-5?), so you can count, how big slow down for
  each file size it is in percents. On the other side, too big file size
  means slower seeks.

* Tape buffer size: It is "Maximum Block Size". It seems that in Linux
  it is a must to increase it. However, in FreeBSD, I have no problems
  with default 65536 bytes, where I have seen speeds over 250 MB/s with
  well compressable datas and without notable system load, so I did not
  increase it yet. For check, how big block you can use, try mt status -v.
  It is smaller of maxio and cpi_maxio. Without kernel patching, you could
  certainly use 131072 bytes. However, maxio could be increased easily
  (MAXPHYS kernel option) and cpi_maxio depends on driver, how well the
  value is obtained. Sometimes it is just a constant, which is known that
  it could be increased. But, I still think that it is not needed so much
  in FreeBSD.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Trouble setting up Bacula on FreeBSD

[Cejka Rudolf]

Simon Templar wrote (2016/03/01):
> Perhaps I should have included this information in the first email,
> but I???ve tried several combinations of parameters in the
> bacula-sd.conf, and it failed each time. Here are the parts I???ve changed:

And perhaps include latest output of btape -c... /dev/nsa0 "test" command.

What is your FreeBSD version?

What is the output of

mt -f /dev/nsa0 status -v

mt -f /dev/nsa0 geteotmodel

mt -f /dev/nsa0 param -l

?

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Trouble setting up Bacula on FreeBSD

[Cejka Rudolf]

Simon Templar wrote (2016/02/29):
> I've set up bacula 7.2 on my FreeBSD server using Postgres as the 
> backend database.

Which FreeBSD version? According to the mt status it seems to be
sufficiently fresh.

> Device {
> ...
>Hardware End of Medium = no   # Noted as FreeBSD specific
>BSF at EOM = no   # Noted as FreeBSD specific
>Backward Space Record = no# Noted as FreeBSD specific
>Fast Forward Space File = yes # Noted as FreeBSD specific
>TWO EOF = no  # Noted as FreeBSD specific

Where did you get this?

Try this:

  BSF at EOM = yes
  TWO EOF = yes
  Hardware End of Medium = no   # maybe not needed now, but there is no
# performance impact, so let's start with it
  Backward Space Record = yes   # not needed, changed to yes is the default
  Fast Forward Space File = yes # not needed, yes is the default

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Windows client 7.0.5 broken?

[Cejka Rudolf]

Kern Sibbald wrote (2016/02/26):
> I have no idea what "developer's gift" is but to the best of my 
> knowledge no Windows 7.0.5 ever existed.

Oops, I'm sorry I was so unclear. I have found the original
status report:

http://blog.bacula.org/bacula-status-report-30-august-2014/

Yes, it was not developers, but "friends" and you decided that it
would be "contributors that are listed in the AUTHORS file".

Please, was it meant as one particular version, which is 7.0.5,
or as access to Bacula Windows binaries with upgrades, possibly
with some delay?

Thank you very much.

> On 02/26/2016 08:49 AM, Cejka Rudolf wrote:
> > Hello,
> >after Windows client upgrade from public 5.2.10 to developers'
> > gift 7.0.5, I have noticed that incremental jobs started to be too
> > big and too slow. Before upgrade with 5.2.10:
> >
> >Elapsed time:   4 hours 28 mins 43 secs
> >FD Files Written:   180,897
> >FD Bytes Written:   18,411,348,578 (18.41 GB)
> >
> > After upgrade with 7.0.5:
> >
> >Elapsed time:   1 day 8 hours 31 mins 19 secs
> >FD Files Written:   10,770,656
> >FD Bytes Written:   24,493,173,950 (24.49 GB)
> >
> > The difference is in ~ 10,000,000 entries.
> >
> > Full job has ~ 33,000,000 entries, where there are ~ 10,000,000
> > directories and ~ 23,000,000 files (strange itself). Job listing
> > confirmed, that 7.0.5 Windows client includes all directory
> > entries on the server in the incremental backup. Please, is there
> > a configuration possibility to work around this problem?
> >
> > I have tried 7.4.0 for personal use also, and it works like 5.2.10,
> > so maybe just 7.0.5 has this bug? Meantmime, I had to return to
> > 5.2.10.
> >
> > Thank you.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Windows client 7.0.5 broken?

[Cejka Rudolf]

Hello,
  after Windows client upgrade from public 5.2.10 to developers'
gift 7.0.5, I have noticed that incremental jobs started to be too
big and too slow. Before upgrade with 5.2.10:

  Elapsed time:   4 hours 28 mins 43 secs
  FD Files Written:   180,897
  FD Bytes Written:   18,411,348,578 (18.41 GB)

After upgrade with 7.0.5:

  Elapsed time:   1 day 8 hours 31 mins 19 secs
  FD Files Written:   10,770,656
  FD Bytes Written:   24,493,173,950 (24.49 GB)

The difference is in ~ 10,000,000 entries.

Full job has ~ 33,000,000 entries, where there are ~ 10,000,000
directories and ~ 23,000,000 files (strange itself). Job listing
confirmed, that 7.0.5 Windows client includes all directory
entries on the server in the incremental backup. Please, is there
a configuration possibility to work around this problem?

I have tried 7.4.0 for personal use also, and it works like 5.2.10,
so maybe just 7.0.5 has this bug? Meantmime, I had to return to
5.2.10.

Thank you.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Faster reading of Volumes from disk

[Cejka Rudolf]

Dan Langille wrote (2016/02/17):
> Writing 334GB to tape takes about 13 minutes.

Hello,
  it means about 334*1024/(13*60) = 438.48 MB/s. Do you really have
so fast tape write?

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] chmod 770 (was Re: Bacula Release 7.4.0)

[Cejka Rudolf]

Dan Langille wrote (2016/02/08):
> Please open a PR (https://bugs.freebsd.org/bugzilla/
> ) and we will work on that.

Yes, but mentioned situation in FreeBSD is non-default and
rather theoretical. Everybody should be satisfied with default
/usr/local/etc/bacula, including me.

Unfortunately this is not the case with direct Bacula installation,
where default sysconfdir is '${prefix}/etc'. So I primarily filled
Bacula bug report (2205).

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] chmod 770 (was Re: Bacula Release 7.4.0)

[Cejka Rudolf]

Dan Langille wrote (2016/02/06):
> > Please, why there is
> > 
> > installdirs:
> >   ...
> >   chmod 770 $(DESTDIR)$(sysconfdir)
> >   ...
> > 
> > in Makefile.in? I had sysconfdir set to /usr/local/etc in FreeBSD
> > and /etc in Linux, so everywhere I have put new bacula,
> > I accidentally cut off access to these system directories to
> > non-root services and programs. (Bug report 2205.)
> 
> Are you using the FreeBSD port?  It defaults to /usr/local/etc/bacula

Hi, no, I have my own cross-platform and cross-administrator installation
script, so I'm independent on server administrators, where bacula-client
has to be installed.

However, as I looked on FreeBSD port, which is heavily patched by the way,
if anybody changes ETCDIR, he could be in trouble too. And there is one
more problem - directory is owned to group bacula:

# cd /usr/ports/sysutils/bacula-client
# make ETCDIR=/usr/local/etc install
...
# ls -ld /usr/local/etc
drwxrwx---  12 root  bacula  1536  8 feb 13:48 /usr/local/etc/

I think that Bacula installation should either use $(sysconfdir)/bacula
instead of just $(sysconfdir), so sysconfdir could be really "sys-conf-dir"
like /etc or /usr/local/etc, and then chmod 770 $(DESTDIR)$(sysconfdir)/bacula
could not change system directory permissions, or chmod should be deleted
(but then I'm not sure about the source of group change - maybe it is the
other thing, which should be solved).

