bacula-users — Users forum for Bacula related matters. Support questions welcome.

Re: [Bacula-users] strange error on MyCatalog

[Josip D. <djo...@li...>]

On Thursday, April 30, 2026 4:01:26 PM CEST Mehrdad Ravanbod wrote:
> Hi
> 
> I recieved a rather strange error yesterday, on the inbuilt backup job
> BackupCatalog
> 
> it seems Bacula is saying it does not find the MyCatalog database, and
> something about a lib-file, but all the other backup jobs are
> completeing without errors so the database does not seem to have any
> problems. These lines showd up in the log
> 
> > Bacula2-dir JobId 2862: Error: Runscript: BeforeJob returned non-zero
> > status=1. ERR=Child exited with code 1
> > Bacula2-dir JobId 2862: BeforeJob: Can't find your catalog (MyCatalog)
> > in director configuration
> > Bacula2-dir JobId 2862: BeforeJob: /opt/bacula/bin/dbcheck:
> > /usr/lib64/libpq.so.5: no version information available (required by
> > /opt/bacula/lib/libbaccats-15.0.2.so)
> > Bacula2-dir JobId 2862: shell command: run BeforeJob
> > "/opt/bacula/scripts/make_catalog_backup.pl MyCatalog"
> 
> Anyone has seen this beforeor knows what it means??

>From this error, I would conclude that you have a job that executes /opt/
bacula/bin/dbcheck before running a job, using BeforeJob option.

dbcheck tool is used to clean up the records from the database that are
not needed any more. It needs to read database credentials from the
bacula-dir configuration file and if the path to the configuration file
is not specified, it will use some compiled-in default path.

In short, dbcheck is not using the correct database connection info.


Regards

-- 
Josip Deanovic

[Bacula-users] strange error on MyCatalog

[Mehrdad R. <meh...@am...>]

Hi

I recieved a rather strange error yesterday, on the inbuilt backup job 
BackupCatalog

it seems Bacula is saying it does not find the MyCatalog database, and 
something about a lib-file, but all the other backup jobs are 
completeing without errors so the database does not seem to have any 
problems. These lines showd up in the log

> Bacula2-dir JobId 2862: Error: Runscript: BeforeJob returned non-zero 
> status=1. ERR=Child exited with code 1
> Bacula2-dir JobId 2862: BeforeJob: Can't find your catalog (MyCatalog) 
> in director configuration
> Bacula2-dir JobId 2862: BeforeJob: /opt/bacula/bin/dbcheck: 
> /usr/lib64/libpq.so.5: no version information available (required by 
> /opt/bacula/lib/libbaccats-15.0.2.so)
> Bacula2-dir JobId 2862: shell command: run BeforeJob 
> "/opt/bacula/scripts/make_catalog_backup.pl MyCatalog"

Anyone has seen this beforeor knows what it means??

-- 
________________________________________________________

Ampfield Aktiebolag
Mehrdad Ravanbod
System administrator

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Arno L. <al...@it...>]

Hi Heitor,

you did offer your software for test and evaluation, and I wrote

On 28.04.26 at 22:35, Arno Lehmann via Bacula-users wrote:
> But as I said -- I'd really like to give your product a try. There are 
> quite some aspects that the public documenttaion does not cover -- for 
> example, what will happen if the Sentinel Plugin gets incomplete file 
> change history, or gets swamped with change notifications? How will it 
> work with uncommon file systems, and what happens if backups are too 
> rare so the list of changed files overflows?

Any chance I can give it a try?

