subject:"Re\: \[Bacula\-users\] backup strategy"

Re: [Bacula-users] Backup strategy advice

2016-03-25 Thread Ana Emília M . Arruda

Hello Eddie,

On Fri, Mar 25, 2016 at 3:03 PM, Eddie Appel  wrote:

> Ana,
>
> Thanks for the response, answers inline...
>

You're welcome :).


>
>
> On 03/24/2016 06:46 PM, Ana Emília M. Arruda wrote:
>
>  Hello Eddie,
>
> On Thu, Mar 24, 2016 at 7:13 PM, Eddie Appel  wrote:
>
>> Hello List,
>>
>> Forgive the long post, but I have a situation here that requires some
>> help...
>>
>> My current setup...
>> -- NetApp filer containing ~35TB of data to be backed up to tape
>> -- Quantumi40 w/ 2 LTO-5 drives
>> -- Quantum SL3 w/ 1 LTO-5 drive
>> -- HP DL-380G5 with 16GB RAM and 3 x spool hdd
>> -- CentOS 6.5 with kernel 2.6.32-431.23.3.el6.x86_64
>> -- Bacula 7.0.5
>> -- All data going to tape comes via NFS from filer with each tape drive
>> having a dedicated spool spindle
>> -- Full jobs run once per week, on Wednesdays
>> -- Incremental jobs run daily
>> -- Tapes leave daily to IronMountain, offsite for 1 week
>>
>> My current issues...
>> -- Tape auto-recycling is not happening consistently/reliably
>> -- Tapes now need to be offsite for longer periods of time.
>>
>> Given the amount of hand holding I have had to do with bacula over the
>> last couple of years, I am convinced that my configs are not very
>> helpful or accurate...
>>
>> So my question is this...
>>
>> What would be the best setup for pools/retention/schedules to accomplish:
>> -- Manage volumes/pools between both autochanger devices
>>
>
> Having the same media type for both autochanger devices will let you
> manage volumes in both autochangers. Update slots command will update the
> device information for the tapes.
>
>
> Having the same media type between the two changers actually produced a
> scenario wherein bacula became confused with slot numbers and pool
> inventory. If memory serves, it was Kern who suggested I use LTOa and LTOb
> as media types between the changers to avoid this contention.
>

Now the update slots [scan] command (since versions 5.0.0 if I am not
wrong) permits the update of slot number, inchager flag and StorageId used
by the volume. This way you can have the same media type between
autochangers. Indeed, if you need to use a volume interchangeably between
both tape libraries you should use the same media type. Having this kind
of configuration helps you if you have problems with one of your
autochangers and need to redirect all your backups to the one working fine.

Unfortunately, it is not possible (by now) to configure pools or jobs to
use more than one storage for backups (it would be a good idea :)). So you
will need to configure one of your autochangers to be used by a specific
job or pool. Hopefully, you can configure the Storage to be used by a
job/pool on your schedule resource. I particularly like this situation
because it is very flexible when you need to change the storage for a
specific job/pool.


> If however, I can mange a single set of pools/volumes/schedules across
> both changers, that would be ideal...
> Would this be a single bacula-sd with 2 storage resources, or 2 individual
> bacula-sd with a single storage resource each?
>
>  My current storage device setup...
>
> Autochanger {
>   Name = Quantumi40
>   Device = Drive0, Drive1
>   Changer Command = "/usr/libexec/bacula/mtx-changer %c %o %S %a %d"
>   Changer Device = /dev/quantumi40
> }
> Autochanger {
>   Name = Quantumsl3
>   Device = Drive2
>   Changer Command = "/usr/libexec/bacula/mtx-changer %c %o %S %a %d"
>   Changer Device = /dev/quantumsl3
> }
> Device {
>   Name = Drive{0|1|2}
>   Drive Index = {0|1}
>   Media Type = LTO5{a|b}
>   Device Type = tape
>   Archive Device = /dev/tape{0|1|2}
>   AutomaticMount = yes;
>   AlwaysOpen = yes;
>   RemovableMedia = yes;
>   RandomAccess = no;
>   AutoChanger = yes
>   LabelMedia = yes;
>   Maximum Changer Wait = 9000
>   Maximum Concurrent Jobs = 1
>   Spool Directory = /spool/{0|1|2}
>   Maximum Spool Size = 100gb
>   Minimum block size = 1048576
>   Maximum block size = 1048576
>   Hardware End of Medium = yes
>   Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
> }
>
> -- Tapes offsite daily, to return 31 days later for recycling
>> -- Tapes written daily/weekly via incremental and/or diff jobs
>> -- Tapes written weekly/monthly via Full jobs
>>
>
> I would have 3 pools: DailyPool, WeeklyPool, MonthlyPool (I am supposing
> you have different retention periods for this. You just specified a 31 days
> retention for daily backups?).
>
> To phrase it another way, I need to have a full set of the data offsite
> for 31 days at a time. Where "full set" is a full week's worth of tapes,
> sent out daily after the Incremental/Full jobs run. I suspect switching to
> a monthly Full, with weekly Diffs and Daily Incremental cycle would be my
> way forward from here...
>
> I think this setup would require a total of 6 pools, 3 for each
> autochanger...
>

I think if you use the same media type and tells bacula the storage to be
used on the

Re: [Bacula-users] Backup strategy advice

2016-03-25 Thread Eddie Appel




On 03/25/2016 11:14 AM, Heitor Faria wrote:


Having the same media type between the two changers actually
produced a scenario wherein bacula became confused with slot
numbers and pool inventory. If memory serves, it was Kern who
suggested I use LTOa and LTOb as media types between the changers
to avoid this contention.

Hello, Eddie: sorry by para shooting this. I think you must use at 
least two different pools if you have the same media type for both 
autochangers (E,g,: Daily-Changer1, Daily-Changer2 etc.), or different 
media types withing the same pool.
Hi Hector, no worries this is actually a Great idea, I will test this 
out over this weekend and report back...

Different media types in the same pool!

so then I would not need 6 pools, but 3 with both LTO5a and LTO5b volumes...



If however, I can mange a single set of pools/volumes/schedules
across both changers, that would be ideal...
Would this be a single bacula-sd with 2 storage resources, or 2
individual bacula-sd with a single storage resource each?

I think this just doesn't matter.
I believe you are correct here, as my testing revealed no difference 
between the two different setups...



Thanks,

--eddie


Regards,
===
Heitor Medrado de Faria - LPIC-III | ITIL-F |  Bacula Systems 
Certified Administrator II

Do you need Bacula training? http://bacula.us/video-classes/
+55 61 8268-4220 
Site: http://bacula.us FB: heitor.faria 


===



Cloud Services for Business www.j2.com
j2 | eFax | eVoice | FuseMail | Campaigner | KeepItSafe | Onebox


This email, its contents and attachments contain information from j2 Global, 
Inc. and/or its affiliates which may be privileged, confidential or otherwise 
protected from disclosure. The information is intended to be for the 
addressee(s) only. If you are not an addressee, any disclosure, copy, 
distribution, or use of the contents of this message is prohibited. If you have 
received this email in error please notify the sender by reply e-mail and 
delete the original message and any copies. (c) 2015 j2 Global, Inc. All rights 
reserved. eFax, eVoice, Campaigner, FuseMail, KeepItSafe, and Onebox are 
registered trademarks of j2 Global, Inc. and its affiliates.
--
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=278785351=/4140___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy advice

2016-03-25 Thread Heitor Faria

> Having the same media type between the two changers actually produced a 
> scenario
> wherein bacula became confused with slot numbers and pool inventory. If memory
> serves, it was Kern who suggested I use LTOa and LTOb as media types between
> the changers to avoid this contention.
Hello, Eddie: sorry by para shooting this. I think you must use at least two 
different pools if you have the same media type for both autochangers (E,g,: 
Daily-Changer1, Daily-Changer2 etc.), or different media types withing the same 
pool. 

> If however, I can mange a single set of pools/volumes/schedules across both
> changers, that would be ideal...
> Would this be a single bacula-sd with 2 storage resources, or 2 individual
> bacula-sd with a single storage resource each?
I think this just doesn't matter. 

Regards, 
=== 
Heitor Medrado de Faria - LPIC-III | ITIL-F | Bacula Systems Certified 
Administrator II 
Do you need Bacula training? http://bacula.us/video-classes/ 
+55 61 8268-4220 
Site: http://bacula.us FB: heitor.faria 
=== 
--
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=278785351=/4140___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy advice

2016-03-25 Thread Eddie Appel


Ana,

Thanks for the response, answers inline...

On 03/24/2016 06:46 PM, Ana Emília M. Arruda wrote:

 Hello Eddie,

On Thu, Mar 24, 2016 at 7:13 PM, Eddie Appel > wrote:


Hello List,

Forgive the long post, but I have a situation here that requires some
help...

My current setup...
-- NetApp filer containing ~35TB of data to be backed up to tape
-- Quantumi40 w/ 2 LTO-5 drives
-- Quantum SL3 w/ 1 LTO-5 drive
-- HP DL-380G5 with 16GB RAM and 3 x spool hdd
-- CentOS 6.5 with kernel 2.6.32-431.23.3.el6.x86_64
-- Bacula 7.0.5
-- All data going to tape comes via NFS from filer with each tape
drive
having a dedicated spool spindle
-- Full jobs run once per week, on Wednesdays
-- Incremental jobs run daily
-- Tapes leave daily to IronMountain, offsite for 1 week

My current issues...
-- Tape auto-recycling is not happening consistently/reliably
-- Tapes now need to be offsite for longer periods of time.

Given the amount of hand holding I have had to do with bacula over the
last couple of years, I am convinced that my configs are not very
helpful or accurate...

So my question is this...

What would be the best setup for pools/retention/schedules to
accomplish:
-- Manage volumes/pools between both autochanger devices


Having the same media type for both autochanger devices will let you 
manage volumes in both autochangers. Update slots command will update 
the device information for the tapes.
Having the same media type between the two changers actually produced a 
scenario wherein bacula became confused with slot numbers and pool 
inventory. If memory serves, it was Kern who suggested I use LTOa and 
LTOb as media types between the changers to avoid this contention.


If however, I can mange a single set of pools/volumes/schedules across 
both changers, that would be ideal...
Would this be a single bacula-sd with 2 storage resources, or 2 
individual bacula-sd with a single storage resource each?


 My current storage device setup...

Autochanger {
  Name = Quantumi40
  Device = Drive0, Drive1
  Changer Command = "/usr/libexec/bacula/mtx-changer %c %o %S %a %d"
  Changer Device = /dev/quantumi40
}
Autochanger {
  Name = Quantumsl3
  Device = Drive2
  Changer Command = "/usr/libexec/bacula/mtx-changer %c %o %S %a %d"
  Changer Device = /dev/quantumsl3
}
Device {
  Name = Drive{0|1|2}
  Drive Index = {0|1}
  Media Type = LTO5{a|b}
  Device Type = tape
  Archive Device = /dev/tape{0|1|2}
  AutomaticMount = yes;
  AlwaysOpen = yes;
  RemovableMedia = yes;
  RandomAccess = no;
  AutoChanger = yes
  LabelMedia = yes;
  Maximum Changer Wait = 9000
  Maximum Concurrent Jobs = 1
  Spool Directory = /spool/{0|1|2}
  Maximum Spool Size = 100gb
  Minimum block size = 1048576
  Maximum block size = 1048576
  Hardware End of Medium = yes
  Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
}


-- Tapes offsite daily, to return 31 days later for recycling
-- Tapes written daily/weekly via incremental and/or diff jobs
-- Tapes written weekly/monthly via Full jobs


I would have 3 pools: DailyPool, WeeklyPool, MonthlyPool (I am 
supposing you have different retention periods for this. You just 
specified a 31 days retention for daily backups?).


To phrase it another way, I need to have a full set of the data offsite 
for 31 days at a time. Where "full set" is a full week's worth of tapes, 
sent out daily after the Incremental/Full jobs run. I suspect switching 
to a monthly Full, with weekly Diffs and Daily Incremental cycle would 
be my way forward from here...