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Release 7.4.0

[Cejka Rudolf]

Kern Sibbald wrote (2016/01/18):
> Yesterday, we released Bacula version 7.4.0.

Hello, thank you very much ;o)

So I have decided that I should finally upgrade my 5.2.12-based backup
site to something newer. So I have upgraded server, then 20+ client
nodes...

And then I have found, that some services started to fail.

And disaster began...

Please, why there is

installdirs:
...
chmod 770 $(DESTDIR)$(sysconfdir)
...

in Makefile.in? I had sysconfdir set to /usr/local/etc in FreeBSD
and /etc in Linux, so everywhere I have put new bacula,
I accidentally cut off access to these system directories to
non-root services and programs. (Bug report 2205.)

(I also filled one minor problem, with id 2204).

Best regards and good night...

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Typical tape write performance

[Cejka Rudolf]

Paul Elliott wrote (2015/12/02):
> I would be interested to hear what block sizes other LTO5/6 users are
> using?

LTO-5/SAS on FreeBSD: I still use Maximum Block Size = 65536, because
333 MB/s transfer rate is sufficient for me. I have tried this small
script

---
#!/bin/sh

B=131072
C=32768

while [ $B -ge 1024 ]; do
  dd if=/dev/zero of=/dev/nsa1 bs=$B count=$C
  B=$(( $B / 2 ))
  C=$(( $C * 2 ))
done
---

with iostat in another terminal window:

   tty sa0   sa1   cpu
 tin  tout  KB/t tps  MB/s   KB/t tps   MB/sus ni sy in id
...
   060  0.00   0  0.00 128.00 3073  384.18   0  0  1  0 99
...
   059  0.00   0  0.00  64.00 5325  332.84   0  0  1  0 99
...
   063  0.00   0  0.00  32.00 7834  244.82   0  0  1  0 98
...
   064  0.00   0  0.00  16.00 11328 177.00   0  0  2  0 98
...
   060  0.00   0  0.00   8.00 14521 113.45   0  0  2  0 98
...
   063  0.00   0  0.00   4.00 16534 64.580  0  1  1 98
...
   063  0.00   0  0.00   4.00 16534 64.580  0  1  1 98
...
   061  0.00   0  0.00   2.00 17723 34.620  0  2  1 98
...
   060  0.00   0  0.00   1.00 18513 18.080  0  2  1 97

> What's the advantage of the maximum job spool size setting? We're not
> using any spooling at the moment so I'm interesting in what parameters
> would be best for our setup. (if any..)

If you have more parallel jobs, full spool space means that all spooling
jobs have to be paused until something is despooled. With Maximum Job
Spool Size set to Maximum Spool Size / number of parallel jobs you can
prevent this situation.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Go from Idea to Many App Stores Faster with Intel(R) XDK
Give your users amazing mobile app experiences with Intel(R) XDK.
Use one codebase in this all-in-one HTML5 development environment.
Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs.
http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Typical tape write performance

[Cejka Rudolf]

Dan Langille wrote (2015/12/02):
> I had not considered that. In my case, I backup to local HDD (ZFS array)
> for long term storage.  Right after those jobs finish, I copy to tape.
> Sounds like I need to implement spooling now. Fortunately, my full
> backups are only about 400GB. I think I can get away with one SSD
> feeding my LTO-4. 

Hello Dan, if you have just one stream at a time, either read stream
or write stream like in this case, things are not so bad with rotational
disks. In this case, you can count almost with read/write speed of the
disk. However do not forget, that nominal speeds are true just at the
beginning of the disk. When datas are read from the end of the disk,
typical speed is about half of the nominal. So if you have disk with
nominal 160 MB/s speed, datas from the end of the disk could be read
typically just at 80 MB/s.

SSDs are needed in case there are more data streams at a time, for examle
if five jobs are spooled (five write streams) and one job is despooled to
the tape (one read stream) at the same time. Then, disks have to reposition
too often and performance drops too much.

iostat and gstat (FreeBSD) are your friends ;o) Very important is %busy
in gstat. If the value is red when writing to the tape, then try SSD
(very simplified).

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Go from Idea to Many App Stores Faster with Intel(R) XDK
Give your users amazing mobile app experiences with Intel(R) XDK.
Use one codebase in this all-in-one HTML5 development environment.
Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs.
http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO Cartridge Memory Specification?

[Cejka Rudolf]

Wolfgang Denk wrote (2015/10/22):
> My LTO library can read and display this data; most of it is
> self-esplaining, except for a data field "TAF" which probably
> contains error or wear information - for most tapes this is empty,
> but for a few (old, long used) ones I see entries like "3, 4, 6" or
> "3, 4, 6, 20".  I would like to understand what these mean?
> 
> Can anybody shed any light on this?  Thanks in advance.

Hello,
  I think that it is a number of Tape Alert Flags raised during
tape lifetime.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Bacula with cryptodev on FreeBSD does not work (#0002141)

[Cejka Rudolf]

Hello, this can hopefully save some time to somebody:

0002141: Bacula with cryptodev on FreeBSD does not work

Description: For all three daemons - stored, dird and filed:
Bacula calls init_crypto() { ... OpenSSL_add_all_algorithms() ...}, which
silently opens file descriptor to /dev/crypto (optional kernel option/module
cryptodev) for communication with kernel crypto engine. Then daemon_start()
is called, which forks itself and closes all open file descriptors (with
exceptions not important here), so it breaks SSL functionality, if there
is used encryption supported by cryptodev engine.

Steps To Reproduce:
- Install FreeBSD >= 10.1-STABLE (>= March 20, 2015 - r280297)
- Add device cryptodev, device crypto and device aesni into your
  configuration or load them as modules
- Try to run backup job with SSL configured between FD and SD
- Job is terminated on the SD side with these errors:
backup-sd: Fatal error: bnet.c:287 TLS Negotiation failed.
backup-sd: Fatal error: TLS negotiation failed with FD at "A.B.C.D:9103"
backup-sd: Fatal error: Incorrect authorization key from File daemon at
  client rejected. Please see http://www.bacula.org/en/rel-manua... for help.
backup-sd: Fatal error: Unable to authenticate File daemon
freebsd-fd: Fatal error: TLS negotiation failed.
freebsd-fd: Fatal error: Failed to authenticate Storage daemon.
backup-dir: Fatal error: Bad response to Storage command: wanted 2000 OK
  storage, got 2902 Bad storage

Additional Information: The problem could be silently ignored in the past,
but since OpenSSL commit
https://git.openssl.org/?p=openssl.git;a=commitdiff;h=323a7e76e61d977ff9f00a8cff396033a6dc37d2;hp=059907771b89549cbd07a81df1a5bdf51e062066
between 1.0.1l and 1.0.1m (I did not check the other branches), there are
added tests of results from EVP_EncryptUpdate() and EVP_EncryptFinal(),
which propagate the error with closed descriptor to cryptodev to the upper
layers.

OpenVPN had exactly the same problem, for further information please see
https://community.openvpn.net/openvpn/ticket/480 .

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

[Cejka Rudolf]

Dimitri Maziuk wrote (2015/05/20):
> Quoting quantum lto-6 sales blurb
> (http://www.quantum.com/products/tapedrives/ltoultrium/lto-6/index.aspx)
> 
> Fast backup and restore performance ?at speeds up to 400MB/s (1.4TB/hr)*
> 
> * Assumes 2.5:1 compression for LTO-6, 2:1 for other LTO
> 
> Assuming no compression = 400/2.5 = 160.
> 
> > Seagate NAS drives have a max throughput ranging from 149MB/sec to 
> > 180MB/sec depending on capacity:
> > 
> > http://www.seagate.com/www-content/product-content/nas-fam/nas-hdd/en-us/docs/nas-hdd-ds1789-3-1409gb.pdf
> 
> 160 < 180.
> 
> you get it now?