Cheers,

Arno

-- 
Arno Lehmann

IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück

Re: [Bacula-users] major lost of performance after moving Bacula Server from Solaris to linux for Solaris clients only

[Josh F. <jf...@ja...>]

On 4/29/26 11:05, Andrew Watkins wrote:
>
> Arno,
>
> Well I finally spotted the difference between the new and old systems.
>
> I said the old server was doing compression, but IT WAS NOT! So, as 
> soon as I switch off compression on the new server I get better 
> performance.
>
>    Rate:                   84648.1 KB/s
>    Software Compression:   None
>    Comm Line Compression:  13.4% 1.2:1
>
> I didn't spot it since the configuration files are the same and I used 
> the same build setup on both platforms. For some reason the old 
> Solaris server didn't set the compression on any client (clients have 
> not been altered, except for passwords!). I don't see why since libz 
> is available.
>
> So, I have good speeds now with NO compression, but I will look into 
> later why GZIP (haven't tested LZO yet) really hits performance 
> (Solaris client CPU is 95% idle during backup, so not a CPU problem.)
>

Gzip in Bacula is single-threaded and operates on quite small block 
sizes. Look for a single core being slammed, even though the overall CPU 
utilization for all cores might be low. If compression is needed to 
decrease storage needs, then put the backup volumes on ZFS (or for tape 
just use the tape drive's hardware compression) and leave Bacula 
compression off.


> Anyway, for now thanks,
>
> Andrew
>
> On 4/23/2026 3:25 PM, Arno Lehmann via Bacula-users wrote:
>> Hi Andrew,
>>
>> just a very quick suggestion for now -- you could try disabling 
>> transport encryption and compression and see if you get better 
>> throughput.
>>
>> And one question about the change: Have you touched the client on the 
>> Solaris host or the configuration (beyond changing credentials for 
>> the DIR)?
>>
>> The reason I ask is that, as far as I recall, the openSSL and 
>> possibly compression libraries on Solaris may be just slow compared 
>> to other platforms.
>>
>> 'status client=<the-solaris-host-fd>' should give a bit of 
>> information about the client build in the header, and that may become 
>> interesting when we try to find out if the FD is potentially 
>> problematic.
>>
>> Cheers,
>>
>> Arno
>>
>> Am 23.04.2026 um 13:28 schrieb Andrew Watkins:
>>> Hello,
>>>
>>> Have a strange one here. I have moved to a new Bacula Server and the 
>>> read performance for a Solaris client is 3 times as bad.
>>>
>>> All other clients (Windows, Linux) are good.
>>>
>>> Old setup:
>>>
>>> Solaris Bacula 13.04 Server (slow disks for spooling) LTO5 when 
>>> backing up a Solaris file server I get:
>>>
>>>      iostat gives me about 100Ms
>>>      Bacula Backup report gives me ~ 54,664.4 KB/s
>>>
>>> New Setup:
>>>
>>> Ubuntu Bacula 13.04 Server (Fast disks for spooling) LTO9 when 
>>> backup up save Solaris file server I get:
>>>
>>> iostat gives me about 30Ms
>>> Bacula Backup report gives me ~4,661.3 KB/s
>>>
>>> But other clients Windows & linux are report good rates.
>>>
>>> The conf files are near enough the same, so any ideas why Solaris is 
>>> so slow when backup up to a Linux server?
>>>
>>> Thanks,
>>>
>>> Andrew
>>>
>>> -- 
>>> *****************************************************
>>> *****    Support Request to...@dc... *****
>>> *****************************************************
>>> * Andrew Watkins                                    *
>>> * Birkbeck, University of London                    *
>>> * Computing and Mathematical Sciences               *
>>> *https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fnotallmicrosoft.blogspot.com&c=E,1,ciyP5u8kQDn-3n1M4lyEC8NXTbHxpJJPqFHVwaSuB0nBvL0meeOunhGJSyiB7XGA7SgCcKpcNE0ZJCYdLWDtlKWoDcaPGj1PgnYzPba9re8NrNfSD6i7ldU,&typo=1 
>>> *
>>> *****************************************************
>>>
>>>
>>>
>>> _______________________________________________
>>> Bacula-users mailing list
>>> Bac...@li...
>>> https://lists.sourceforge.net/lists/listinfo/bacula-users
>>
> -- 
> *****************************************************
> *****    Support Request to...@dc...    *****
> *****************************************************
> * Andrew Watkins                                    *
> * Birkbeck, University of London                    *
> * Computing and Mathematical Sciences               *
> *http://notallmicrosoft.blogspot.com               *
> *****************************************************
>
>
> _______________________________________________
> Bacula-users mailing list
> Bac...@li...
> https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] major lost of performance after moving Bacula Server from Solaris to linux for Solaris clients only

[Rob G. <ro...@cr...>]

Gzip typically has high compression at high CPU cost.
Lzo typically has lower compression, and lower CPU cost.
Zstd (if available for your version of Bacula) typically has the best of
both worlds.