I think this setup would require a total of 6 pools, 3 for each 
autochanger...


Here's my current thinking...

 Pool {
  Name = {$i40|$sl3}-{$full|$diff|$inc}Pool
  Pool Type = Backup
  Recycle = yes
  Recycle Oldest Volume = yes
  Auto Prune = yes
  ScratchPool = Scratch
  RecyclePool = Scratch
  Volume Retention = {31|14|7} days
}

Which would make the Schedule(s) something like... Unless there's a way 
to have a unified schedule/pools for both changers...


Schedule {
  Name = "i40-Daily"
  Run = Level=Incremental Pool=i40-IncPool tue-sun at 18:00
  Run = Level=Full Pool=i40-FullPool 1st mon at 18:00
  Run = Level=Differential Pool=i40-DiffPool 2nd mon at 18:00
  Run = Level=Differential Pool=i40-DiffPool 3rd mon at 18:00
  Run = Level=Differential Pool=i40-DiffPool 4th mon at 18:00
  Run = Level=Differential Pool=i40-DiffPool 5th mon at 18:00
}
Schedule {
  Name = "sl3-Daily"
  Run = Level=Incremental Pool=sl3-IncPool sat-thu at 03:00
  Run = Level=Full Pool=sl3-FullPool 1st fri at 03:00
  Run = Level=Differential Pool=sl3-DiffPool 2nd fre at 03:00
  Run = Level=Differential Pool=sl3-DiffPool 3rd fri at 03:00
  Run = Level=Differential Pool=sl3-DiffPool 4th fri at 03:00
  Run = Level=Differential Pool=sl3-DiffPool 5th fri at 03:00
}


Pool {
  Name = DailyPool
  ...
  Volume

Re: [Bacula-users] Backup strategy advice

2016-03-24 Thread Ana Emília M . Arruda

Hello Eddie,

On Thu, Mar 24, 2016 at 7:13 PM, Eddie Appel  wrote:

> Hello List,
>
> Forgive the long post, but I have a situation here that requires some
> help...
>
> My current setup...
> -- NetApp filer containing ~35TB of data to be backed up to tape
> -- Quantumi40 w/ 2 LTO-5 drives
> -- Quantum SL3 w/ 1 LTO-5 drive
> -- HP DL-380G5 with 16GB RAM and 3 x spool hdd
> -- CentOS 6.5 with kernel 2.6.32-431.23.3.el6.x86_64
> -- Bacula 7.0.5
> -- All data going to tape comes via NFS from filer with each tape drive
> having a dedicated spool spindle
> -- Full jobs run once per week, on Wednesdays
> -- Incremental jobs run daily
> -- Tapes leave daily to IronMountain, offsite for 1 week
>
> My current issues...
> -- Tape auto-recycling is not happening consistently/reliably
> -- Tapes now need to be offsite for longer periods of time.
>
> Given the amount of hand holding I have had to do with bacula over the
> last couple of years, I am convinced that my configs are not very
> helpful or accurate...
>
> So my question is this...
>
> What would be the best setup for pools/retention/schedules to accomplish:
> -- Manage volumes/pools between both autochanger devices
>

Having the same media type for both autochanger devices will let you
manage volumes in both autochangers. Update slots command will update the
device information for the tapes.


> -- Tapes offsite daily, to return 31 days later for recycling
> -- Tapes written daily/weekly via incremental and/or diff jobs
> -- Tapes written weekly/monthly via Full jobs
>

I would have 3 pools: DailyPool, WeeklyPool, MonthlyPool (I am supposing
you have different retention periods for this. You just specified a 31 days
retention for daily backups?).

Pool {
  Name = DailyPool
  ...
  Volume Retention = 31 days
  ...
}


> -- Automating the 'update slots' and 'volume status' changes when tapes
> leave daily
>

I have an admin job that runs every day immediately before the first
backup job. And this admin job just runs an update slots from bconsole.


> -- Restore validation tests performed quarterly
>

Do you want this to happen automatically? I do not understand your need
here... But if this is the case, you can do this with some scripting and
bls/bextract.

I have a 30 days retention for daily diff backups, 40 day retention for
weekly full backups and 5 years retention for monthly full backups. Please
let me know if you have a similar case.

Best regards,
Ana


>
> Let me know if you need further information, and thanks in advance for
> reading and any help!
>
>
>
> --eddie
>
>
> Cloud Services for Business www.j2.com
> j2 | eFax | eVoice | FuseMail | Campaigner | KeepItSafe | Onebox
>
>
> This email, its contents and attachments contain information from j2
> Global, Inc. and/or its affiliates which may be privileged, confidential or
> otherwise protected from disclosure. The information is intended to be for
> the addressee(s) only. If you are not an addressee, any disclosure, copy,
> distribution, or use of the contents of this message is prohibited. If you
> have received this email in error please notify the sender by reply e-mail
> and delete the original message and any copies. (c) 2015 j2 Global, Inc.
> All rights reserved. eFax, eVoice, Campaigner, FuseMail, KeepItSafe, and
> Onebox are registered trademarks of j2 Global, Inc. and its affiliates.
>
>
> --
> 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=278785351=/4140
> ___
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
--
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=278785351=/4140___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy with Bacula

2013-03-28 Thread Geert Stappers

Op 2013-03-27 om 13:12 schreef Paul Mather:
 On Mar 26, 2013, at 11:44 PM, Bill Arlofski wrote:
  On 03/26/13 22:33, Wood Peter wrote:
  Hi,
  
  I'm planning to use Bacula to do file level backup of about 15 Linux
  systems. Total backup size is about 2TB.
  
  For Bacula server I'm thinking to buy Dell PE R520 with 24TB internal
  storage (8x 3TB disks) and use virtual tapes.
  
  Hi Peter.   First, you're going to want some RAID level on that server and
  RAID0 is not (IMHO) RAID at all.  :)
  
  At a minimum you're going to want to set up (for example) RAID5 in the 
  server
  which will give you a maximum of 7 x 3 = 21 TB since the equivalent of 1 
  full
  drive is used for parity, spread across all of the drives in a RAID5 array.
  
  Having said that, RAID5 does not have the best write speeds, but other RAID
  levels that will give more redundancy and better write speeds will use
  significantly more drives and give you less total storage. You may need to
  spend some time to consider your redundancy and read/write throughput
  requirements.
  
  Also, you will generally want to configure at least one drive as a hot spare
  so the RAID controller can automatically and immediately fail over to it in
  the case of a drive failure in the array.
  
  So that takes away at least 3 more TB, so now your down to 18TB total 
  storage
  with a minimally configured RAID5 array with 1 hot spare.
  
  Just some things to consider. :)
 
 
 Another thing to consider is that with large capacity drives (3 TB)
 combined into large RAID-5 arrays there is an increased likelihood of
 a catastrophic array failure during an array rebuild due to an initial
 drive failure.  For this reason, RAID-6 would be preferred for such
 large arrays.
 
 Reliability of large RAID arrays is one of the motivations
 behind raidz3 (triple-parity redundancy) in ZFS.  See, e.g.,
 http://queue.acm.org/detail.cfm?id=1670144 for details.
 

Thanks for the advice on RAID configuration. Now we can go back to
  24TB internal storage (8x 3TB disks) and use virtual tapes.
so (cheap) disks as alternative to backup tapes.


Groeten Stappers
--
Own the Future-Intelreg; Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest.
Compete for recognition, cash, and the chance to get your game 
on Steam. $5K grand prize plus 10 genre and skill prizes. 
Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy with Bacula

2013-03-28 Thread Peter Wood

Thanks all for the help.

To use the space more efficiently what I decided to do is use vtapes, limit
the file size to 10GB and after a full backup just scp the related files to
a server in the office. I still have to do the proof of concept setup but
from what I'm reading this should be doable.

Bill, great article. Thanks.

-- Peter


On Thu, Mar 28, 2013 at 12:25 AM, Geert Stappers 
geert.stapp...@vanadgroup.com wrote:

 Op 2013-03-27 om 13:12 schreef Paul Mather:
  On Mar 26, 2013, at 11:44 PM, Bill Arlofski wrote:
   On 03/26/13 22:33, Wood Peter wrote:
   Hi,
  
   I'm planning to use Bacula to do file level backup of about 15 Linux
   systems. Total backup size is about 2TB.
  
   For Bacula server I'm thinking to buy Dell PE R520 with 24TB internal
   storage (8x 3TB disks) and use virtual tapes.
  
   Hi Peter.   First, you're going to want some RAID level on that server
 and
   RAID0 is not (IMHO) RAID at all.  :)
  
   At a minimum you're going to want to set up (for example) RAID5 in the
 server
   which will give you a maximum of 7 x 3 = 21 TB since the equivalent of
 1 full
   drive is used for parity, spread across all of the drives in a RAID5
 array.
  
   Having said that, RAID5 does not have the best write speeds, but other
 RAID
   levels that will give more redundancy and better write speeds will use
   significantly more drives and give you less total storage. You may
 need to
   spend some time to consider your redundancy and read/write throughput
   requirements.
  
   Also, you will generally want to configure at least one drive as a hot
 spare
   so the RAID controller can automatically and immediately fail over to
 it in
   the case of a drive failure in the array.
  
   So that takes away at least 3 more TB, so now your down to 18TB total
 storage
   with a minimally configured RAID5 array with 1 hot spare.
  
   Just some things to consider. :)
 
 
  Another thing to consider is that with large capacity drives (3 TB)
  combined into large RAID-5 arrays there is an increased likelihood of
  a catastrophic array failure during an array rebuild due to an initial
  drive failure.  For this reason, RAID-6 would be preferred for such
  large arrays.
 
  Reliability of large RAID arrays is one of the motivations
  behind raidz3 (triple-parity redundancy) in ZFS.  See, e.g.,
  http://queue.acm.org/detail.cfm?id=1670144 for details.
 

 Thanks for the advice on RAID configuration. Now we can go back to
   24TB internal storage (8x 3TB disks) and use virtual tapes.
 so (cheap) disks as alternative to backup tapes.


 Groeten Stappers

 --
 Own the Future-Intelreg; Level Up Game Demo Contest 2013
 Rise to greatness in Intel's independent game demo contest.
 Compete for recognition, cash, and the chance to get your game
 on Steam. $5K grand prize plus 10 genre and skill prizes.
 Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
 ___
 Bacula-users mailing list
 Bacula-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/bacula-users

--
Own the Future-Intel(R) Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest. Compete 
for recognition, cash, and the chance to get your game on Steam. 
$5K grand prize plus 10 genre and skill prizes. Submit your demo 
by 6/6/13. http://altfarm.mediaplex.com/ad/ck/12124-176961-30367-2___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy with Bacula

2013-03-27 Thread Paul Mather

On Mar 26, 2013, at 11:44 PM, Bill Arlofski waa-bac...@revpol.com wrote:

 On 03/26/13 22:33, Wood Peter wrote:
 Hi,
 
 I'm planning to use Bacula to do file level backup of about 15 Linux
 systems. Total backup size is about 2TB.
 
 For Bacula server I'm thinking to buy Dell PE R520 with 24TB internal
 storage (8x 3TB disks) and use virtual tapes.
 
 Hi Peter.   First, you're going to want some RAID level on that server and
 RAID0 is not (IMHO) RAID at all.  :)
 
 At a minimum you're going to want to set up (for example) RAID5 in the server
 which will give you a maximum of 7 x 3 = 21 TB since the equivalent of 1 full
 drive is used for parity, spread across all of the drives in a RAID5 array.
 
 Having said that, RAID5 does not have the best write speeds, but other RAID
 levels that will give more redundancy and better write speeds will use
 significantly more drives and give you less total storage. You may need to
 spend some time to consider your redundancy and read/write throughput
 requirements.
 
 Also, you will generally want to configure at least one drive as a hot spare
 so the RAID controller can automatically and immediately fail over to it in
 the case of a drive failure in the array.
 