It should be "min. 160 < max. 180", which is false.

Tape users usually do not want to switch off hardware tape drive
compression and use slower/faster better/worse software compression
instead of it. Also, max. disk speeds are just at the beginning of
the disk. At the end, it is often just 1/3 to 2/3 of the max. speed.
And when the disk is not new, there could be very annoying speed
instabilities due to internal sector reallocations and some other 
mysteries with no-so-bad/good sectors and heads. I have seen small LBA
interval slow downs even up to just 1 MB/s on one 2 TB drive, and the
drive still reported very good SMART values - no reallocations, no
errors.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

[Cejka Rudolf]

Bryn Hughes wrote (2015/05/19):
> Yes, there is definitely a minimum throughput for LTO drives.  If you 
> aren't writing from local storage then you almost certainly won't be 
> able to meet it for anything at all modern.  LTO-3 was the last 
> generation that could be fed by a gigabit ethernet link (full speed is 
> 80MB/sec), newer technologies have much higher throughput rates.  As 

Even LTO-3 could be a problem to fully (= all the time) feed over gigabit
(using drive compression), but for all existing LTO-[123456] generations,
1 Gb/s is usually sufficient. When there is slower data stream, LTO tape
drive slows down physical tape movement speed. However, there are lower
limits already mentioned (~ 50 MB/s). There is often bigger problem in
many small files in source filesystem and native user/system disk load.

> When using spooling you give Bacula some space, it writes till it hits 
> whatever your max spool size is, then it stops spooling and writes out 
> that data to tape, when done it erases what it had on disk previously 
> and continues spooling data from where it left off. As mentioned above 
> there is an impact on your overall backup times with this as no new data 
> will be flowing in during the periods when Bacula is writing to tape, 
> however it'll keep your tape throughput in the 'happy' zone.

However when spooling job is finished, tape write is started immediately
and other spooling jobs do continue. So the second possibility is that
spooling never stops. In real scenarios, there is something between
these two cases.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

[Cejka Rudolf]

Florian Rist wrote (2015/05/18):
> The question is probably, what happens if a tape change is forgotten. 
> Say I have scheduled backup job for the night and the tape is not change 
> in the evening. Will this mess up everything or will the job 
> automatically be run as soon as the needed tape becomes available?

In theory, it waits until tape becomes available, but in real world,
another events usually happen with result that backup is canceled.

> I seams to be possible to get cheap old used libraries. Is it possible 
> to change the drive?
> I once did this a long time ago with a MOD library system. I was easy to 
> do as the library was controlled via RS232 and did not at all 
> communicate with the drive. Do modern LTO libraries work the same way?

It depends on library. It is even common that you connect just directly
to the drive - and the library is bridged using this drive. However, you
need a library drive and not just standalone internal or external drive.
And of course, expect that library drive is more expensive than standalone
drives. However, it is always better to look at particular library or
autoloader, which possibilities do you have.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

[Cejka Rudolf]

Florian Rist wrote (2015/05/18):
> Thanks for the numbers and pointing me to the spool/unspool problem. 
> This can be fixed by using two hard drives, right?

In RAID-0? Let's hope so :o) However, I would rather use "better" instead
of "fixed". Another solution would be to allow just one parallel job, I'm
not sure about your expectations in this way.

> The whole thing about costs probably seams to be a bit ridiculous, but 
> unfortunately there is niter money nor understanding for the need here 
> right now. So I'm trying to figure out a cheap way to get to a working 
> system that can be handled somehow. And I'm fully aware of the fact that 
> the hardware investment costs will be a minor part in the actual long 
> term overall costs.

The question is: Do you really need a tape solution? Isn't better for you
to start with disk based solution? For example, some HW RAID card with
8 or 16 internal ports (~ 300 or 500 EUR) with some 2-6 TB HDDs? And you
can reuse them after. Tape solutions are very interesting, but they usually
count with great number of tape media or with archival purposes or data
exchange.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

[Cejka Rudolf]

Florian Rist wrote (2015/05/17):
> I don't have the hardware yet. My plan is to get a small cheap server, 
> some low cost CPU (some i3 Core), 6 TB HD (for spooling), a SAS 
> controller and a LTO 5 or 6 drive (probably from Tandberg). I'll need to 
> change tapes manually, any library solution seams to be to expensive. 

Just try to ask several vendors and look at
http://www.oracle.com/us/products/servers-storage/storage/tape-storage/sl150-modular-tape-library/quantum-comp-i40-i500-wp-1704937.pdf
It is rather old (2012), but it shows that price differences may be
dramatical. For you, table 1 and line 1: Cheaply expandable 30-slot and one
LTO-5 drive SL150 cumulative cost 6000 USD against non-expandable 40-slot and
one LTO-5 drive i40 cumulative cost 13000 USD - but once again, it was 2012.

If you want to avoid tape back hitching, you have to have atleast:
HP LTO-5: 47 MB/sIBM LTO-5: 40 MB/s
HP LTO-6: 54 MB/sIBM LTO-6: 40 MB/s
(I'm sorry, I do not know anything about Tandberg LTOs.)
However, these speeds are just native speeds without compression,
so you have to be unpredictably faster in reality. Also, various
models can be somewhat different and there can be some prediction
logic, which can effectively change these requirements too. So the
best thing is to be as fast as possible :o) And I am afraid, that single
6 TB drive is not a good solution. It can be maybe sufficient if you first
spool the data, and then just unspool them to the tape, or using D2D2T,
but if you have spooling and unspooling at the same time, I do not think
that it will be considerate to your tape drive.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] STK SL-3000 experiences

[Cejka Rudolf]

tonyalbers wrote (2015/03/05):
> Thanks Davide, that's excellent news.
> 
> If anyone else has got anything to add, please do.

Hello, almost nothing. I have five-modules SL-150 with one LTO-5
and one LTO-6 over SAS and under FreeBSD and without any problem.

Just I have bought it when it was very new (mid 2012) and first library
and tape firmware(s) had problem with SAS connection stability, random
disconnections once per month or so. But it was solved in 2013 +/- 1 :o)

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula 7 on FreeBSD, suddenly can't do anything

[Cejka Rudolf]

dweimer wrote (2015/01/27):
> ...

Interesting. What did you do right before "suddenly" actually?
Nothing, or something and somewhere? :o)

> > It had me scratching my head for a while, because it was still half
> > working. My nagios and bacula daemons couldn't find any *.local names
> > anymore, but I could still ping the same machines by name at the
> > command prompt.

It seems that you have some nontrivial DNS configuration? Some local
daemon like named or unbound? Could you show us some example?

ping some.server # what ip is used here? what says tcpdump port domain?
grep some.server /etc/hosts # or is the translation from hosts?
grep hosts /etc/nsswitch.conf # do you have "hosts: files dns" here?
cat /etc/resolv.conf # something suspicious here?
ldd -a /usr/local/bin/bacula-dir | grep libc.so # just /lib/libc.so.7 here?
file /usr/local/bin/bacula-dir # for which FreeBSD is it compiled?

Ping uses gethostbyname2(). Bacula uses getaddrinfo(), if there is
#define HAVE_GETADDRINFO 1 in src/config.h. So ping and bacula could
behave differently, but it is weird. Unfortunately, nslookup and host
use different mechanisms. They skip /etc/hosts for example. So they
can check remote DNS server, but not problems with local translation,
which seems to be your case.

> Yes to .local DNS suffix, ...

And please how?

> morning ran without a hitch. Then Monday night when I went to login 
> after switching my external hard drive for the new week that's when the  
> problems were noticeable. At first it connected, just horribly slow. 
> After a restart of all the Bacula services, I could no longer connect 
> with bconsole on the local machine.