Robert Gerber
402-237-8692
ro...@cr...

On Wed, Apr 29, 2026, 10:39 AM Andrew Watkins <a.w...@bb...> wrote:

> Arno,
>
> Well I finally spotted the difference between the new and old systems.
>
> I said the old server was doing compression, but IT WAS NOT! So, as soon
> as I switch off compression on the new server I get better performance.
>
>   Rate:                   84648.1 KB/s
>   Software Compression:   None
>   Comm Line Compression:  13.4% 1.2:1
>
> I didn't spot it since the configuration files are the same and I used the
> same build setup on both platforms. For some reason the old Solaris server
> didn't set the compression on any client (clients have not been altered,
> except for passwords!). I don't see why since libz is available.
>
> So, I have good speeds now with NO compression, but I will look into later
> why GZIP (haven't tested LZO yet) really hits performance (Solaris client
> CPU is 95% idle during backup, so not a CPU problem.)
>
> Anyway, for now thanks,
>
> Andrew
> On 4/23/2026 3:25 PM, Arno Lehmann via Bacula-users wrote:
>
> Hi Andrew,
>
> just a very quick suggestion for now -- you could try disabling transport
> encryption and compression and see if you get better throughput.
>
> And one question about the change: Have you touched the client on the
> Solaris host or the configuration (beyond changing credentials for the
> DIR)?
>
> The reason I ask is that, as far as I recall, the openSSL and possibly
> compression libraries on Solaris may be just slow compared to other
> platforms.
>
> 'status client=<the-solaris-host-fd>' should give a bit of information
> about the client build in the header, and that may become interesting when
> we try to find out if the FD is potentially problematic.
>
> Cheers,
>
> Arno
>
> Am 23.04.2026 um 13:28 schrieb Andrew Watkins:
>
> Hello,
>
> Have a strange one here. I have moved to a new Bacula Server and the read
> performance for a Solaris client is 3 times as bad.
>
> All other clients (Windows, Linux) are good.
>
> Old setup:
>
> Solaris Bacula 13.04 Server (slow disks for spooling) LTO5 when backing up
> a Solaris file server I get:
>
>      iostat gives me about 100Ms
>      Bacula Backup report gives me ~ 54,664.4 KB/s
>
> New Setup:
>
> Ubuntu Bacula 13.04 Server (Fast disks for spooling) LTO9 when backup up
> save Solaris file server I get:
>
> iostat gives me about 30Ms
> Bacula Backup report gives me ~4,661.3 KB/s
>
> But other clients Windows & linux are report good rates.
>
> The conf files are near enough the same, so any ideas why Solaris is so
> slow when backup up to a Linux server?
>
> Thanks,
>
> Andrew
>
> --
> *****************************************************
> *****    Support Request to...@dc...    *****
> *****************************************************
> * Andrew Watkins                                    *
> * Birkbeck, University of London                    *
> * Computing and Mathematical Sciences               *
> *
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fnotallmicrosoft.blogspot.com&c=E,1,ciyP5u8kQDn-3n1M4lyEC8NXTbHxpJJPqFHVwaSuB0nBvL0meeOunhGJSyiB7XGA7SgCcKpcNE0ZJCYdLWDtlKWoDcaPGj1PgnYzPba9re8NrNfSD6i7ldU,&typo=1
> *
> *****************************************************
>
>
>
> _______________________________________________
> Bacula-users mailing list
> Bac...@li...
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
>
> --
> *****************************************************
> *****    Support Request to sg...@dc...    *****
> *****************************************************
> * Andrew Watkins                                    *
> * Birkbeck, University of London                    *
> * Computing and Mathematical Sciences               *
> * http://notallmicrosoft.blogspot.com               *
> *****************************************************
>
> _______________________________________________
> Bacula-users mailing list
> Bac...@li...
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>

Re: [Bacula-users] major lost of performance after moving Bacula Server from Solaris to linux for Solaris clients only

[Andrew W. <a.w...@bb...>]

Arno,

Well I finally spotted the difference between the new and old systems.