 So that takes away at least 3 more TB, so now your down to 18TB total storage
 with a minimally configured RAID5 array with 1 hot spare.
 
 Just some things to consider. :)


Another thing to consider is that with large capacity drives (3 TB) combined 
into large RAID-5 arrays there is an increased likelihood of a catastrophic 
array failure during an array rebuild due to an initial drive failure.  For 
this reason, RAID-6 would be preferred for such large arrays.

Reliability of large RAID arrays is one of the motivations behind raidz3 
(triple-parity redundancy) in ZFS.  See, e.g., 
http://queue.acm.org/detail.cfm?id=1670144 for details.

Cheers,

Paul.


--
Own the Future-Intelreg; Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest.
Compete for recognition, cash, and the chance to get your game 
on Steam. $5K grand prize plus 10 genre and skill prizes. 
Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy with Bacula

2013-03-26 Thread Bill Arlofski

On 03/26/13 22:33, Wood Peter wrote:
 Hi,
 
 I'm planning to use Bacula to do file level backup of about 15 Linux
 systems. Total backup size is about 2TB.
 
 For Bacula server I'm thinking to buy Dell PE R520 with 24TB internal
 storage (8x 3TB disks) and use virtual tapes.

Hi Peter.   First, you're going to want some RAID level on that server and
RAID0 is not (IMHO) RAID at all.  :)

At a minimum you're going to want to set up (for example) RAID5 in the server
which will give you a maximum of 7 x 3 = 21 TB since the equivalent of 1 full
drive is used for parity, spread across all of the drives in a RAID5 array.

Having said that, RAID5 does not have the best write speeds, but other RAID
levels that will give more redundancy and better write speeds will use
significantly more drives and give you less total storage. You may need to
spend some time to consider your redundancy and read/write throughput
requirements.

Also, you will generally want to configure at least one drive as a hot spare
so the RAID controller can automatically and immediately fail over to it in
the case of a drive failure in the array.

So that takes away at least 3 more TB, so now your down to 18TB total storage
with a minimally configured RAID5 array with 1 hot spare.

Just some things to consider. :)


 - I'm backing up mostly text configuration files.
   How many full backups can I expect to fit in this amount of storage?

Completely depends on how compressible the data is, or if you care to use
compression at all. Consider that compression takes place on the client, so
that can add significantly to the clients' load on a busy client.

 For offsite backup I was thinking to use external tape drive and
 replicate the full backup to tape. Once a month I'll go to the colo
 where the server is, get the tape and bring it to the office.

Bacula can do Copy Jobs  so, for example, you run your backups to your file
volumes om your array, then the copy jobs run and copy the data to tapes that
you can remove and take off-site.

Once a month for off-sites is a bit long of a duration for one thing, and
you'd need to consider the amount of data and be sure your off-site tape(s)
had enough capacity. Don't want to be running to your colo at strange, random
times of the night or weekend to make sure the backup jobs run on schedule. :)

 - But, we have 100Mbps from the office to the colo.
   Is there a way to copy the full backup to the office over the
 network and completely avoid tapes?

Yes.  You can set up your Bacula server at your office, and configure your
co-located servers to write to the Bacula server over VPN, or stunnel (poor
man's VPN, but works quite well)

 If that's not possible can I attach external storage to the server
 with hot swappable hard drives and use them as tape replacement?

Yes. See http://www.revpol.com/node/140 for just such an example using eSATA
drives in an external drive caddy.

 It's been a while since I last used Bacula so your help is appreciated.

Hope this helps. :)

-- 
Bill Arlofski
Reverse Polarity, LLC


--
Own the Future-Intelreg; Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest.
Compete for recognition, cash, and the chance to get your game 
on Steam. $5K grand prize plus 10 genre and skill prizes. 
Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy for Laptops

2010-04-02 Thread ebollengier


Hi Anthony,


Avarca, Anthony wrote:
 
 All,
 
 I'm using bacula to backup desktop and laptop clients. The desktops work
 well with a schedule, but laptops are another story. Does anyone have a
 strategy to backup laptops? Is it possible to have the user trigger a
 backup?
 
 Any feedback is welcomed.
 

I'm using a small script that runs ping every 10min from the Director to see
if the laptop is connected,
and when I get an answer, I start a backup with a bconsole command. Once
it's done, I
wait for something like 6 hours and I restart the detection part.

Simple and very flexible.

Bye

PS: Is Bacula used at your work? Then, can you fill a testimonial on
www.bacula.org? 

-- 
View this message in context: 
http://old.nabble.com/Backup-Strategy-for-Laptops-tp28121373p28121850.html
Sent from the Bacula - Users mailing list archive at Nabble.com.


--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy for Laptops

2010-04-02 Thread Jayson Broughton

We back up about 200 workstation clients currently backing up, I would
say out of that; 80 of them are laptops.  I found that rescheduling the
backup on failure seems to work with a 10 hour window.  This is added to
a JobDefs as such:
This way, when a user comes in later on in the day and missed their
first backup, if they are online on the next 'reschedule', the backup
will kick off and life will go on as normal.  I also have the priority
levels of laptops set higher on the foodchain than desktops.  So that
laptop clients are backed up before desktop clients (users can leave
their desktops on at night).

Reschedule On Error = yes
  Reschedule Interval = 1 hours
  Reschedule Times = 10


Hope this helps!

~Jayson



On Fri, 2010-04-02 at 11:15 -0600, Avarca, Anthony wrote:
 All,
 
 
 I'm using bacula to backup desktop and laptop clients. The desktops
 work well with a schedule, but laptops are another story. Does anyone
 have a strategy to backup laptops? Is it possible to have the user
 trigger a backup?
 
 
 Any feedback is welcomed.
 
 
 
 Anthony Avarca
 aava...@anl.gov
 630.252.4940
 
 
 --
 Download Intel#174; Parallel Studio Eval
 Try the new software tools for yourself. Speed compiling, find bugs
 proactively, and fine-tune applications for parallel performance.
 See why Intel Parallel Studio got high marks during beta.
 http://p.sf.net/sfu/intel-sw-dev
 ___ Bacula-users mailing list 
 Bacula-users@lists.sourceforge.net 
 https://lists.sourceforge.net/lists/listinfo/bacula-users



The information in this electronic mail message and any attached files is 
confidential and may be legally privileged.  If you are not the intended 
recipient, delete this message and contact the sender immediately.  Access to 
this message by anyone other than its intended recipient is unauthorized.  You 
must not use or disseminate this information as it is proprietary property of 
the True companies.  Communications on or through the True companies' computer 
systems may be monitored or recorded to secure effective system operation and 
for other lawful purposes.  Thank you.


--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy for Laptops

2010-04-02 Thread Gavin McCullagh

Hi,

On Fri, 02 Apr 2010, Avarca, Anthony wrote:

 I'm using bacula to backup desktop and laptop clients. The desktops work
 well with a schedule, but laptops are another story. Does anyone have a
 strategy to backup laptops? Is it possible to have the user trigger a
 backup?

It's not the prettiest thing in the world, but we get the user to trigger a
backup using bconsole.  Firstly, we allow console connections from the
client computer to the director, strictly only for the resources which are
relevant to that client.  When the user decides to run a backup, 

1. Start bconsole
2. Type runreturn
3. Type exitreturn
4. The messages are emailed to the user so they know when the job is
   finished.

the default backup triggered in our case is always incremental and
accurate.  We then run virtual full backups on a schedule.

This scheme has solved a long-term problem for us.  It's a laptop for which
a full backup takes over 5 hours.  The user simply can't wait that long.
The incremental by contrast takes about 1 hour, which is workable and he
can work on it during the backup, albeit a bit slow.

Gavin



--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy for Laptops

2010-04-02 Thread Avi Rozen

Gavin McCullagh wrote:
 1. Start bconsole
 2. Type runreturn
 3. Type exitreturn
 4. The messages are emailed to the user so they know when the job is
finished.
   

Assuming the laptop is running a Debian based Linux distro: can't this
be automated by running a backup script from  /etc/networks/interfaces
using the post-up directive, or by placing such a script under
/etc/networks/post-up.d ?

Avi.


--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy for Laptops

2010-04-02 Thread Gavin McCullagh

Hi,

On Fri, 02 Apr 2010, Avi Rozen wrote:

 Gavin McCullagh wrote:
  1. Start bconsole
  2. Type runreturn
  3. Type exitreturn
  4. The messages are emailed to the user so they know when the job is
 finished.
 
 Assuming the laptop is running a Debian based Linux distro: can't this
 be automated by running a backup script from  /etc/networks/interfaces
 using the post-up directive, or by placing such a script under
 /etc/networks/post-up.d ?

If it were a debian-based distro, perhaps.  However, this laptop runs
Windows.  I daresay you could script something similar in Windows too
though.

The other disadvantage of that scheme is that I don't really want a backup
to trigger every time the user appears on the network (and a backup is
due).  I want them to be able to decide I'm happy to plug in and leave the
laptop on the network for an hour now and I accept it will run a little
slowly during the time.  

In our particular case, if backups ran automatically, we'd have failed
backups because the laptop was unplugged midway through.  We'd also have
complaints that the laptop was running slow at times.

Of course, differently people will have different requirements here.

Gavin


--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy with new big BaculaServer (HD+Tape backup)

2008-03-03 Thread Drew Bentley

On Mon, Mar 3, 2008 at 7:56 AM, thomas [EMAIL PROTECTED] wrote:
 Hi,

  we use bacula a long time(6 years) to backup linux and windows server.
  It works create!!!

  Now one harddisk (Raid5) is defect in the bacula server and we consider
  to buy a new server.
  I think we can buy a large server with tapelibrary (DELL TL2000 LTO3).

  Our consideration are:
  - We want to backup to disk
  - We want hold the backups 7 days on Disk
  - We want a full backup each day (300GB)
  - We want each full harddisk backup are copied to a tape in the tapelibrary

  I think we will have a harddisk (RAID5) with 4,5 TB.
  So we can hold a lot of full backups on harddisk.
  4500GB(total) - 300GB(one fullBCK) = 4200GB
  4200GB / 300GB = 14 days

  We know how to backup to tape, and we know how to backup to
  volumes(filesystem).

  BUT, how can we backup to harddisk and then to tape in one backup job?
  We dont want to run the same job twice with different device!

  The tapelibrary can hold 24 LTO3 (400/800GB) tapes.
  So we have
  5 tapes = Daily
  4 tapes = weekly
  12 tapes = yearly
  3 tapes = manual backup

  Regards,
  Thomas


Sounds like you want to do some type of offsite vault job, but just
dump from disk to tape. Honestly I think doing a full backup each day
is just a waste of space, on tape and disk. You might want to consider
maybe doing a full backup each week with incrementals in between.
Maybe even setting up that once a week job to dump off onto tape
instead of disk. Perhaps have the Full weekly's on tape and the
incrementals on disk. For those quick restores or recoveries, you may
only need to recover from disk for a quick restore, anything requiring
more or a full recovery would require the tape, etc.

I'm not sure how busy your systems are but 300GB full dump each day is
just not a good way to setup backups. And even then, you may want to
consider doing one full backup per month with differentials 3-4 other
times with incrementals inbetween. this will save you time, money and
resources for your backup needs.

Your best option is to first determine how often your data changes. A
good daily average that a lot of vendors will use is 3% daily change.
Then determine what type of retention policy you want for your backups
and each host. Some hosts may not need a year retention, some might.
Some may only really require that you keep a month or less retention
for.

From that, then take the best approach to backing up your data saving
yourself that time, money and resources. A good backup solution and
plan is like artwork, take the best approach that will make sense in
the long run, not just what's convenient for you now.

-Drew

-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy : One Full Backup + 364 Diferentials

2007-11-21 Thread Augusto Camarotti

Thanks verymuch for then explanation Martin. I see your point. In fact,
we are not so worried about disaster recovery, since this files I mention in
this backup are text documents.
And generally, users ask restore of some files in particular, not
folders. Besides, they usually don't tend do delete files, they just let
them pile up in the shared folders :D
I guess using differential backup all year will not be a problem.