Could you do gdb bconsole and check, that timeout is with name resolution?

And looking into bacula-7.0.5 sources, I think that debug level 100
and tracing on should be sufficient for tracing name resolution
(src/lib/bsock.c and BSOCK::open, then bnet_host2ipaddrs and resolv_host
in bnet.c).

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula 7 on FreeBSD, suddenly can't do anything

[Cejka Rudolf]

dweimer wrote (2015/01/26):
> I did find a work around, but not sure where the problem is for sure, 
> going to post on FreeBSD list as well. the fix, convert all my hostnames 
> in configuration files from fully qualified DNS to IP addresses. I just 
> did this on a hunch, but apparently something somehow has broken in the 
> Bacula name resolution on FreeBSD or at least suddenly on my system.

Hello, maybe DNS server or DNS service is down? Please check
/etc/resolv.conf and availability of DNS service. There is also
one important thing when upgrading from FreeBSD 9.x to 10.x, where
bind (named) is removed from the base system. Since 10.x, there is
just caching DNS resolver (unbound) and if you had local DNS server,
after upgrade to 10.x you have to install bind from ports.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] spool disk filesystem, checksums

[Cejka Rudolf]

Daniel Pocock wrote (2014/12/05):
> Now it is in the bug tracker:
> http://bugs.bacula.org/view.php?id=2111

"... Data corruption occurs on disks sometimes, e.g. the write to disk
didn't record data correctly on the platter and any attempt to read
it returns the corrupted data. This may just be 1 bit and it may only
occur once in millions of writes but it is known to happen. ..."

You wrote just about disks first, so I responded just about disks too.
After this, you responded with "disks and IO subsystems". And now in bug
tracker, you again write just about disks.

Please:
1) Read atleast Error rates and handling in
   http://en.wikipedia.org/wiki/Hard_disk_drive
2) Be clear on what problem do you have
3) Think about RAM at first - in PCs without ECC, in servers with just
   very simple ECC, much weaker than used in disks and on the tapes
4) ... and what about the caches, all storages and all communication
   paths along the path? What all we know about them? Almost nothing?
   Checking just the spool disks (with all related subsystems needed to
   check these) is very narrow view of such a complex thing

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] spool disk filesystem, checksums

[Cejka Rudolf]

Daniel Pocock wrote (2014/12/04):
> On 04/12/14 18:35, Kern Sibbald wrote:
> > On 12/03/2014 08:49 PM, Daniel Pocock wrote:
> >>
> >> Does Bacula checksum content on the spool disk before sending it to tape?
> >>
> >> To be more explicit, if there is a single bit error on the spool disk,
> >> will it be noticed before going onto tape or would it only be noticed in
> >> future when a file is taken off the tape?
> > 
> > Unless you are running ZFS for the spool disk, the error will only be
> > noticed when the data is read from the tape.
> > 
> 
> In that case, it sounds like a good idea to use ZFS or Btrfs with
> checksums enabled

Hard drives use error correction/detection codes, so single bit error
without any error indication is unlikely. Especially in case of spool
disks, where datas are read shortly after write.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Failed on backup volume 2 suddenly

[Cejka Rudolf]

Keith T wrote (2014/11/21):
> BTW, any advise/references of tuning "DB dir inserting" faster? Thanks
> in advance!

MySQL: MyISAM & delay_key_write = ALL

and either do not do unexpected power offs (not so hard today), or when
the disaster happens, run "repair table ..." for all tables in bacula
dabatase (easy thing).

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO-6 - tape cleaning frequency

[Cejka Rudolf]

>On Nov 10, 2014, at 10:15 AM, Dan Langille  wrote:
>> How often does your LTO-6 tape drive need to be cleaned?

Hi, I bought five cleaning tapes ten years ago. Four are on sale :o)

HP LTO-3: Almost never, 7 years of operation

HP LTO-5: Almost never, 2 years of operation
  650 mounts, roughly 1 petabyte written
  2 uses of cleaning tape, however I would bet that both were
  just my experiments with reading of cartridge memory

Simply buy library, reserve one slot for cleaning tape, insert cleaning
tape into and sing "I don't care! I love it! I don't care!" :o)

You can look into /usr/src/sys/cam/scsi/scsi_sa.c. The timeouts are
in minutes. Even IO has default timeout 4 minutes. Maybe the reason
is possible cleaning?

If you are really interested, please read

http://www.tarconis.com/documentos/LTO_Cleaning_wp.pdf

It is rather old, but many things still have to be true.
It is from HP and it is possible that IBM is different.
But I did not find similar document from IBM.

The most important facts from the document above are:
1) The tape is flying over the head during the operation and the only
   contact is just when tape starts or stops. So yes, correct streaming
   is very important. Tape drive has to write data without repositions.
2) There is internal non-abrasive brush, which is usually sufficient.
   (From 150 km and LTO-1, I think that it may be preventively used
   after roughly 5 fully written tapes, but it is just my unconfirmed
   estimate.)
3) The cleaning tape is used practically only as a last resort and could
   not replace the brush - it is for another type of dirt. And it is
   abrasive, so if there is no reason to use cleaning tape, do not do it.
   Your head will be happier.

Regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO speed optimisation

[Cejka Rudolf]

Bryn Hughes wrote (2014/11/05):
> Finally there are some OS-level settings for the 'st' driver (I'm 
> assuming you are on Linux).  With my LTO3 drives I need to add this to 
> the kernel command line:

Hi, Solaris 11.

> On 2014-11-05 03:48 AM, Roberts, Ben wrote:
> > Hi all,
> > I'd like to try and make some speed improvements to my Bacula setup 
> > (5.2.13, Solaris11). I have data (and attribute) spooling enabled 
> > using a pool of 46x 1TB directly-attached SAS disks dedicated to this 
> > purpose. Data is being despooled to 2x directly-attached SAS LTO6 

Hmm, I do not believe in this. I think that this is too expensive and
too inefective solution. Too many mechanical disks. Which volume
organization do you use? ZFS: zpool list -v?

46 mechanical disks can provide really big aggregate speed for many
parallel data streams, but if you have just one or two 200 MB/s read
streams and several slow write streams, it is very hard work for
mechanical disks.

As wrote Bryn, it is best to obserse (peak) drives/RAID utilisation,
which gives you best valuation, if current spooling storage is
sufficient or not.

For LTO-6 tape, there is digital speed matching / data rate matching,
which can slow down physical write speed from 160 MB/s up to
40 MB/s / 54 MB/s. The tape drive tries to learn the best speed from
data stream continuity, so even very small almost unobservable delays
can result in slow tape writes. Even if you see good average
throughput per second.

> > drives at around 100mB/sec each. I think I should be able to get 
> > closer to the ~160mB/s maximum uncompressed thoughput the drives and 
> > tape media support (ref: 
> > http://docs.oracle.com/cd/E38452_01/en/LTO6_Vol4_E1/LTO6_Vol4_E1.pdf) 

You can try to switch off tape drive compression. There is some possibility,
that write speed could increase. With tape drive compression, you have to
have bigger throughput, atleast 200 MB/s.

> > I've just done a speed test and can read from the spool array at a 
> > sustained 300mB/sec even while other jobs are running, so I'm sure 
> > there's no bottleneck at the disk layer. My suspicion is that the 
> > bottleneck is at the application layer, probably due to the way I have 
> > Bacula configured.

Don't be sure. I had similar speeds and it never worked for me. Yes,
there was much lower number of mechanical drives, but 46 have not
to be better that 4 in this case.

However in any case, observe CPU utilization too. And do not forget
to compile Bacula with compiler optimizations, e. g. gcc -O3.
And I expect that you have Spool Data = yes and you do not have
Spool Attributes = no.