I said the old server was doing compression, but IT WAS NOT! So, as soon 
as I switch off compression on the new server I get better performance.

   Rate:                   84648.1 KB/s
   Software Compression:   None
   Comm Line Compression:  13.4% 1.2:1

I didn't spot it since the configuration files are the same and I used 
the same build setup on both platforms. For some reason the old Solaris 
server didn't set the compression on any client (clients have not been 
altered, except for passwords!). I don't see why since libz is available.

So, I have good speeds now with NO compression, but I will look into 
later why GZIP (haven't tested LZO yet) really hits performance (Solaris 
client CPU is 95% idle during backup, so not a CPU problem.)

Anyway, for now thanks,

Andrew

On 4/23/2026 3:25 PM, Arno Lehmann via Bacula-users wrote:
> Hi Andrew,
>
> just a very quick suggestion for now -- you could try disabling 
> transport encryption and compression and see if you get better 
> throughput.
>
> And one question about the change: Have you touched the client on the 
> Solaris host or the configuration (beyond changing credentials for the 
> DIR)?
>
> The reason I ask is that, as far as I recall, the openSSL and possibly 
> compression libraries on Solaris may be just slow compared to other 
> platforms.
>
> 'status client=<the-solaris-host-fd>' should give a bit of information 
> about the client build in the header, and that may become interesting 
> when we try to find out if the FD is potentially problematic.
>
> Cheers,
>
> Arno
>
> Am 23.04.2026 um 13:28 schrieb Andrew Watkins:
>> Hello,
>>
>> Have a strange one here. I have moved to a new Bacula Server and the 
>> read performance for a Solaris client is 3 times as bad.
>>
>> All other clients (Windows, Linux) are good.
>>
>> Old setup:
>>
>> Solaris Bacula 13.04 Server (slow disks for spooling) LTO5 when 
>> backing up a Solaris file server I get:
>>
>>      iostat gives me about 100Ms
>>      Bacula Backup report gives me ~ 54,664.4 KB/s
>>
>> New Setup:
>>
>> Ubuntu Bacula 13.04 Server (Fast disks for spooling) LTO9 when backup 
>> up save Solaris file server I get:
>>
>> iostat gives me about 30Ms
>> Bacula Backup report gives me ~4,661.3 KB/s
>>
>> But other clients Windows & linux are report good rates.
>>
>> The conf files are near enough the same, so any ideas why Solaris is 
>> so slow when backup up to a Linux server?
>>
>> Thanks,
>>
>> Andrew
>>
>> -- 
>> *****************************************************
>> *****    Support Request to...@dc...    *****
>> *****************************************************
>> * Andrew Watkins                                    *
>> * Birkbeck, University of London                    *
>> * Computing and Mathematical Sciences               *
>> *https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fnotallmicrosoft.blogspot.com&c=E,1,ciyP5u8kQDn-3n1M4lyEC8NXTbHxpJJPqFHVwaSuB0nBvL0meeOunhGJSyiB7XGA7SgCcKpcNE0ZJCYdLWDtlKWoDcaPGj1PgnYzPba9re8NrNfSD6i7ldU,&typo=1 
>> *
>> *****************************************************
>>
>>
>>
>> _______________________________________________
>> Bacula-users mailing list
>> Bac...@li...
>> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
-- 
*****************************************************
*****    Support Request to...@dc...    *****
*****************************************************
* Andrew Watkins                                    *
* Birkbeck, University of London                    *
* Computing and Mathematical Sciences               *
*http://notallmicrosoft.blogspot.com               *
*****************************************************

Re: [Bacula-users] Bacula within Proxmox VMs - performance discussion

[Rob G. <ro...@cr...>]

Stefan,

Watch for duplicate COW setups. If client selected QCOW VM storage on top
of ZFS, that would be a copy on write filesystem on top of another copy on
write filesystem.