Thanks again Martin.

I would like if anyone else has another critic to this strategy.

Regards,

Augusto


On Nov 20, 2007 2:05 PM, Martin Simmons [EMAIL PROTECTED] wrote:

  On Tue, 20 Nov 2007 13:14:24 -0300, Augusto Camarotti said:
 
  Hi Mark, thanks for the reply
 
  When I was saying the generally 3 GB of data was added, I was counting
 on
  data changed, since new data is being added when a change is made. I
 just
  don`t express it right.

 Ah, OK.


  I didnt understand why it'll be a problem not tracking deleted files,
 since
  the reason of a backup is to save data that have been added, and not
  deleted.

 It depends on the purpose of your backup, e.g. is it to allow for restores
 of
 corrupted/lost files or is it for disaster recovery?

 Some people don't like Differentials for disaster recovery.  Suppose the
 Full
 backup contains file A and B.  Someone decides that file A is junk so they
 delete it and they create file C instead.  The next Differential backup
 will
 contain file C and nothing else.  This looks OK because all files are
 saved in
 some backup.

 However, when your disk gets trashed and you have to do a restore from the
 Full and Differential backups.  The problem is that you will get back all
 three files A, B and C.

 __Martin


 
  Waiting for new sugestions or critics,
 
  Augusto
 
  On Nov 20, 2007 8:09 AM, Martin Simmons [EMAIL PROTECTED] wrote:
 
On Mon, 19 Nov 2007 18:55:30 -0300, Augusto Camarotti said:
   
Nowadays we have a backup strategy like this :
Daily Backup = Monday-Thursday = Full Backups with individual tapes
 for
   each
weekday.
Weekly Backup = 2nd-5th Friday =3D Full Backups with individual
 tapes
   for 7
Fridays
Montly Backup = 1st Friday = Full Backups with individual tapes for
 12
months
   
This way I'm using 23 tapes.
   
Then I was reading about Differential backups, and this strategy
 came
   into
my mid :
   
- One Full Backup on the first day of the year, say 2nd January
 (since
   1st
it's a holyday) on disk, in the storage machine. The data would be
   stored in
a tape, which would be taken offsite.
- 364 Differential backups, one for each day of the year. Using the
 23
   tapes
to do it.
   
Since the amount of data added on a year it's a maximum of 3 GB, the
 23
tapes will be enough to backup every day on a year.
  
   Beware that the amount of new data added in a year is not sufficient
 to
   calculate this.  You also need to include the amount of data *changed*
   since
   the beginning of the year, because that will be backed up each day.
  
  
Well, as I didn't find any recomendations to this strategy, I was in
   doubt
if it would be fine to do this way.
   
Suggestions or Critics are welcome.
  
   Differential/incremental backups have a disadvantage that they do not
   track
   deleted files, which could be a major problem over a whole year.
  
   __Martin
  

-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2007-11-21 Thread Sebastian Perkins


Thanks for the information : I'm using a disk based solution, but after 
a few days worth's of Incremental backup, it looks like I could do a 
diff backup every weekend (even with 2To backup space is still a concern 
!).

-- 
Sebastian Perkins
Responsable Informatique
[EMAIL PROTECTED]

ALPES CONTROLES
3 impasse des prairies
74940 Annecy le Vieux
Tél. 04.50.64.11.46 - Fax. 04.50.64.23.83
http://www.alpes-controles.fr 



Marek Simon a écrit :
 Differential and Incremental  levels differ in this thing: Incremental 
 backup takes the files which has changed since any previous backup of 
 any level. Differental backup takes all files, which has changed since 
 previous Full (and only Full) backup.
 So with your current strategy (one Full per month, Incr other days), 
 if you would have to restore files just a day before new full backup, 
 you must restore from full backup volume and then from 30 records in 
 volumes containing those Incremental ones, which takes longer time. 
 You see this strategy is not usable if you need to backup to tapes.
 If you have some Differential backup (lets say) every weekend, in 
 worst case you must to restore from 1 Full backup, one Differential 
 (which contains all files from last Full backup, which means in this 
 case 3 weeks) and from 6 Incremental backups. The restore would be 
 quicker, but the space needed for backups is bit bigger (in some cases 
 can be even smaller). However, if you have all your backups on some 
 kind of disk array and it is reliable, bacula is able to restore data 
 from 30 Incremental backups with no problem. You still need to keep 
 all your volumes with Incremental backups in non-interupted sequence, 
 or you risk the lose of some files. Of course the Full backup must be 
 all right too.
 MArek


 Sebastian Perkins napsal(a):
 Thanks both for the info.

 I've been going through the Maximum Volume Jobs option with mutiple 
 volumes for incr  full as in the automated disk backup example.  
 Just that I trimmed everything down to one month (ie just 2 full 
 volumes  31 incr ones).

 With a test  situation with 10min retention times between full's ... 
 volumes are recycling rel nice !

 One question remains : do I need differential in my stratrgy ?

 Regards

 Sebastian Perkins
 Responsable Informatique
 [EMAIL PROTECTED]

 ALPES CONTROLES
 3 impasse des prairies
 74940 Annecy le Vieux
 Tél. 04.50.64.11.46 - Fax. 04.50.64.23.83
 http://www.alpes-controles.fr

 - Message Original -
 De: Marek Simon [EMAIL PROTECTED]
 Cc: bacula-users@lists.sourceforge.net
 Sent: lundi 19 novembre 2007 10 h 30 (GMT+0100) Europe/Berlin
 Sujet: Re: [Bacula-users] backup strategy

 David is right. Bacula is not able to clear a part of a volume. She can
 only append to a volume or clean it completely. I can recommend to have
 one volume for one day (Use duration 23 hours). Bacula will create new
 volumes everywhen she needs.
 Marek

 David Legg napsal(a):
  Hi Sebastian,
 
  I'm only a newbie to this but looking at your director config settings
  it looks like you haven't expressed a maximum volume size.  Therefore
  Bacula will keep filling up a volume until your disk runs out.  To 
 force
  Bacula to stop filling a volume and look around for another one you
  should use a 'Maximum Volume Bytes = nnn' or 'Maximum Volume Jobs = 
 nnn'
  or 'Volume Use Duration = ttt' setting in your pools definition.
 
  Re-read the beginning of Chapter 22 'Automatic Volume Recycling' for
  more details.
 
  Hope that helps.
 
  Regards,
 
  David Legg
 
 
  Sebastian Perkins wrote:
Hi,
 
  We are using bacula 2.2.6 with a disk based backup solution. 3 Months
  into testing, backups  restores are working.
 
  Here goes the problem : we need to backup 750Gb of data (22 server
  samba shares) and guarantee one (maybe two) months of restore data
  just in case somebody moves a directory, loses a file, wants lask
  week's file etc... The samba shares are basically full of doc/xls/pdf
  etc...
 
  The backup is a 2 To raid5 disk solution. I'm a bit worried about
  keeping control over volumes and disk space. I don't want the volumes
  to fill up any further than 1.8To while keeping one or 2 months of
  history.
 
  We've resolved one problem regarding size by using the compress 
 option :o)
 
  So, our bacula.conf strategy is :
 
  * one full backup per month, plus incremental every other day (no
differential).
  * one volume per share (so that we can keep a eye on a fast
growing volume).
  * keeping job/file/volume retention to 1 month (maybe 2).
 
  But the volumes just grow... and grow. I've 'volume updated' for
  retention periods but the auto-pruning doesn't seem to reduce volume
  size. For example after the 2nd month's full backup I thought the
  volume would be reduced in size by the 1st months full backup size
  (it's outside the retention period).
 
  Schedule, Volume  Client syntax are :
 
  Schedule {
Name = WeeklyCycle
Run

Re: [Bacula-users] Backup Strategy : One Full Backup + 364 Diferentials

2007-11-20 Thread Martin Simmons

 On Mon, 19 Nov 2007 18:55:30 -0300, Augusto Camarotti said:
 
 Nowadays we have a backup strategy like this :
 Daily Backup = Monday-Thursday = Full Backups with individual tapes for each
 weekday.
 Weekly Backup = 2nd-5th Friday =3D Full Backups with individual tapes for 7
 Fridays
 Montly Backup = 1st Friday = Full Backups with individual tapes for 12
 months
 
 This way I'm using 23 tapes.
 
 Then I was reading about Differential backups, and this strategy came into
 my mid :
 
 - One Full Backup on the first day of the year, say 2nd January (since 1st
 it's a holyday) on disk, in the storage machine. The data would be stored in
 a tape, which would be taken offsite.
 - 364 Differential backups, one for each day of the year. Using the 23 tapes
 to do it.
 
 Since the amount of data added on a year it's a maximum of 3 GB, the 23
 tapes will be enough to backup every day on a year.

Beware that the amount of new data added in a year is not sufficient to
calculate this.  You also need to include the amount of data *changed* since
the beginning of the year, because that will be backed up each day.


 Well, as I didn't find any recomendations to this strategy, I was in doubt
 if it would be fine to do this way.
 
 Suggestions or Critics are welcome.

Differential/incremental backups have a disadvantage that they do not track
deleted files, which could be a major problem over a whole year.

__Martin

-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup Strategy : One Full Backup + 364 Diferentials

2007-11-20 Thread Martin Simmons

 On Tue, 20 Nov 2007 13:14:24 -0300, Augusto Camarotti said:
 
 Hi Mark, thanks for the reply
 
 When I was saying the generally 3 GB of data was added, I was counting on
 data changed, since new data is being added when a change is made. I just
 don`t express it right.

Ah, OK.


 I didnt understand why it'll be a problem not tracking deleted files, since
 the reason of a backup is to save data that have been added, and not
 deleted.

It depends on the purpose of your backup, e.g. is it to allow for restores of
corrupted/lost files or is it for disaster recovery?

Some people don't like Differentials for disaster recovery.  Suppose the Full
backup contains file A and B.  Someone decides that file A is junk so they
delete it and they create file C instead.  The next Differential backup will
contain file C and nothing else.  This looks OK because all files are saved in
some backup.

However, when your disk gets trashed and you have to do a restore from the
Full and Differential backups.  The problem is that you will get back all
three files A, B and C.

__Martin


 
 Waiting for new sugestions or critics,
 
 Augusto
 
 On Nov 20, 2007 8:09 AM, Martin Simmons [EMAIL PROTECTED] wrote:
 
   On Mon, 19 Nov 2007 18:55:30 -0300, Augusto Camarotti said:
  
   Nowadays we have a backup strategy like this :
   Daily Backup = Monday-Thursday = Full Backups with individual tapes for
  each
   weekday.
   Weekly Backup = 2nd-5th Friday =3D Full Backups with individual tapes
  for 7
   Fridays
   Montly Backup = 1st Friday = Full Backups with individual tapes for 12
   months
  
   This way I'm using 23 tapes.
  
   Then I was reading about Differential backups, and this strategy came
  into
   my mid :
  
   - One Full Backup on the first day of the year, say 2nd January (since
  1st
   it's a holyday) on disk, in the storage machine. The data would be
  stored in
   a tape, which would be taken offsite.
   - 364 Differential backups, one for each day of the year. Using the 23
  tapes
   to do it.
  
   Since the amount of data added on a year it's a maximum of 3 GB, the 23
   tapes will be enough to backup every day on a year.
 
  Beware that the amount of new data added in a year is not sufficient to
  calculate this.  You also need to include the amount of data *changed*
  since
  the beginning of the year, because that will be backed up each day.
 
 
   Well, as I didn't find any recomendations to this strategy, I was in
  doubt
   if it would be fine to do this way.
  
   Suggestions or Critics are welcome.
 
  Differential/incremental backups have a disadvantage that they do not
  track
  deleted files, which could be a major problem over a whole year.
 