> > Having read through Bareos' tuning paper 
> > (http://www.bareos.org/en/Whitepapers/articles/Speed_Tuning_of_Tape_Drives.html),
> >  
> > I've updated the max file size from 1->50GB which increased the 
> > throughput from ~75 to ~100mB/sec. I believe I need to look at tuning 
> > the block size to gain the last bit of improvement.

Count 3-5 seconds per filemark. So yes, after change from 1 GB to 50 GB
file size you could see such a big improvement, but potential of change
of this type is exhausted now.

Bacula 5.2.12, FreeBSD 9.3, LSI MegaRAID SAS 9265-8i with
4 KINGSTON SKC100 480GB SSDs in RAID0, so read/write performance
is between 1-2 GB/s, LTO-5 tape drive with 140 MB/s native
speed. As you can see, there is no problem to run over
180 MB/s even on just LTO-5 an I saw up to 250 MB/s with
highly compressable data (even with just 64 KB per transfer
on tape, which is often discussed here too, that 64 KB/t is
too low). Temporal drops to 30 MB/s are due to filemarks,
file size is 16 GB.

mfid0   sa0
   time   KB/t  tps  MB/s   KB/t  tps  MB/s
...
2014/11/06 01:49:40  119.5 2685 313.3  64.00 2657 166.1
2014/11/06 01:49:45  120.6 2345 276.3  64.00 2853 178.3
2014/11/06 01:49:50  120.3 2453 288.2  64.00 2778 173.6
2014/11/06 01:49:55  115.5 1478 166.6  64.00  602 37.62
2014/11/06 01:50:00  124.8 1419 172.8  64.00 2292 143.3
2014/11/06 01:50:05  123.5 1765 212.9  64.00 2868 179.2
2014/11/06 01:50:10  118.3 2884 333.2  64.00 2626 164.1
2014/11/06 01:50:15  121.8 1993 237.0  64.00 2331 145.7
2014/11/06 01:50:20  122.7 1979 237.1  64.00 2603 162.7
2014/11/06 01:50:25  122.6 2140 256.3  64.00 2844 177.7
2014/11/06 01:50:30  121.2 2237 264.8  64.00 2807 175.4
2014/11/06 01:50:35  122.2 2306 275.2  64.00 2973 185.8
2014/11/06 01:50:40  120.5 2711 319.0  64.00 2999 187.4
2014/11/06 01:50:45  79.67 1397 108.7  63.99  581 36.31
...

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] DesSpooling Attr

[Cejka Rudolf]

Iban Cabrillo wrote (2013/11/27):
>  Data spooling: 0 active jobs, 0 bytes; 1 total jobs, 1,845,487,814 max
> bytes/job.
>  Attr spooling: 1 active jobs, 18,289,433 bytes; 0 total jobs, 18,289,433
> max bytes.
> 
> Data spooling about 10 minutes, Attr Spooling more than 30 minutes and
> growing
> This happen for LT03 and LT=5 tapes

Could you look at mysql server please? Its CPU usage, disk usage,
swapping, memory usage...

Here are examples from my backup server with mysql:

1) 18,688,335 files in 37 minutes, which is ~ 500,000 files per minute:

   03-Nov 08:37 backup-sd: Sending spooled attrs to the Director.
   +Despooling 6,159,834,448 bytes ...
   03-Nov 09:14 backup-dir: Bacula backup-dir 5.2.12 (12Sep12):

2) 309,936 files in time is under 1 minute:

   02-Nov 14:00 backup-sd: Sending spooled attrs to the Director.
   +Despooling 86,532,417 bytes ...
   02-Nov 14:00 backup-dir: Bacula backup-dir 5.2.12 (12Sep12):

I think that there are three key settings on my server:

1) MyISAM tables
2) delay_key_write = ALL in my.cnf
3) Database on SSD

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Rapidly troubleshoot problems before they affect your business. Most IT 
organizations don't have a clear picture of how application performance 
affects their revenue. With AppDynamics, you get 100% visibility into your 
Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Multiple Catalog Configuration

[Cejka Rudolf]

Mingus Dew wrote (2013/06/03):
> Does anyone have an example they can share of using mutliple catalogs? I
> have a particular set of NAS backups that has over 500,000,000 file records
> and pretty much need to use a separate catalog to hopefully overcome 3 days
> of table locking while inserting preventing other jobs from completing.

Hello, I just started with more catalogs too ;o) There is bacula-dir.conf
example below. Remember that client1 and client2 can be actually one real
host, not just two different. You can switch between catalogs using use
in bconsole. You just need to know, that these are two completely different
databases and they are both filled with your configuration bacula-dir.conf,
but some things like tapes are not usually directly in configuration, so you
have to distribute or copy tape records between catalogs by hand. At least
I think - I did not find anything better, than to use many delete commands
in one catalog and many add commands in the second catalog (I have moved
some tapes from one catalog to the second.).

Client {
  Name = client1
  ...
  Catalog = cat1
  ...
}

Client {
  Name = client2
  ...
  Catalog = cat2
  ...
}

Catalog {
  Name = cat1
  dbname = db1; user = user; password = pass
}

Catalog {
  Name = cat2
  dbname = db2; user = user; password = pass
}

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
How ServiceNow helps IT people transform IT departments:
1. A cloud service to automate IT design, transition and operations
2. Dashboards that offer high-level views of enterprise services
3. A single system of record for all IT processes
http://p.sf.net/sfu/servicenow-d2d-j
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bacula seems not to get mtx-changer output

[Cejka Rudolf]

Marco Carcano- Statusnobilis SA wrote (2013/04/16):
> /usr/libexec/bacula/mtx-changer /dev/sg3 listall 0 /dev/nst0

Do you have write permissions on /dev/sg3?

> 3991 Bad autochanger command: /usr/libexec/bacula/mtx-changer /dev/sg3 
> loaded 1 /dev/nst0 0
> 3991 result="": ERR=Child died from signal 9: Kill, unblockable

Try this command from command line, still the same result?
Then you can try strace -f ... 2>strace.out a see, where it dies.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
http://www2.precog.com/precogplatform/slashdotnewsletter
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Time for "Accurate Backup"

[Cejka Rudolf]

Masopust, Christian wrote (2013/02/12):
> As there are lots of files within the filesets to be backuped
> (appr. 600) I would kindly ask you about your experiences
> with accurate backups.

Hello, you can try it yourself - configure it and do "estimate".
I tried it with client with 15 000 000 files in a backup and I changed
my questions from "time" theme (because everything faster than full
backup is good for me and I expect, that it has to be faster)
to "client cpu and memory used" theme. And I'm not happy from backup
client using > 2 or 4 GB, so I stopped my experiments with accurate
backups for now.

However, I would be very very happy, if I did some mistake and if memory
usage is not so bad as I have seen (something like streaming - get new
attributes all the time, remember just currently needed information and
delete already used data - to have just small information window on
client side).

> How long will it last to send the accurate information
> of appr. 600 files and directories to the file daemon?

I closed my experiments with that it would be somewhat similar to waiting
on interactive restore, which is about 12 minutes for 15 000 000 files
in a backup in my case. (Is there a chance to perform better with
Postgres instead of MySQL? I saw many suggestiong about Postgres instead
of MySQL because of writing, but what about reading?)

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013 
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO-5 Tapes full after between 30-130gb Dell PV-124T

[Cejka Rudolf]

bdoran wrote (2013/01/15):
> I'm using a Bacula setup in conjunction with a Dell PV-124T which has 16 
> LTO-5 tapes. The system works correctly except that the tapes become 
> extremely quickly. There is no consistency between the amount written to each 
> tape though it is generally in the range of 30-130gb - far less then the 
> expected volume.

Why do you have commented out Maximum Spool Size? And I could not find
any Spool Data directive.