Robert Gerber
402-237-8692
ro...@cr...

On Wed, Apr 29, 2026, 8:22 AM Stefan G. Weichinger via Bacula-users <
bac...@li...> wrote:

> Am 29.04.26 um 14:39 schrieb Heitor Faria:
> > Hello Stefan,
> >
> >     Things work, but I only see around 50-70 MB/s for backups.
> >
> > You obviously have a bottleneck. You need to start measuring: network
> > (iperf); disk writes, reads, your tree walk (e.g. time find /your/path -
> > type f > /dev/null) etc.
>
> I replaced the virtual disk controller in the client from virtio to scsi.
>
> And enabled SSD emulation for that disk.
>
> Changed little ... the disk is read with around 70 MB/s according to
> atop -> shown as "busy 80%" (with virtio it was even higher).
>
> I also checked the mount options for things like "noatime" (ext4).
>
> SSD Emulation might be wrong:
>
> I didn't configure that server, so I just found that the larger
> datastore consists of HDDs: 4x Seagate Exos 28 TB in some kind of ZFS
> pool (!) with raidz2, as far as I see.
>
> hm, this might be the reason for some unwanted overhead here.
>
> Let me repeat: not my idea, not my setup ;-)
>
> > By coincidence, I just published my personal Proxmox Backup Plugin for
> > Bacula Community, which supports the latest Proxmox Backup APIs and
> > provides all PBS functions too: https://www.linkedin.com/posts/
> > heitorbacula_podheitor-plugin-de-backup-replica%C3%A7%C3%A3o-
> > activity-7455206154382782464-6GS_?
> >
> utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAAACcSp4BYjaTUSVxav5iossfBRxcPrLdSJM
> <
> https://www.linkedin.com/posts/heitorbacula_podheitor-plugin-de-backup-replica%C3%A7%C3%A3o-activity-7455206154382782464-6GS_?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAAACcSp4BYjaTUSVxav5iossfBRxcPrLdSJM
> >
> > PVT if you want to test it.
>
> thanks, not right now.
>
>
> _______________________________________________
> Bacula-users mailing list
> Bac...@li...
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>

Re: [Bacula-users] Bacula within Proxmox VMs - performance discussion

[Stefan G. W. <li...@xu...>]

Am 29.04.26 um 14:39 schrieb Heitor Faria:
> Hello Stefan,
> 
>     Things work, but I only see around 50-70 MB/s for backups.
> 
> You obviously have a bottleneck. You need to start measuring: network 
> (iperf); disk writes, reads, your tree walk (e.g. time find /your/path - 
> type f > /dev/null) etc.

I replaced the virtual disk controller in the client from virtio to scsi.

And enabled SSD emulation for that disk.

Changed little ... the disk is read with around 70 MB/s according to 
atop -> shown as "busy 80%" (with virtio it was even higher).

I also checked the mount options for things like "noatime" (ext4).

SSD Emulation might be wrong:

I didn't configure that server, so I just found that the larger 
datastore consists of HDDs: 4x Seagate Exos 28 TB in some kind of ZFS 
pool (!) with raidz2, as far as I see.

hm, this might be the reason for some unwanted overhead here.

Let me repeat: not my idea, not my setup ;-)

> By coincidence, I just published my personal Proxmox Backup Plugin for 
> Bacula Community, which supports the latest Proxmox Backup APIs and 
> provides all PBS functions too: https://www.linkedin.com/posts/ 
> heitorbacula_podheitor-plugin-de-backup-replica%C3%A7%C3%A3o- 
> activity-7455206154382782464-6GS_? 
> utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAAACcSp4BYjaTUSVxav5iossfBRxcPrLdSJM <https://www.linkedin.com/posts/heitorbacula_podheitor-plugin-de-backup-replica%C3%A7%C3%A3o-activity-7455206154382782464-6GS_?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAAACcSp4BYjaTUSVxav5iossfBRxcPrLdSJM>
> PVT if you want to test it.

thanks, not right now.