  __Martin
 

-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2007-11-19 Thread Sebastian Perkins

Thanks both for the info. 

I've been going through the Maximum Volume Jobs option with mutiple volumes 
for incr  full as in the automated disk backup example. Just that I trimmed 
everything down to one month (ie just 2 full volumes  31 incr ones). 

With a test situation with 10min retention times between full's ... volumes are 
recycling rel nice ! 

One question remains : do I need differential in my stratrgy ? 

Regards 

Sebastian Perkins 
Responsable Informatique 
[EMAIL PROTECTED] 

ALPES CONTROLES 
3 impasse des prairies 
74940 Annecy le Vieux 
Tél. 04.50.64.11.46 - Fax. 04.50.64.23.83 
http://www.alpes-controles.fr 


- Message Original - 
De: Marek Simon [EMAIL PROTECTED] 
Cc: bacula-users@lists.sourceforge.net 
Sent: lundi 19 novembre 2007 10 h 30 (GMT+0100) Europe/Berlin 
Sujet: Re: [Bacula-users] backup strategy 

David is right. Bacula is not able to clear a part of a volume. She can 
only append to a volume or clean it completely. I can recommend to have 
one volume for one day (Use duration 23 hours). Bacula will create new 
volumes everywhen she needs. 
Marek 

David Legg napsal(a): 
 Hi Sebastian, 
 
 I'm only a newbie to this but looking at your director config settings 
 it looks like you haven't expressed a maximum volume size. Therefore 
 Bacula will keep filling up a volume until your disk runs out. To force 
 Bacula to stop filling a volume and look around for another one you 
 should use a 'Maximum Volume Bytes = nnn' or 'Maximum Volume Jobs = nnn' 
 or 'Volume Use Duration = ttt' setting in your pools definition. 
 
 Re-read the beginning of Chapter 22 'Automatic Volume Recycling' for 
 more details. 
 
 Hope that helps. 
 
 Regards, 
 
 David Legg 
 
 
 Sebastian Perkins wrote: 
 
 Hi, 
 
 We are using bacula 2.2.6 with a disk based backup solution. 3 Months 
 into testing, backups  restores are working. 
 
 Here goes the problem : we need to backup 750Gb of data (22 server 
 samba shares) and guarantee one (maybe two) months of restore data 
 just in case somebody moves a directory, loses a file, wants lask 
 week's file etc... The samba shares are basically full of doc/xls/pdf 
 etc... 
 
 The backup is a 2 To raid5 disk solution. I'm a bit worried about 
 keeping control over volumes and disk space. I don't want the volumes 
 to fill up any further than 1.8To while keeping one or 2 months of 
 history. 
 
 We've resolved one problem regarding size by using the compress option :o) 
 
 So, our bacula.conf strategy is : 
 
 * one full backup per month, plus incremental every other day (no 
 differential). 
 * one volume per share (so that we can keep a eye on a fast 
 growing volume). 
 * keeping job/file/volume retention to 1 month (maybe 2). 
 
 But the volumes just grow... and grow. I've 'volume updated' for 
 retention periods but the auto-pruning doesn't seem to reduce volume 
 size. For example after the 2nd month's full backup I thought the 
 volume would be reduced in size by the 1st months full backup size 
 (it's outside the retention period). 
 
 Schedule, Volume  Client syntax are : 
 
 Schedule { 
 Name = WeeklyCycle 
 Run = Full 1st sun at 00:05 
 Run = Incremental mon-sat at 23:05 
 #Run = Differential 2nd-5th sun at 23:05 
 #Run = Incremental mon-sun at 23:05 
 } 
 
 Client { 
 Name = pool01-fd 
 Address = 192.168.x.y 
 FDPort = 9102 
 Catalog = MyCatalog 
 Password = xxx 
 File Retention = 1 month # 30 days 
 Job Retention = 1 months # 30 days 
 AutoPrune = yes # Prune expired Jobs/Files 
 } 
 
 Pool { 
 Name = Pool01 
 Pool Type = Backup 
 Recycle = yes # Bacula can automatically 
 recycle Volumes 
 AutoPrune = yes # Prune expired volumes 
 Volume Retention = 1 month # one month 
 LabelFormat = Vol01 
 } 
 
 
 So, just 2 questions : Is my strategy right ? Why are the volumes 
 growing ? 
 
 Thanks for any help ! 
 
 Sebastian Perkins 
 Responsable Informatique 
 [EMAIL PROTECTED] 
 
 ALPES CONTROLES 
 3 impasse des prairies 
 74940 Annecy le Vieux 
 Tél. 04.50.64.11.46 - Fax. 04.50.64.23.83 
 http://www.alpes-controles.fr 
 
  
 
 - 
 This SF.net email is sponsored by: Microsoft 
 Defy all challenges. Microsoft(R) Visual Studio 2005. 
 http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ 
  
 
 ___ 
 Bacula-users mailing list 
 Bacula-users@lists.sourceforge.net 
 https://lists.sourceforge.net/lists/listinfo/bacula-users 
 
 
 
 - 
 This SF.net email is sponsored by: Microsoft 
 Defy all challenges. Microsoft(R) Visual Studio 2005. 
 http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ 
 ___ 
 Bacula-users mailing list 
 Bacula-users@lists.sourceforge.net 
 https

Re: [Bacula-users] backup strategy

2007-11-18 Thread David Legg

Hi Sebastian,

I'm only a newbie to this but looking at your director config settings 
it looks like you haven't expressed a maximum volume size.  Therefore 
Bacula will keep filling up a volume until your disk runs out.  To force 
Bacula to stop filling a volume and look around for another one you 
should use a 'Maximum Volume Bytes = nnn' or 'Maximum Volume Jobs = nnn' 
or 'Volume Use Duration = ttt' setting in your pools definition.

Re-read the beginning of Chapter 22 'Automatic Volume Recycling' for 
more details.

Hope that helps.

Regards,

David Legg


Sebastian Perkins wrote:
 Hi,

 We are using bacula 2.2.6 with a disk based backup solution. 3 Months 
 into testing, backups  restores are working.

 Here goes the problem : we need to backup 750Gb of data (22 server 
 samba shares) and guarantee one (maybe two) months of restore data 
 just in case somebody moves a directory, loses a file, wants lask 
 week's file etc... The samba shares are basically full of doc/xls/pdf 
 etc...

 The backup is a 2 To raid5 disk solution. I'm a bit worried about 
 keeping control over volumes and disk space. I don't want the volumes 
 to fill up any further than 1.8To while keeping one or 2 months of 
 history.

 We've resolved one problem regarding size by using the compress option :o)

 So, our bacula.conf strategy is :

 * one full backup per month, plus incremental every other day (no
   differential).
 * one volume per share (so that we can keep a eye on a fast
   growing volume).
 * keeping job/file/volume retention to 1 month (maybe 2).

 But the volumes just grow... and grow. I've 'volume updated' for 
 retention periods but the auto-pruning doesn't seem to reduce volume 
 size. For example after the 2nd month's full backup I thought the 
 volume would be reduced in size by the 1st months full backup size 
 (it's outside the retention period).

 Schedule, Volume  Client syntax are :

 Schedule {
   Name = WeeklyCycle
   Run = Full 1st sun at 00:05
   Run = Incremental mon-sat at 23:05
   #Run = Differential 2nd-5th sun at 23:05
   #Run = Incremental mon-sun at 23:05
 }

 Client {
   Name = pool01-fd
   Address = 192.168.x.y
   FDPort = 9102
   Catalog = MyCatalog
   Password = xxx
   File Retention = 1 month# 30 days
   Job Retention = 1 months# 30 days
   AutoPrune = yes # Prune expired Jobs/Files
 }

 Pool {
   Name = Pool01
   Pool Type = Backup
   Recycle = yes   # Bacula can automatically 
 recycle Volumes
   AutoPrune = yes # Prune expired volumes
   Volume Retention = 1 month  # one month
   LabelFormat = Vol01
 }


 So, just 2 questions : Is my strategy right ? Why are the volumes 
 growing ?

 Thanks for any help !

 Sebastian Perkins
 Responsable Informatique
 [EMAIL PROTECTED]

 ALPES CONTROLES
 3 impasse des prairies
 74940 Annecy le Vieux
 Tél. 04.50.64.11.46 - Fax. 04.50.64.23.83
 http://www.alpes-controles.fr 

 

 -
 This SF.net email is sponsored by: Microsoft
 Defy all challenges. Microsoft(R) Visual Studio 2005.
 http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
 

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

-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy without recycling volumes

2006-10-19 Thread Ryan Novosielski

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

DVD's are 4.7GB no? Could get some more data on there, probably.

Lucas Di Pentima wrote:
 Hello all,
 
 I'm thinking about creating a backup strategy using DVD-R's and
 harddisk, maybe you can help me sorting out the strategy.
 
 Scenario:
   * Bacula 1.38.x
   * Debian Linux director
   * Debian Linux  Windows clients
   * Two schedules: ServerCycle (at night) and WorkstationCycle (start at
 9:30am)
 
 My idea is to apply a grandparent-parent-child strategy, but without
 recycling the volumes, that's because I expect to use this on DVD-Rs
 (when DVD support get stable), in the meanwhile this will be used on
 hard disk.
 
 Here is my proposed configuration:
 
 Schedule {
   Name = ServerCycle
   Run = Full 1st mon at 2:00
   Run = Differential 2nd-5th mon at 2:00
   Run = Incremental tue-fri at 2:00
 }
 Schedule {
   Name = WorkstationCycle
   Run = Full 1st mon at 9:30
   Run = Differential 2nd-5th mon at 9:30
   Run = Incremental tue-fri at 9:30
 }
 Pool {
   Name = Full-Pool
   Pool Type = Backup
   Recycle = no
   AutoPrune = yes
   Volume Retention = 60 days
   Maximum Volume Bytes = 4 g
   Volume Use Duration = 30 days
   Accept Any Volume = yes
   Label Format = Full-
 }
 Pool {
   Name = Diff-Pool
   Pool Type = Backup
   Recycle = no
   AutoPrune = yes
   Volume Retention = 60 days
   Maximum Volume Bytes = 4 g
   Volume Use Duration = 27 days
   Accept Any Volume = yes
   Label Format = Differential-
 }
 Pool {
   Name = Inc-Pool
   Pool Type = Backup
   Recycle = no
   AutoPrune = yes
   Volume Retention = 60 days
   Maximum Volume Bytes = 4 g
   Volume Use Duration = 5 days
   Accept Any Volume = yes
   Label Format = Incremental-
 }
 
 Catalog information will keep the last 2 months activity, older volumes
 will be stored forever, just in case they're needed.
 
 Every Full-Pool volume will have full jobs of one month, the same with
 Diff-Pool, and Inc-Pool volumes should only store files from within a week.
 
 This is all supposing that the information to be stored won't exceed 4
 GB per volume, of course.
 
 Is this strategy correct? are there any details I should take into account?
 
 Many thanks,
 
 
 
 
 -
 Using Tomcat but need to do more? Need to support web services, security?
 Get stuff done quickly with pre-integrated technology to make your job easier
 Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
 http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642
 
 
 
 
 ___
 Bacula-users mailing list
 Bacula-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/bacula-users

- --
  _  _ _  _ ___  _  _  _
 |Y#| |  | |\/| |  \ |\ |  | |Ryan Novosielski - Systems Programmer III
 |$| |__| |  | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922)
 \__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFN4P8mb+gadEcsb4RAobHAKCqmv8sErdHpNi8W6X15Xyt905YpwCfTd8T
UqMn0kz5PltQKJaHiObK5uQ=
=m959
-END PGP SIGNATURE-


-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy without recycling volumes

2006-10-19 Thread Lucas Di Pentima

Hi Ryan,

Ryan Novosielski wrote:
 DVD's are 4.7GB no? Could get some more data on there, probably.
DVD's are really 4.4 I think, but my concern is if my proposed
configuration is correct, or convenient.