In case of LTO-5, I'm not currently sure about the needed minimal tape
write speed, but you should expect it somewhere among 40-70 MB/s, and
it is almost impossible to do it without data spooling.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Master Java SE, Java EE, Eclipse, Spring, Hibernate, JavaScript, jQuery
and much more. Keep your Java skills current with LearnJavaNow -
200+ hours of step-by-step video tutorials by Java experts.
SALE $49.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122612 
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO5 speed

[Cejka Rudolf]

f.staed...@dafuer.de wrote (2013/01/02):
> thanks for your quick answer. So the sound of the drive like starting 
> and stopping in about 10 second intervals are normal?

Hi, no, it is not normal. Except that you have too small Maximum File Size
setting. I have 8 GB, which means atleast one file per one minute, and it
can sustain ~ 140 MB/s over the whole file. Furthermore, I have increased
Maximum File Size to 16 GB right now and I'm thinking even about 32 GB.

> While writing this - is maybe Maximum File Size of SD the reason? Since 
> transferred data is about 1GB and it is writing an EOF mark to tape? If 

Almost certainly yes. I have ~ 4-8 seconds pauses for each EOF mark like
you, so it means, that 1 GB is too small value, because it is written
in ~ 7 seconds and you have almost 50 % overhead.

> so, a full tape will have about 1500 EOF marks - can I safely put it to 
> a higher value? As far as I understand the only drawback are maybe 
> slower restores of files. But I don't think that would be a problem, 
> since new data arrive in 5-10GB chunks.

If you have disk spooling and tape file seeks (but I think that bacula can
do even record seeks), then average delay for two minute files (Maximum
File Size = 16gb) is just about one minute per random seek, which I have
usually just one per one restore. And if you count tape loads and unloads
plus robotics and regular tape seeks (again typical times around one
minute each), then impact of Maximum File Size = 16gb (or even more) for
restores is not so significant.

Furthermore, I had problems with read performance atleast 80 MB/s for
LTO-3 and now even atleast 140 MB/s for LTO-5 from conventional HW RAID0
with mechanical disks, when there were another faster spooling and
writing jobs, so I decided to switch to HW RAID0 with SSDs.

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122712
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] How to specify drive of Autochanger?

[Cejka Rudolf]

Wolfgang Denk wrote (2012/11/16):
> > > Autochanger {
> > >Name = Autochanger
> > >Device = LTO3-0
> > >Device = LTO3-1
> > >Changer Command = "/usr/libexec/bacula/mtx-changer %c %o %S %a %d"
> > >Changer Device = /dev/tape/by-id/scsi-1BDT_FlexStor_II_00DE64100465_LL0
> > > }
> > 
> > Try replacing the two Device= lines with one line:
> >  Device = LT03-0,LT02-1
 ^^
 LTO3-1

> Thanks for the idea - done, but unfortunately this does not change the
> behaviour in any way.

And this is good, because Bacula manual explicitly says, that "You may
specify multiple device names on a single line separated by commas,
and/or you may specify multiple Device directives. This directive is
required." for Autochanger resource.

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] How to specify drive of Autochanger?

[Cejka Rudolf]

John Drescher wrote (2012/11/15):
> > However, no matter what I enter at the prompt "Enter autochanger
> > drive[0]:", the director will always send a command to access drive
> > index 0.
> 
> Where do you get this prompt? I have never ever been asked this and I
> have had an autochanger at work since 2006 and I used it with bacula
> from the start since I was already using bacula for years before I got
> the changer..

And do you have two or more drives?

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] How to specify drive of Autochanger?

[Cejka Rudolf]

Wolfgang Denk wrote (2012/11/16):
> I did the same (to test the drives individualy, to make sure these are
> actually available and working); see the full configuration I posted
> before here:
> http://article.gmane.org/gmane.comp.sysutils.backup.bacula.general/72107

Hi, change
Device = LTO3-0, LTO3-1
to
Device = Autochanger
What does say status storage=LTOLIB then?

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] rc-chio-changer for FreeBSD

[Cejka Rudolf]

Hello,
  please, could anybody test newer version of rc-chio-changer for FreeBSD
(and possibly for OpenBSD/NetBSD) in bacula and send me, with which
configuration? It has added listall and transfer commands plus some other
smaller modifications. Personally, I do use 120 slot library with
2 tape drives.

Thank you.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic
#!/bin/sh
#
# Bacula interface to tape libraries and autoloaders for FreeBSD
# (by Rudolf Cejka , v1.2, 2012/11/14)
#
# $Id$
#
# If you set in your Device resource
#   Changer Command = "path-to-this-script/chio-changer %c %o %S %a %d"
# you will have the following input to this script:
#   chio-changer "changer-device" "command" "slot" "tape-device" "drive-index"
#   $1   $2   $3 $4$5
# for example (on a FreeBSD system):
#   chio-changer /dev/ch0 load 1 /dev/nsa0 0
#
# If you change the script, take care to return either the chio exit
# code or a 0. If the script exits with a non-zero exit code, Bacula
# will assume the request failed.
#

PROGNAME=`basename $0`

# Uncomment the following line, if you want to log debug output.
#DEBUG=/var/run/bacula/${PROGNAME}.log

# Uncomment the following line, if you need to eject a tape before moving
# it from the drive.
#OFFLINE=yes

# Uncomment one or more of the following lines, if you need to wait for
# some time (in seconds) after unloading, loading or transferring a tape.
#OFFLINE_SLEEP=10
#LOAD_SLEEP=10
#MOVE_SLEEP=10

# Uncomment the following line, if you do not have a changer with volume
# reader.
#FAKE_BARCODES=/usr/local/etc/bacula-barcodes

usage()
{
  cat <  [slot] [tape-device] [drive-index]

Commands ():
  unload  Unload a tape into the slot from where it was loaded
  load  Load a tape from the slot  (1-based)
  transfer   Transfer a tape from the slot  to
  the slot  (1-based)
  listList full storage slots
  listall List all storage slots and drives with source information
  loaded  Give slot from where the tape was loaded (0 = empty drive)
  slots   Give number of available slots

Example:
  ${PROGNAME} /dev/ch0 load 1Load a tape from the slot 1

EOF
  exit 1
}

# Default settings
CHANGER=/dev/ch0
TAPE=/dev/nsa0
DRIVE=0

CHIO=/bin/chio
MT=/usr/bin/mt

if [ -n "${DEBUG}" ]; then
  MSG=$0
  for PAR; do MSG="${MSG} \"${PAR}\""; done
  echo `date +"%Y/%m/%d %H:%M:%S"` ${MSG} >> ${DEBUG}
fi

if [ -n "$1" ]; then
  CHANGER=$1;
fi
COMMAND=$2
SLOT=$3
SLOTDST=$4
if [ -n "$4" ]; then
  TAPE=$4
fi
if [ -n "$5" ]; then
  DRIVE=$5
fi

