Hello Mark,

Thanks for taking the time to write up the Feature Request, and I see you put 
a good deal of effort into it.  However, I'm not really too enthusiastic 
about this request (though I see certain merits) for three reasons:

1. Bacula already handles running out of spool space quite well. It simply 
despools, truncates the file (i.e. releases the space used), then attempts to 
spool again.  This allows it to work in a dynamic situation even with 
competing spool files on the same partition.

2. What you are requesting is very system dependent, and not something that I 
personally am interested in writing.  There is no simple easy way to get the 
space (ugly OS dependent calls if I remember right) -- even running a "df" 
command, which is relatively simple, is system dependent (it doesn't work at 
all on Win32 for example) and relatively hard to parse.

3. To do what you ask would require that Bacula constantly recalculate the 
amount of remaining space to compute the remaining percentage.  How often 
should it do so?  On a dynamic system as you indicate, it could change very 
quickly (big changes every second).

If you *really* want this included in the Projects list for next vote, OK will 
do, but please send an email to that effect, and I will include it (with any 
modifications you might want to add), otherwise, if I don't hear from you, I 
will assume that you have changed your mind.

Best regards,

Kern


On Wednesday 23 August 2006 20:24, [EMAIL PROTECTED] wrote:
> 
> Item 1:  flush spooled data based on free percentage 
>   Origin: Mark Bergman <[EMAIL PROTECTED]>
>   Date:  Wed Aug 23 11:44:28 EDT 2006
>   Status:
> 
>   What: add directive to flush spooled data on the basis of free space 
>       remaining on the spool volume
> 
>   Why:  Spooling data to disk can make backups more efficient. The larger 
the
>       spool file, the greater the potential gains. While there are directives
>       to limit the spool file sizes (individually and collectively), disk
>       space usage is often dynamic, particularly if the spool directory
>       isn't on a disk volume dedicate to spooling. By setting a threshold
>       of free space remaining on disk as a trigger to flush spool files to
>       the storage volume, spooling can be more efficient. Currently, the spool
>       partition can run out of space even if the "Maximum Spool Size" and
>       "Maximum Job Spool Size" limits are not reached.
> 
>       In addition, the current directives are based on per-storage
>       device limits. If there are multiple storage devices using the same
>       spool device, the only safe scenario is to assign a "Maximum Spool Size"
>       that is the available space divided by the number of devices. This would
>       ensure that simultaneous jobs destined for different devices won't run
>       out of spool space, but it's much less efficient if the jobs aren't 
>       going to all the storage devices.
> 
>       For example:
> 
>               available spool space: 100GB
>               number of storage devices: 4
>               recommended "Maximum Spool Size" in each Device definition: 25GB
> 
>       however, if jobs are using just one storage device, they will still
>       be limited to 25GB of spooled files, rather than being able to make use
>       of the 100GB of spool space.
> 
>       I suggest adding a directive "Minimum Free Percentage"
>       which has precendence over the other spool file size directives, and
>       which would allow jobs going to any device to continue spooling
>       (subject to the other limits, if set) until the specified amount of 
>       disk space remains. For example, if the directive in bacula-sd.conf
>       read:
> 
> 
>               Device {
>                       Minimum Free Percentage = 10
>                       ...
>                       }
> 
>       then in the above scenario, jobs destined for any of the 4 devices
>       would continue spooling as long as there was at least 10% of the spool
>       partition free (and subject to the other limits, if any were set).
> 
> 
>       In the event that the Minimum Free Percentage is reached, spooling would
>       be stopped, and the largest spool file would be flushed to the storage
>       device.
> 
>       It would be considered a syntax error if multiple Device{} definitions
>       that were using the same spool directory set the "Minimum Free
>       Percentage" directive to different values. 
> 
> 
> 
> 
> 
> 
> The information contained in this e-mail message is intended only for the 
personal and confidential use of the recipient(s) named above. If the reader 
of this message is not the intended recipient or an agent responsible for 
delivering it to the intended recipient, you are hereby notified that you 
have received this document in error and that any review, dissemination, 
distribution, or copying of this message is strictly prohibited. If you have 
received this communication in error, please notify us immediately by e-mail, 
and delete the original message.
> 
> -------------------------------------------------------------------------
> 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=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
> 

-------------------------------------------------------------------------
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=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to