Regards,
-- 
Lucas Di Pentima - Santa Fe, Argentina - Jabber ID: [EMAIL PROTECTED]
LUNIX S.R.L. - Soluciones en GNU/Linux - http://www.lunix.com.ar
GnuPG Public Key:
http://random.sks.keyserver.penguin.de:11371/pks/lookup?op=getsearch=0x6AA54FC9
Key fingerprint = BD3B 08C4 661A 8C3B 1855  740C 8F98 3FCF 6AA5 4FC9



signature.asc
Description: OpenPGP digital signature
-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Arno Lehmann


Hello,

On 1/19/2006 12:08 PM, Ralf Gross wrote:

Hi,

I want to backup 2 of our server to tape (AIT-2, 50GB native, 65-70GB
compressed, TLS-4210 12 slot autochanger). bacula 1.36.3 (debian sarge,
maybe I'll build bacula 1.38.4 from source...) is running and I'm already
able to backup to tape.


First, upgrade to 1.38.whatever-is-current.

Using volumes from more than one pool in one autochanger is not 
supported with 1.36, and you will probably run into problems sooner or 
later.



Both machines are also backed up to disk by an other backup tool, but I
want to have a backup on tape, just to be safe...

What I want:

Server 1:
 * ~55GB data
 * full backup each Sunday 6am
 * no inc/diff backup
 * 2 tapes/slots for 2 backups/weeks

Server 2:
 * ~190GB data (takes 10 hours and more...)
 * monthly backup at the first Saturday 8pm of the new month, this should
be archived for 2 years
 * regular full backup each Friday at 7pm, incr. backup Moday to Thursday
at 7pm

Because I'm only rarely at the place where the autochanger is located, I
can't change the tapes every week. My goal is to change only the archive
tapes once a month and - if possible - leave/reuse the other tapes.

I'm not quite sure how to setup bacula to achieve this. Should I really
use 2 differnt jobs and differnt pools for the regular/archive backup of
server 2?


One job per each combination of client and fileset. You set the levels 
in the schedule. See the manual for more information.



I'd create a pool with 2 tapes for server 1, and a pool with 3 tapes for
the archive job. Thus 7 tapes are left for the regular jobs. But the 2
regular full backups already need 6 tapes, thus only 1 tape is left for
the incr. job.


Determining the right strategy is, obviously, up to you. (You wouldn't 
be the first to discover that your autochanger doesn't fit your needs, 
though.)


Apart from that, I don't exactly understand why you make a distinction 
between archive and regular jobs. The simplest solution is to keep the 
full backups as long as you need them, and have the differential and 
incremental jobs go to pools that cycle more quickly.


In that case, you could leave the diff/incr volumes in the changer and 
only remove the full backup volumes.



I'm also very unsure about the right settings for
retention/prune/purge/recycle.


That's something you should look up in the manual, then - I couldn't 
explain it much differently.



Maybe I'm a just bit to confused at the moment ;)


Maybe... start with a simple experimental setup, with a small fileset, 
disk based volumes, and run your backups multiple times a day. This 
would allow you to observer the volume management live.


Arno


Ralf



---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users



--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Alan Brown


On Thu, 19 Jan 2006, Arno Lehmann wrote:

Using volumes from more than one pool in one autochanger is not supported 
with 1.36, and you will probably run into problems sooner or later.


That's not quite right, I have 4 pools loaded in my autochanger and have 
had for several years.


The problem comes when there are multiple simultaneous backups congigured 
and there is contention for different pools on the same drive in the 
autochanger - it causes the second job to abort instead of queueing.


I had no problems with different pools running simultaneously on different 
drives.


The workaround for pool contention is to disable multiple simultaneous 
backups, which unfortunately defeats the purpose of having multiple 
drives, but is better than having backups aborting.


AB


---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Kern Sibbald

On Thursday 19 January 2006 15:10, Alan Brown wrote:
 On Thu, 19 Jan 2006, Arno Lehmann wrote:
  Using volumes from more than one pool in one autochanger is not supported
  with 1.36, and you will probably run into problems sooner or later.

 That's not quite right, I have 4 pools loaded in my autochanger and have
 had for several years.

 The problem comes when there are multiple simultaneous backups congigured
 and there is contention for different pools on the same drive in the
 autochanger - it causes the second job to abort instead of queueing.

 I had no problems with different pools running simultaneously on different
 drives.

 The workaround for pool contention is to disable multiple simultaneous
 backups, which unfortunately defeats the purpose of having multiple
 drives, but is better than having backups aborting.

A better workaround is to upgrade to 1.38.4 or 1.38.5, where drive/pool 
contention and multiple drives are properly handled.

-- 
Best regards,

Kern

  (
  /\
  V_V


---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Alan Brown


On Thu, 19 Jan 2006, Kern Sibbald wrote:


The workaround for pool contention is to disable multiple simultaneous
backups, which unfortunately defeats the purpose of having multiple
drives, but is better than having backups aborting.


A better workaround is to upgrade to 1.38.4 or 1.38.5, where drive/pool
contention and multiple drives are properly handled.


I should have specificed 1.36 and earlier in the original posting, sorry.

I'm aware of the better path and plan to make the migration about this 
time next week when the current backup sets have completed running.


AB



---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Arno Lehmann


Hi,

On 1/19/2006 3:10 PM, Alan Brown wrote:

On Thu, 19 Jan 2006, Arno Lehmann wrote:

Using volumes from more than one pool in one autochanger is not 
supported with 1.36, and you will probably run into problems sooner or 
later.



That's not quite right, I have 4 pools loaded in my autochanger and have 
had for several years.


Hmm. I assume that you never had slots where tapes from different pools 
were loaded, because that's what will break your jobs. The reason is the 
way the DIR determins which volumes have their InChanger and Slot 
information reset.


The problem comes when there are multiple simultaneous backups 
congigured and there is contention for different pools on the same drive 
in the autochanger - it causes the second job to abort instead of queueing.


I had no problems with different pools running simultaneously on 
different drives.


The workaround for pool contention is to disable multiple simultaneous 
backups, which unfortunately defeats the purpose of having multiple 
drives, but is better than having backups aborting.


Quite true.

I guess you'll love 1.38.5 ;-)

Arno


AB



--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Arno Lehmann


Hello,

On 1/19/2006 3:11 PM, Ralf Gross wrote:
...

Determining the right strategy is, obviously, up to you. (You wouldn't
be the first to discover that your autochanger doesn't fit your needs,
though.)



At the moment, tapes are the only way to archive data and put them into a
safe here. 200GB+ disk are still more expensive and need more space than
tapes.


No, I wouldn't recommen to use disks as backup media *shudder*... 
rather, I suppose you might find that an autoloader with more slots 
might better suit your needs.





Apart from that, I don't exactly understand why you make a distinction
between archive and regular jobs. The simplest solution is to keep the
full backups as long as you need them, and have the differential and
incremental jobs go to pools that cycle more quickly.



I want to keep the monthly archive tapes for 2 years. But the regular full
backup should be kept only 14 days. I want to change the tapes only one
time a month. Thus the regular backup must be overwritten more frequently
and I think I have to use 2 different jobs/pools to achieve that.


Difficult.
You could do this differently, too, but that would require clever 
scripting (for the retention times etc.) and a rather complicated setup 
of your schedules.


Still I wonder why you need that distinction - I mean, you're doing the 
full backups anyway, and why not simply remove the volumes for one of 
the full backups once a month, set their status to Archived, and thus 
keep things simpler (and needing less autochanger slots)?





In that case, you could leave the diff/incr volumes in the changer and
only remove the full backup volumes.



Hm, you would use 2 separate pools for the full and inrc. volumes, take
out the monthly tape and put it in the safe? How can I be sure the monthly
tape is not overwritten by a regular job if I forget to take it out? (I'm
sure this is covered by the manual...).


True. Retention times is the keyword...


The focus should be - as mentioned
above - to change the archive tapes only once a month.


Should be possible... an actual estimate with real numbers and all that 
stuff would usually be something I get paid for :-) but I suppose you 
might find that you don't have enough slots:


Only talking about server 2. Three tapes per full backup. You keep them 
for two weeks and need space for the next backup: 9 tapes already.



I'm also very unsure about the right settings for
retention/prune/purge/recycle.


That's something you should look up in the manual, then - I couldn't
explain it much differently.



Ok.



Maybe I'm a just bit to confused at the moment ;)


Maybe... start with a simple experimental setup, with a small fileset,
disk based volumes, and run your backups multiple times a day. This
would allow you to observer the volume management live.



That's a good idea. Is it possible to simulate the changer behavior with
slots in any way (I guess not)?


That's not possible, but you will not need that, too. You can emulate 
your setup by manually creating the volumes you'd have in the 
autochanger. These volumes can belong to different pools, and you can 
limit the size each volume can have.


Whenever you get an operator intervention request you know that you'd 
have to take some tapes and go to the autochanger in real life.


Arno


Ralf



---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users



--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy/config questions

2006-01-19 Thread Ralf Gross

Arno Lehmann schrieb:
 On 1/19/2006 3:11 PM, Ralf Gross wrote:
 [snip]
 At the moment, tapes are the only way to archive data and put them into a
 safe here. 200GB+ disk are still more expensive and need more space than
 tapes.
 
 No, I wouldn't recommen to use disks as backup media *shudder*... 
 rather, I suppose you might find that an autoloader with more slots 
 might better suit your needs.

You are right, the changer is 6 six years old! Unfortunately the
budget situation is a bit tight at the moment. But that's an other
story.
 
 
 Apart from that, I don't exactly understand why you make a distinction
 between archive and regular jobs. The simplest solution is to keep the
 full backups as long as you need them, and have the differential and
 incremental jobs go to pools that cycle more quickly.
 
 
 I want to keep the monthly archive tapes for 2 years. But the regular full
 backup should be kept only 14 days. I want to change the tapes only one
 time a month. Thus the regular backup must be overwritten more frequently
 and I think I have to use 2 different jobs/pools to achieve that.
 
 Difficult.
 You could do this differently, too, but that would require clever 
 scripting (for the retention times etc.) and a rather complicated setup 
 of your schedules.

I'd like to keep this as simple as possible, because the setup is not
for a typical IT department. I'll be the only person that (hopefully)
will have detailed knowledge about bacula.

 
 Still I wonder why you need that distinction - I mean, you're doing the 
 full backups anyway, and why not simply remove the volumes for one of 
 the full backups once a month, set their status to Archived, and thus 
 keep things simpler (and needing less autochanger slots)?
 
If remove the tapes and put them into a safe - which might be placed
at a differnt location - and I need them for a restore I'll lose some
time. That was one reason for the 2 differnt jobs/pools.

I think I've to take a deeper look at the documentation to get the
whole picture, what I can accomplish with bacula. 


 In that case, you could leave the diff/incr volumes in the changer and
 only remove the full backup volumes.
 
 
 Hm, you would use 2 separate pools for the full and inrc. volumes, take
 out the monthly tape and put it in the safe? How can I be sure the monthly
 tape is not overwritten by a regular job if I forget to take it out? (I'm
 sure this is covered by the manual...).
 
 True. Retention times is the keyword...

I'm quite sure that I'll have some questions to the retention times in
the future (no threat! ;)
 
 The focus should be - as mentioned
 above - to change the archive tapes only once a month.
 
 Should be possible... an actual estimate with real numbers and all that 
 stuff would usually be something I get paid for :-) but I suppose you 
 might find that you don't have enough slots:

Agreed!
 
 Only talking about server 2. Three tapes per full backup. You keep them 
 for two weeks and need space for the next backup: 9 tapes already.

Ok, with 12 slots it seems to be the best method to have just 2 jobs
for the 2 server and handel the archive tapes as you mentioned above.
 
 [snip]
 Maybe I'm a just bit to confused at the moment ;)
 
 Maybe... start with a simple experimental setup, with a small fileset,
 disk based volumes, and run your backups multiple times a day. This
 would allow you to observer the volume management live.
 