case ${COMMAND} in
unload)
  if [ "${OFFLINE}" = yes ]; then
${MT} -f ${TAPE} offline
if [ $? = 0 -a -n "${OFFLINE_SLEEP}" ]; then
  sleep ${OFFLINE_SLEEP}
fi
  fi
  if [ -z "${SLOT}" ]; then
${CHIO} -f ${CHANGER} return drive ${DRIVE}
  else
${CHIO} -f ${CHANGER} move drive ${DRIVE} slot $((${SLOT} - 1))
  fi
  if [ $? -ne 0 ]; then
# In case of an error, try to unload the cartridge to the first free slot
FREE=`${CHIO} -f ${CHANGER} status slot | \
  sed -ne '/FULL/d;s/^slot *\([0-9]*\):.*/\1/p' | \
  awk 'BEGIN { n = 0 } { n = $1 + 1; exit } END { print n }'`
if [ ${FREE} -gt 0 ]; then
  ${CHIO} -f ${CHANGER} move drive ${DRIVE} slot $((${FREE} - 1))
else
  exit 1
fi
  fi
  ;;
load)
  if [ -z "${SLOT}" ]; then
usage
  fi
  ${CHIO} -f ${CHANGER} move slot $((${SLOT} - 1)) drive ${DRIVE}
  if [ $? -ne 0 ]; then
exit 1
  fi
  if [ -n "${LOAD_SLEEP}" ]; then
sleep ${LOAD_SLEEP}
  fi
  ;;
transfer)
  if [ -z "${SLOT}" -o -z "${SLOTDST}" ]; then
usage
  fi
  ${CHIO} -f ${CHANGER} move slot $((${SLOT} - 1)) slot $((${SLOTDST} - 1))
  if [ $? -ne 0 ]; then
exit 1
  fi
  if [ -n "${MOVE_SLEEP}" ]; then
sleep ${MOVE_SLEEP}
  fi
  ;;
list)
  if [ -z "${FAKE_BARCODES}" ]; then
${CHIO} -f ${CHANGER} status -v slot | \
  sed -ne 's/^slot *\([0-9]*\):.*FULL.*voltag.*<\([^:]*\):.*/\1:\2/p' | \
  awk -F: '{ print $1 + 1 ":" $2 }'
  else
if [ -f "${FAKE_BARCODES}" ]; then
  grep -v -e "^#" -e "^$" < ${FAKE_BARCODES}
else
  echo "${PROGNAME}: Barcode file ${FAKE_BARCODES} is missing"
  exit 1
fi
  fi
  ;;
listall)
  if [ -z "${FAKE_BARCODES}" ]; then
${CHIO} -f ${CHANGER} status -vS | \
  sed -ne '
  s/^slot *\([0-9]*\):.*ENAB.*FULL.*voltag.*<\([^:]*\):.*/I:\1:F:\2/p;t
  s/^slot *\([0-9]*\):.*FULL.*voltag.*<\([^:]*\):.*/S:\1:F:\2/p;t
  s/^drive 
*\([0-9]*\):.*FULL.*voltag.*<\([^:]*\):.*source.*<[^0-9]*\([0-9]*\)>.*/D:\1:F:\3:\2/p;t
  s/^slot *\([0-9]*\):.*ENAB.*voltag.*<\([^:]*\):.*/I:\1:E/p;t
  s/^slot *\([0-9]*\):.*voltag.*<\([^:]*\):.*/S:\1:E/p;t
  s/^drive *\([0-9]*\):.*voltag.*<\([^:]*\):.*/D:\1:E/p' | \
  awk -F: '{ for (n = 1; n <= NF; n++) printf "%s%s",
  (n == ($1 == "D" ? 4 : 2)) ? ($n == "" ?

Re: [Bacula-users] LTO3 tape capacity (variable?)

[Cejka Rudolf]

Alan Brown wrote (2012/10/01):
> 2Mb is the maximum block size I've found to work on LTO* - it's a bacula 
> limit, not an LTO one.

Or it could be a limit of an operating system or of used hw interface.

> Increasing from 65535 will improve throughput significantly.
> There have been discussions about this in the last few years.

Sometimes yes, sometimes no. It depends on many things, like hardware
of your server, used interface type and speed and used operating system.
I'm not sure just about one thing right now - on which level rewrites
are done, either on a block level, or on something smaller.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Got visibility?
Most devs has no idea what their production app looks like.
Find out how fast your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219671;13503038;y?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO3 tape capacity (variable?)

[Cejka Rudolf]

Alan Brown wrote (2012/09/28):
> > Aren't these considered reasonable settings for LTO3?
> >
> > Maximum block size = 262144   # 256kb
> > Maximum File Size = 2gb
> 
> Not really.

Hi, I have

  Maximum Block Size = 65536
  Maximum File Size = 4gb

and without any problem. All tapes have over 400,000,000,000 bytes,
except 2 of 116, where I think that one was forced to change, so just
1 of 116 has under 400 GB (around 380 GB). Almost all tapes are
7 years old now. My typical end of tape mesage is like

End of medium on Volume "" Bytes=618,539,200,512 Blocks=9,438,187

> Change maximum file size to 10Gb and maximum block size to 2M

However, then please test readibility of these data on all systems
you could possibly use. Increasing blocks size is potentially dangerous,
so before increasing it, be sure, that you can read all these data back.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://ad.doubleclick.net/clk;258768047;13503038;j?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO3 tape capacity (variable?)

[Cejka Rudolf]

Stephen Thompson wrote (2012/09/25):
> The tape in question have only been used once or twice.

Do you mean just one or two drive loads and unloads?

> The library is a StorageTek whose SLConsole reports no media (or drive) 
> errors, though I will look into those linux-based tools.

There are several types of errors, recoverable and non-recoverable, and
I'm afraid that you see just non-recoverable, but it is too late to see
them.

> Our Sun/Oracle service engineer claims that our drives do not require 
> cleaning tapes.  Does that sound legit?

If you are interested, you can study
http://www.tarconis.com/documentos/LTO_Cleaning_wp.pdf ;o)
So in HP case, it is possible to agree. However, you still
have to have atleast one cleaning cartridge prepared ;o)

> Our throughput is pretty reasonable for our hardware -- we do use disk 
> staging and get something like 60Mb/s to tape.

HP LTO-3 drive can slow down physical speed to 27 MB/s, IBM LTO-3
to 40 MB/s. Native speed is 80 MB/s, bot all these speeds are after
compression. If you have 60 MB/s before compression and there are
some places with somewhat better compression than 2:1, then you are not
able to feed HP LTO-3. For IBM drive, it is suffucient to have places
with just 2:1 to need repositions.

> Lastly, the tapes that get 200 vs 800 are from the same batch of tapes, 
> same number of uses, and used by the same pair of SL500 drives.  That's 
> primarily why I wondered if it could be data dependent (or a bacula bug).

And what about the reason to switch to the next tape? Do you have something
like this in your reports?

22-Sep 02:22 backup-sd JobId 74990: End of Volume "1" at 95:46412 on device 
"drive0" (/dev/nsa0). Write of 65536 bytes got 0.
22-Sep 02:22 backup-sd JobId 74990: Re-read of last block succeeded.
22-Sep 02:22 backup-sd JobId 74990: End of medium on Volume "1" 
Bytes=381,238,317,056 Blocks=5,817,238 at 22-Sep-2012 02:22.

Do not you use something from the following things in bacula configuration?
UseVolumeOnce
Maximum Volume Jobs
Maximum Volume Bytes
Volume Use Duration
?

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] LTO3 tape capacity (variable?)

[Cejka Rudolf]

> > We've been using LTO3 tapes with bacula for a few years now.  Recently I've
> > noticed how variable our tape capacity it, ranging from 200-800 Gb.
> >   Is that strictly governed by the compressibility of the actual data being
> > backed up?

Hello,
  the lower bound 200 GB on 400 GB LTO-3 tapes is not possible due
to the drive compression, because it always compares, if compressed
data are shorter that original. In other case, it writes data uncompressed.
So, in all cases, you should see atleast 400 000 000 000 bytes written
on all tapes.

> > Or is there some chance that bacula isn't squeezing as much
> > onto my tapes as I would expect? 200Gb is not very much!

In bacula, look mainly for the reasons, why there is just 200 GB written.
If the tape is full, think about these:

- Weared tapes. Typical tape service life is written as 200 full cycles.
  However, read
  http://www.xma4govt.co.uk/Libraries/Manufacturer/ultriumwhitepaper_EEE.sflb
  where they experienced problems with some tapes just only after
  30 cycles! How many cycles could you have with your tapes?