 
 That's a good idea. Is it possible to simulate the changer behavior with
 slots in any way (I guess not)?
 
 That's not possible, but you will not need that, too. You can emulate 
 your setup by manually creating the volumes you'd have in the 
 autochanger. These volumes can belong to different pools, and you can 
 limit the size each volume can have.
 
 Whenever you get an operator intervention request you know that you'd 
 have to take some tapes and go to the autochanger in real life.

I'll try that after building 1.38.5 from source.

Thanks, Ralf



---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy not-so-strict schedule

2005-09-26 Thread Ryan Novosielski


Alan Brown wrote:




On Sun, 25 Sep 2005, Timo Neuvonen wrote:

- nominally, the next tape would be taken into use in the beginning 
of every

month

Reasons behind this are minimal manual handling of the tapes, 
knowing there
will be loss of up to one month's data if the current tape gets 
destroyed. I

believe my tape drive (Exabyte VXA-2) could handle this.




Does bacula have the syntax to say first tuesday of every month ?

Setting Max use duration helps too.

Why tuesdays? Most holidays/long weeekends use Mondays or Fridays. If 
this isn't true for your area, use wednesdays.


We switched to this policy soem time back on other backup systems for 
weekly/monthly backups and it made life a lot easier. Using an 
autochangwer for bacula I don't have to worry too much about exact 
timings but by policy the tape status (should be in safe/changer) is 
looked at on tuesdays too.


AB 


That's exactly the policy we currently use. Our former policy was on 
the first of the month, unless it's a weekend or holiday. I like this 
one better. ;)



---
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42 plasma tv or your very

own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy not-so-strict schedule

2005-09-25 Thread Kern Sibbald

On Sunday 25 September 2005 00:24, Timo Neuvonen wrote:
 After some occasional experiments with Bacula, I finally started reading
 the manual (my printed version is for 1.36.0) page by page, my goal being
 to set up a decently working backup strategy instead of occasional backups.

 So far, my idea is roughly the following:
 - pool of 12 tapes, each of which should have a capacity for a one full
 backup, plus appr. 20 incrementals or deltas
 - nominally, the next tape would be taken into use in the beginning of
 every month

 Reasons behind this are minimal manual handling of the tapes, knowing there
 will be loss of up to one month's data if the current tape gets destroyed.
 I believe my tape drive (Exabyte VXA-2) could handle this.


 What makes things complicated (afaik), is that I wouldn't like to obey too
 strictly the schedule new tape on the every 1st day of month. If I know
 in advance I'll be a few days out of the office, I propably would change
 the tape in advance -or late. Neither want I have a situation that the tape
 fills up, but it normally must be changed before this.  Manual (going to
 vacation) shows a way to handle this, marking the tape full or used
 manually, so the recycling allows re-using the oldest tape although the
 previous one wasn't written full yet.
 - what is the command required to manually mark tape as used?

 I would find it extremely handy, that every tape would always *start* with
 a full backup, followed by a set of incrementals and deltas. However, this
 conflicts with the strict schedule above. What would be the most simple way
 to synchronize the backup schedule with the tape change? I mean, always
 in the night right after changing the tape (which would *not* happen
 exactly to pre-set schedule) a full backup would be run, followed by
 regular
 day-of-week based series of incremental and delta backups in the following
 nights (which would go on indefinitely, in practice up to eg. 25-40
 calendar days or 20-30 nightly backups) until the tape would be changed
 again.

 Easy way would be to manually run full backup job, and accept the extra
 incremental to run the same night afterwards (set to run with lower
 priority). However, this doesn't sound nice to me. Any idea how to handle
 this without too many console commands? I think occasionally I still had to
 delegate this to somebody else, with little experience of maintenance
 tasks, that's why the easy way would be a must... If a better way does not
 exist, is there possibly any scripting language available in the console,
 so the novice operator could call a more complex pre-written script by a
 single console command?


 Finally, provided I get everything else working as planned, are there any
 special risks related to reconstructing the catalog with the bscan command
 after a disaster? Since this way all I would ever need for a disaster
 recovery would be just a single (the most recent) tape, I'm wondering if
 there were no actual need to keep separate catalog backups or bootstrap
 files?

You will regret not keeping separate catalog backups as well as bootstrap 
files for the backup AND a bootstrap for the catalog backup.  Please see the 
last section of the Restore chapter of the development manual:
http://www.bacula.org/dev-manual/Bacula_Consol_Restor_Comman.html#SECTION0002111000


 --
 TiN




 ---
 SF.Net email is sponsored by:
 Tame your development challenges with Apache's Geronimo App Server.
 Download it for free - -and be entered to win a 42 plasma tv or your very
 own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
 ___
 Bacula-users mailing list
 Bacula-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/bacula-users

-- 
Best regards,

Kern

  (
  /\
  V_V


---
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42 plasma tv or your very
own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Backup strategy not-so-strict schedule

2005-09-25 Thread Alan Brown


On Sun, 25 Sep 2005, Timo Neuvonen wrote:


- nominally, the next tape would be taken into use in the beginning of every
month

Reasons behind this are minimal manual handling of the tapes, knowing there
will be loss of up to one month's data if the current tape gets destroyed. I
believe my tape drive (Exabyte VXA-2) could handle this.


Does bacula have the syntax to say first tuesday of every month ?

Setting Max use duration helps too.

Why tuesdays? Most holidays/long weeekends use Mondays or Fridays. If this 
isn't true for your area, use wednesdays.


We switched to this policy soem time back on other backup systems for 
weekly/monthly backups and it made life a lot easier. Using an 
autochangwer for bacula I don't have to worry too much about exact timings 
but by policy the tape status (should be in safe/changer) is looked at 
on tuesdays too.


AB



---
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. 
Download it for free - -and be entered to win a 42 plasma tv or your very

own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Phil Stracchino

Aditya Ivaturi wrote:
 and so on. In this case Lun11 and Lun12 are the two partitions defined
 by individual storage configs. Thus this should create 5 volumes labled
 LIN-0001 and so on for linux boxes on Lun11 and recycle them right? Well
 the jobs ran fine the first day and created those volumes but the next
 day, the director sent an error message saying that it  Cannot find any
 appendable volumes. Please use the label  command to create a new
 Volume. Why is it requesting me to create new volumes? Shouldn't the
 director just use the existing volumes for each job until it is pruned
 and recycled? What am I missing here or what did I do wrong?

What is the output of a List Volumes command?


-- 
 Phil Stracchino   [EMAIL PROTECTED]
Renaissance Man, Unix generalist, Perl hacker
 Mobile: 603-216-7037 Landline: 603-886-3518


---
SF.Net email is Sponsored by the Better Software Conference  EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile  Plan-Driven Development * Managing Projects  Teams * Testing  QA
Security * Process Improvement  Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Aditya Ivaturi



 
 What is the output of a List Volumes command?
 
 

Little background - Default Pool was set
up initially when I was just getting everything set up. Then I moved things
around to have a more structure. Default is not in use any more but the
catalog is still the same. The jobs for windows boxes are still waiting
to be finished coz the linux jobs were first in queue.

Pool: Default
+-+-+---+--+--+--+-+--+---+---+-+
| MediaId | VolumeName | VolStatus | VolBytes
| VolFiles | VolRetention | Recycle | Slot | InChanger | MediaType | LastWritten
|
+-+-+---+--+--+--+-+--+---+---+-+
|1 | EDU   
 | Error   |0 |  
 0 |  31,536,000 |1 |  0
| 0 | File   | 2005-09-02
14:17:38 |
|2 | ITG-WIN   |
Error   |0 |   
0 |  31,536,000 |1 |  0 |
0 | File   | 2005-09-02 15:07:31
|
|3 | ITG-UNIX  |
Error   |0 |   
0 |  31,536,000 |1 |  0 |
0 | File   | 2005-09-03 01:05:05
|
|4 | EDU002   |
Error   |0 |   
0 |  31,536,000 |1 |  0 |
0 | File   | 2005-09-04 14:07:39
|
|5 | ITG-UNIX002 | Error 
 |0 |0 |
 31,536,000 |1 |  0 |  
  0 | File   | 2005-09-06 08:52:05 |
|6 | EDU003   |
Error   |0 |   
0 |  31,536,000 |1 |  0 |
0 | File   | 2005-09-06 08:53:25
|
|7 | ITG-UNIX003 | Error 
 |0 |0 |
 31,536,000 |1 |  0 |  
  0 | File   | 2005-09-06 09:43:30 |
+-+-+---+--+--+--+-+--+---+---+-+

Pool: Lin-Pool
+-++---+-+--+--+-+--+---+---+-+
| MediaId | VolumeName | VolStatus | VolBytes 
| VolFiles | VolRetention | Recycle | Slot | InChanger | MediaType
| LastWritten |
+-++---+-+--+--+-+--+---+---+-+
|8 | LIN-0001  | Used
  |  9,736,677 |0 |
 15,552,000 |1 |  0 |  
  0 | File   | 2005-09-06 16:51:32 |
|9 | LIN-0002  | Used
  | 29,961,179 |0 |
 15,552,000 |1 |  0 |  
  0 | File   | 2005-09-06 16:52:09 |
|   10 | LIN-0003  | Used
  | 164,744,328 |0 | 
15,552,000 |1 |  0 |   
 0 | File   | 2005-09-06 17:24:07 |
|   11 | LIN-0004  | Used
  |  3,946,566 |0 |
 15,552,000 |1 |  0 |  
  0 | File   | 2005-09-06 17:24:12 |
+-++---+-+--+--+-+--+---+---+-+

Pool: Win-Pool
No results to list.

--Turi

Re: [Bacula-users] backup strategy

2005-09-07 Thread Aditya Ivaturi



 
 Well, right away it's apparent that there's no volumes in your Win-Pool
 (therefore, no appendable volumes); all the volumes in your Lin-pool
are
 used (all on the same day), but not recycled -- presumably they have
not
 reached their retention date; and all the volumes in your Default
pool
 are errored.

Here is what happened when I set up the new configuration.
I ran the jobs in Lin-Pool and didn't bother with the Win-Pool as I was
expecting it to be finished that night according to schedule. Well bacula
started the whole cycle again and started with the Lin-Pool jobs first
and the very first job it was waiting for an appendable volume. That is
the reason why I don't have any volumes for windows yet. 

 
 Bacula is telling you No appendable volume could be found
because ...
 no appendable volumes exist.
 
 At a guess, you might want to replace that Use volume once
with a
 Volume use duration setting, because I'll bet what happened
is your
 system put one job on each of your Lin-pool Volumes the first day
you
 ran it, using up every tape in the pool, then had nothing left to
put
 the next job on.
 

From the docs for Volume
Use Duration:

The Volume Use Duration directive defines the
time period that the Volume can be written beginning from the time of first
data write to the Volume. If the time-period specified is zero (the default),
the Volume can be written indefinitely. Otherwise, when the time period
from the first write to the volume (the first Job written) exceeds the
time-period-specification, the Volume will be marked Used, which means
that no more Jobs can be appended to the Volume, but it may be recycled
if recycling is enabled.

So if I am understanding it correctly, my config was
default i.e 0 which means the volumes should have been used
indefinitely until the retention period is reached which is 6 months since
it is created (and it has been just 2 days since it was created) and will
be again recycled. But instead it has been marked used, which
tells me that somehow the director thinks that time period has exceeded
the specified time. And that is the reason director thinks that it cannot
write to the volume and is requesting a new volume to append to.

If my analysis is correct, what is going wrong? Or
could you please let me know where I comprehended it incorrectly? 