- Do you use disk staging, so that tape writes are done at full speed?
  Do you have a good disk staging? Considering using SSDs for staging
  is very wise. If data rate is lower that 1/3 to 1/2 of native tape
  speed (based on drive vendor, HP or IBM), then drive has to perform
  tape repositions, which means another important excessive drive and
  tape wearing.

  My experiences are, that even HW RAID-0 with four 10k disks could not
  be sufficient and when there are data writes and reads in parallel,
  it could not put 80 MB/s to the drive, typically just 50-70 MB/s,
  which is still acceptable for LTO-3, but not good.

  Currently, I have 4 x 450 GB SSDs HW RAID-0 with over 1500 GB/s without
  problem running writes and reads in parallel and just after that I hope
  that it is really sufficient for >= LTO-3 staging and putting drives and
  tapes wearing to minimum.

- Dirty heads. You can enforce cleaning cycle, but then return to the
  two points above and other suggestiong, like using some monitoring
  like ltt on Linux (or I have some home made reporting tool using
  camcontrol on FreeBSD), where it would be possible to ensure, that
  your problem are weared tapes, or something else.

Best regards.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Slow tape write, no more than 60-70 MB/s in 5.0.3

[Cejka Rudolf]

Brian Debelius wrote (2011/11/09):
> Well compiling with -O3 does not seem to have had any increase in 
> throughput for me. :(  I'll be moving back to the recommended -O2 after 
> these jobs finish.

Yes, it is possible. My problem was that I had no -O at all.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Slow tape write, no more than 60-70 MB/s in 5.0.3

[Cejka Rudolf]

Marcio Merlone wrote (2011/11/09):
> Em 09-11-2011 09:26, Cejka Rudolf escreveu:
> > Cejka Rudolf wrote (2011/11/09):
> >>does anybody have any idea, why after upgrade from Bacula 2.4.3 to 
> >> 5.0.3,
> >> tape write speed dropped down from about 70-120 MB/s (LTO-3 tape limits)
> >> to just 60-70 MB/s? I think that I have seen somewhere, that something
> >> (signatures?) has been moved from clients or director to storage daemon,
> >> but could not find it anymore. Or could be the problem somewhere else?
> > Oops, I'm sorry, I forgot to add -O3. After returning -O3, I can
> > see again speeds over 100 MB/s.
> 
> Please enlighten me (us?): Where did you add -03 to improve the speed? I 
> have an LTO-4 drive and get no more than 50~60 MB/s!

Hello, before compilation, instead of

./configure ...

try to use

CFLAGS=-O3 ./configure ...

and then see your tape drive more happy ;o)
If there are no bugs caused by -O3...

If Bacula behavior did not change since 2.4.3, the main limiting
factor to achieve higher throughput on LTO tapes, is in CRC32
computations performed between disk spool reads and tape writes.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Slow tape write, no more than 60-70 MB/s in 5.0.3

[Cejka Rudolf]

Cejka Rudolf wrote (2011/11/09):
> Hello,
>   does anybody have any idea, why after upgrade from Bacula 2.4.3 to 5.0.3,
> tape write speed dropped down from about 70-120 MB/s (LTO-3 tape limits)
> to just 60-70 MB/s? I think that I have seen somewhere, that something
> (signatures?) has been moved from clients or director to storage daemon,
> but could not find it anymore. Or could be the problem somewhere else?

Oops, I'm sorry, I forgot to add -O3. After returning -O3, I can
see again speeds over 100 MB/s.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Slow tape write, no more than 60-70 MB/s in 5.0.3

[Cejka Rudolf]

Hello,
  does anybody have any idea, why after upgrade from Bacula 2.4.3 to 5.0.3,
tape write speed dropped down from about 70-120 MB/s (LTO-3 tape limits)
to just 60-70 MB/s? I think that I have seen somewhere, that something
(signatures?) has been moved from clients or director to storage daemon,
but could not find it anymore. Or could be the problem somewhere else?

Thank you.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Restore error with 10 GbE server and 1 GbE client

[Cejka Rudolf]

Hello,
  is there anybody else, who has experienced similar restore problem as me?
If I use bacula server --(10 GbE)--> switch --(1 GbE)--> bacula client,
big restores with many small files (for example, 10 million files in
500 GB) end after a while with the following error (small restores under
10 thousand files are always good):

...
02-Aug 13:12 backup-sd JobId 62258: Error: bsock.c:393 Write error sending 
130568 bytes to client:IP:36643: ERR=Invalid argument
02-Aug 13:12 backup-sd JobId 62258: Fatal error: read.c:137 Error sending to 
File daemon. ERR=Invalid argument
02-Aug 13:12 backup-sd JobId 62258: Error: bsock.c:339 Socket has errors=1 on 
call to client:IP:36643
...

If I change bacula server connection from 10 GbE to just 1 GbE, all restores
go well. I think that it is related to data bursting, 10 GbE -> 1 GbE overload
and network buffers in operating system and in switch, TCP error reporting and
supposed bigger sensitivity of bacula to network errors.

Bacula server is FreeBSD 8.2-STABLE (Jul 30), network interfaces ix0 or em0.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
The must-attend event for mobile developers. Connect with experts. 
Get tools for creating Super Apps. See the latest technologies.
Sessions, hands-on labs, demos & much more. Register early & save!
http://p.sf.net/sfu/rim-blackberry-1
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Quantum Scalar i500 slow write speed

[Cejka Rudolf]

ekke85 wrote (2010/08/05):
> slow. It writes at 22mb/sec. The drives should be able to do a lot more
> then that. I have to backup 11TB that takes a couple of days to

Try tar -cf /dev/null /data-with-11-tb and you will see, if the
bottleneck is data source, or something else. How many files do
you have here?

> The spooling attribute was not enabled, I have enabled it now.

Yes, attribute spoolig is good thing too. However, data spooling
is more important, as i said before.

> ]# dd if=/dev/zero of=/home/bigfile bs=1M count=1
> 1+0 records in
> 1+0 records out
> 1048576 bytes (10 GB) copied, 189.551 seconds, 55.3 MB/s

Very slow. My spooling system says 372 MB/s, which is for spooling
4 parallel backups and one writing thread to LTO-3 tape. With 55 MB/s,
you can do at most one backup at a time, and the tape is still
"undernourished" (if write speed would equal to read speed)..

> This is writing that 10gb file with tar to tape:
> ]# time tar -czf /dev/Drive1 /home/bigfile
> tar: Removing leading `/' from member names

No, it was not tar, but a gzipped tar, where source was just zeros...
I expect, that you wrote on the tape just about 10 MB and you could
not say, what you tested more: Reading speed, processor speed, or
tape write speed.

> The 11Tb I have to backup is on a NetApp, the NetApp is mounted via
> NFS on the backup host and is getting the data from there to write to disk.

Which role NetApp plays here? NFS is not very good ticket to the speed...

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
The Palm PDK Hot Apps Program offers developers who use the
Plug-In Development Kit to bring their C/C++ apps to Palm for a share
of $1 Million in cash or HP Products. Visit us here for more details:
http://p.sf.net/sfu/dev2dev-palm
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Quantum Scalar i500 slow write speed

[Cejka Rudolf]

ekke85 wrote (2010/08/05):
> I do not have spooling on and I don't have software compression on.

It seems that you have LTO-3 drive(s). If you are not able to constantly
backup data atleast at rate 27 MB/s (HP LTO-3) or 40 MB/s (IBM LTO-3), you
need the spooling - which is a must. Note that required constant rates on
loaded nodes or small files are almost impossible. 

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
The Palm PDK Hot Apps Program offers developers who use the
Plug-In Development Kit to bring their C/C++ apps to Palm for a share
of $1 Million in cash or HP Products. Visit us here for more details:
http://p.sf.net/sfu/dev2dev-palm
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users