--Turi

Re: [Bacula-users] backup strategy

2005-09-07 Thread Arno Lehmann


Hello,

Aditya Ivaturi wrote:



I have 10 systems to be backed up evenly divided into linux and windows 
boxes (5  5). I want to maintain two pools one for each type and have 
separate volume for each job. So Pool-A will have vol-001, vol-002 and 
so on for the 5 linux machines and similarly for the windows machine 
pool. Each pool is going to reside on a different partition. So here is 
the config I went ahead with:


Pool {
  Name = Lin-pool
  Pool Type = Backup
  Recycle = yes   # automatically recycle Volumes
  AutoPrune = yes # Prune expired volumes
  Volume Retention = 6 months
  Accept Any Volume = yes # write on any volume in the pool
  Maximum Volume Jobs = 1
  Label Format = LIN-
  Maximum Volumes = 5
}


That would be the problem... you tell bacula to use each volume for one 
pool, not to use more than five volumes, and to keep the volumes for six 
months...


...
and so on. In this case Lun11 and Lun12 are the two partitions defined 
by individual storage configs. Thus this should create 5 volumes labled 
LIN-0001 and so on for linux boxes on Lun11 and recycle them right? Well 
the jobs ran fine the first day and created those volumes but the next 
day, the director sent an error message saying that it   Cannot find 
any appendable volumes. Please use the label  command to create a new 
Volume. Why is it requesting me to create new volumes? Shouldn't the 
director just use the existing volumes for each job until it is pruned 
and recycled? What am I missing here or what did I do wrong?


You told it something that can't work, I think.

Arno


--Turi



--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
SF.Net email is Sponsored by the Better Software Conference  EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile  Plan-Driven Development * Managing Projects  Teams * Testing  QA
Security * Process Improvement  Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Arno Lehmann


Hi,

theres no need to include [EMAIL PROTECTED] in 
the To-addresses. I admit that sometimes I forget to trim the addresses, 
but usually you will never have to deal with that address...


Aditya Ivaturi wrote:
...
So if I am understanding it correctly, my config was default i.e 0 
which means the volumes should have been used indefinitely until the 
retention period is reached which is 6 months since it is created (and 
it has been just 2 days since it was created) and will be again 
recycled. But instead it has been marked used, which tells me that 
somehow the director thinks that time period has exceeded the specified 
time. And that is the reason director thinks that it cannot write to the 
volume and is requesting a new volume to append to.


If my analysis is correct, what is going wrong? Or could you please let 
me know where I comprehended it incorrectly?



Yo missed something else, I think. See my previous post.

Arno


--Turi


--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
SF.Net email is Sponsored by the Better Software Conference  EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile  Plan-Driven Development * Managing Projects  Teams * Testing  QA
Security * Process Improvement  Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Aditya Ivaturi



  Pool {
   Name = Lin-pool
   Pool Type = Backup
   Recycle = yes  # automatically
recycle Volumes
   AutoPrune = yes # Prune expired
volumes
   Volume Retention = 6 months
   Accept Any Volume = yes # write on any volume in the pool
   Maximum Volume Jobs = 1
   Label Format = LIN-
   Maximum Volumes = 5
  }
 
 That would be the problem... you tell bacula to use each volume for
one 
 pool, not to use more than five volumes, and to keep the volumes for
six 
 months...
 

I was actually thinking the same thing what you pointed
out, but I followed the example that was given in Using Pools to
Manage Volume http://www.bacula.org/rel-manual/Using_Pools_Manage_Volumes.html.
If you see the example there, you'll notice that I did somewhat similar
to what it did and the Maximum volume jobs is set to 1. 

May be I am confused...

--Turi

Re: [Bacula-users] backup strategy

2005-09-07 Thread Phil Stracchino

Aditya Ivaturi wrote:
 The Volume Use Duration directive defines the time period that the
 Volume can be written beginning from the time of first data write to the
 Volume. If the time-period specified is zero (the default), the Volume
 can be written indefinitely. Otherwise, when the time period from the
 first write to the volume (the first Job written) exceeds the
 time-period-specification, the Volume will be marked Used, which means
 that no more Jobs can be appended to the Volume, but it may be recycled
 if recycling is enabled.
 
 So if I am understanding it correctly, my config was default i.e 0
 which means the volumes should have been used indefinitely until the
 retention period is reached which is 6 months since it is created (and
 it has been just 2 days since it was created) and will be again
 recycled. But instead it has been marked used, which tells me that
 somehow the director thinks that time period has exceeded the specified
 time. And that is the reason director thinks that it cannot write to the
 volume and is requesting a new volume to append to.


No, the volumes will not be used indefinitely.  Here's your Pool
definitions again:


Pool {
  Name = Lin-pool
  Pool Type = Backup
  Recycle = yes   # automatically recycle Volumes
  AutoPrune = yes # Prune expired volumes
  Volume Retention = 6 months
  Accept Any Volume = yes # write on any volume in the pool
  Maximum Volume Jobs = 1
  Label Format = LIN-
  Maximum Volumes = 5
}

Pool {
  Name = Win-pool
  Pool Type = Backup
  Recycle = yes   # automatically recycle Volumes
  AutoPrune = yes # Prune expired volumes
  Volume Retention = 6 months
  Accept Any Volume = yes # write on any volume in the pool
  Maximum Volume Jobs = 1
  Label Format = WIN-
  Maximum Volumes = 5
}


You have specified Maximum Volume Jobs = 1 in both pools.  That's
equivalent to specifying the Use Volume Once directive.  You're
telling Bacula to write *one single job* to each Volume, then mark it as
Used.  Bacula is doing exactly what you told it to do.  It's just that
what you told it to do wasn't actually what you WANTED it to do.


Try removing Maximum Volume Jobs = 1 from your Pool definitions, then
issuing an Update Pools command.



-- 
 Phil Stracchino   [EMAIL PROTECTED]
Renaissance Man, Unix generalist, Perl hacker
 Mobile: 603-216-7037 Landline: 603-886-3518


---
SF.Net email is Sponsored by the Better Software Conference  EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile  Plan-Driven Development * Managing Projects  Teams * Testing  QA
Security * Process Improvement  Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Arno Lehmann


Hi,

Aditya Ivaturi wrote:

...
I was actually thinking the same thing what you pointed out, but I 
followed the example that was given in Using Pools to Manage Volume 
http://www.bacula.org/rel-manual/Using_Pools_Manage_Volumes.html. If you 
see the example there, you'll notice that I did somewhat similar to what 
it did and the Maximum volume jobs is set to 1.


May be I am confused...


Might be :-)

The examples given seem to be right - you have to consider when jobs are 
run to go to the pools in the example.


There will, for example, be a Diff pool for differentials. These are run 
once a week, and they expire after 40 days - and there is only one job 
in he schedule.


If you have more than one job you need to modify the setup accordingly, 
although - especially when you expect your number of clients to grow - I 
wouldn't use Maximum Volume Jobs and Maximum Volumes. But that's my 
preference.


Arno


--Turi


--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
SF.Net email is Sponsored by the Better Software Conference  EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile  Plan-Driven Development * Managing Projects  Teams * Testing  QA
Security * Process Improvement  Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] backup strategy

2005-09-07 Thread Aditya Ivaturi



 
 Might be :-)
 
 The examples given seem to be right - you have to consider when jobs
are 
 run to go to the pools in the example.
 
 There will, for example, be a Diff pool for differentials. These are
run 
 once a week, and they expire after 40 days - and there is only one
job 
 in he schedule.
 
 If you have more than one job you need to modify the setup accordingly,

 although - especially when you expect your number of clients to grow
- I 
 wouldn't use Maximum Volume Jobs and Maximum Volumes. But that's my

 preference.
 

You are right I was confused. Thanks to you and Phil
Stracchino for clarifing it. Just for other users note - I had to change
the configs and then use update command to individually change each parameter
that I have changed for each volume and tape. This works flawlessly. 

--Turi

Re: [Bacula-users] backup strategy

2005-06-09 Thread Arno Lehmann


Hello,

viktorija wrote:

Hello, 


Now am trying to plan my backup strategy. I have about 1TB data and in server i 
have 5 harddrives (200Gb), now they are configured as RAID-stripe. But it is 
not good, because if one of hard will die then i will loose all my backups, so 
i want to configure these 5 drives as different types, but how can i say to 
bacula use next drive if previous is full? Please give me some links, or some 
point how to do it. Thank you and sorry for my bad english.


Sounds like it would be best if you set up your hard drives as RAID5 - 
when one disk dies you lose nothing. The need for additional space might 
be a problem, of course, but when you want to backup 1TB now and have 
1TB room you will probably be in the same situation soon anyway.


So, I'd suggest buying more disks. Or a tape drive :-) I like tape 
drives. I like to sell tape drives ;-)


Or you can try the following ideas.
Set up five storage devices, each using one disk. Set up your schedules 
to use disk 1 on monday, 2 on tuesday, etc.


Put the different clients or filesets to different disks.

Create a setup where each disk holds a volume in one storage device. I 
guess you could create five volumes (with a fixed maximum size), move 
the files to a disk each, and symlink them back to baculas storage 
directory.


Personally, I'd use the last approach - if it works reliably. This would 
give you the best flexibility in your setup (because you could also 
create 5 or 15 or ... volumes) and adding more volumes is possible.


Arno



---
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61 plasma display: http://www.necitguy.com/?r=20

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


--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61 plasma display: http://www.necitguy.com/?r=20

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

Re: [Bacula-users] backup strategy

2005-05-23 Thread Arno Lehmann


Hello,

Sza'sz Endre Istva'n wrote:


Dear List!

I try to to implement Bacula for a small company. I have succesfully
installed and configured bacula (congrats to all, who writed the
manual), but I need a little help, to develop the backup strategy.


I'll see if I can suggest something...


I plan to do the backup to 200 MB IDE disks, and I think, that I have
enough space, to do once a Full backup, and then the Incremental backups
for a month.


Ok, first of all, I'd consider using a third disk - even in case one 
breaks (or is just in use and overwriten) you will probably want to have 
two backups. At least I prefer that.



It will be simple, to keep this disk allways in the backup server, but I
think, that this is not a good idea because of security reason. So I
want to introduce a second backup disk, but I dont know when and how to
do the backup to the first backup disk, and when (and how) to the
second.

The requirements are:

- I want to change the backup disk once a week
- I want to be able to do a full restore using _just_ the one of the two
disks


And you want a monthly cycle... not the best situation to find a simple 
solution.



I think, that the following strategie will do, but I am courious, what
do you think about this: 


Week A: (disk1)
 - 1. day: - Full backup
 - 2-7 days: - Incremental (since week A 1. day full backup)

week B: (disk2)
 - 1. day: - Full backup
 - 2-7 days: - Incremental (since week B 1. day full backup)

week C: (disk1)
 - 1-7 days: - Incremental (since week A 1. day full backup and _NOT_
since week B 1. day full backup)

week D: (disk2)
 - 1-7 days: - Incremental (since week B 1. day full backup)

Is this overcomplicated? ;-) (I know, that this is :-) What I dont know
at this moment, how can I simply doing this.


Well, as long as it does what you want... but, sorry, I don't like that. 
Too complicated for me - after all, each disk change can result in an 
error, and setting this up doesn't look very straightforward. Deciding 
which jobs to restore is also not trivial.


The sort of advice I can give you depends on what you consider most 
important, so I'd suggest you think about your priorities:


- simplicity
- monthly cycle
- full restore from only one volume
- two volumes

If I had such a scenario, I could imagine the following backup scheme:

- I'd use three disks
- I would store one week of data on each disk
- First day: Full, afterwards incremental backups
- Set up only one pool, three volumes, retention time of two weeks.

This way, I'd have a three-week-cycle (which is close enough to a month 
for me ;-) , I can restore from one disk only, I should, even in case of 
a serious problem, have two backup generations left, and it's easy to 
handle - one day each week you change the disks. A-B-C-A Even a 
secretary can do that :-) (no offense... and apologies to all secretaries!)


Now you all know that for me, simplicity is the single most important 
feature in a customer's backup scheme...


Arno


So, please give me a piece of advice!

Thanks for the responses in advance!

Szaszka


---
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412alloc_id=16344op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


--
IT-Service Lehmann[EMAIL PROTECTED]
Arno Lehmann  http://www.its-lehmann.de


---
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412alloc_id=16344op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

44 matches

Mail list logo