Re: [Bacula-users] Bacula within Proxmox VMs - performance discussion

[Heitor F. <hei...@gm...>]

Hello Stefan,

> Things work, but I only see around 50-70 MB/s for backups.
>
You obviously have a bottleneck. You need to start measuring: network
(iperf); disk writes, reads, your tree walk (e.g. time find /your/path
-type f > /dev/null) etc.
I really like to use the BBR TCP congestion control algorithm. It might
make some difference between client and storage comms:
https://baculaenterprise.com.br/bacula-e-prototoclo-bbr-para-backups-via-internet-perdas-de-pacote-desconecoes-etc/?lang=en

By coincidence, I just published my personal Proxmox Backup Plugin for
Bacula Community, which supports the latest Proxmox Backup APIs and
provides all PBS functions too:
https://www.linkedin.com/posts/heitorbacula_podheitor-plugin-de-backup-replica%C3%A7%C3%A3o-activity-7455206154382782464-6GS_?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAAACcSp4BYjaTUSVxav5iossfBRxcPrLdSJM
PVT if you want to test it.

Rgds.
-- 
Atenciosamente,

Heitor faria (Miami)
https://www.youtube.com/@podheitor
WhatsApp: +1 786-726-1749 | +55 61 98268-4220

Re: [Bacula-users] Bacula within Proxmox VMs - performance discussion

[Stefan G. W. <li...@xu...>]

Let me add:

the network test for that client shows 270 - 295 MB/s

I don't compress things

[Bacula-users] Bacula within Proxmox VMs - performance discussion

[Stefan G. W. <li...@xu...>]

Let me describe a setup I prepare for a customer.

So far I see less performance than expected or wanted, so I'd like to 
check for things I miss.

A customer runs one single Proxmox Virtualization host (PVE).
YES, I know that I could also ask things in a Proxmox Forum ...

That host has plenty of CPU and RAM and a large (for me) array of SSDs.
I don't know the actual RAID level right now, but it should be "fast 
enough".

So far there are maybe 10 VMs, not much.

One is a PBS (Proxmox Backup server), it should write VM backups to an 
LTO7 library attached.

As I noticed that I can only attach the SAS-card to a single VM I 
decided and managed to install Bacula into this PBS-VM as well. Works.

(no, no conflicting jobs right now. pbs does nothing with tape so far, 
disabled that while setting up)

A second VM will be a fileserver. Debian 13.4, bacula-fd inside.
Data from an external legacy server was rsynced in, now I have set up 
the fd, plus some filesets and jobs.

Things work, but I only see around 50-70 MB/s for backups.

I test with and without Data Spooling, doesn't make much difference.

Both VMs are set up with virtual NICs on the same virtual bridge ... so 
in theory they should have 10 Gbit/s network (virtio nics in place, yes).

The ressources shared are storage (one VM reading from the array, the 
other writing it to the same storage, just another virtual disk = spool 
area), CPU and RAM, right?

Nothing looks topped out or so ..

When I disable Data Spooling for the Job, the storage conflict should be 
gone, right?

I know that this setup isn't exactly prepared for top performance, but 
anyway I'd expect more throughput possible, right?

One job backs up 3.2 TB alone .. this takes days with the current speed.

Let me know your thoughts and what parts of the config I should look at, 
or maybe share here.

Thanks!

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Josip D. <djo...@li...>]

On 2026-04-29 08:51, Radosław Korzeniewski wrote:
> The idea of an "incremental accelerator" for common local filesystems 
> is
> great.
> But the inotify api is not the best solution for this case IMVHO, YMMV. 
> The
> main issue is the scalability issue.

True. IMHO, this is quick and dirty solution path which can be 
acceptable
for some use cases. I consider reliability of a backup system too 
important
and failures to miss a file unforgiving.

> You want to use "incremental accelerator" for all your high volume 
> storage,
> but inotify functionality does not scale.
> The more directories you want to inotify the more resources it 
> requires. A
> tradeoff, sure.

Yes, it doesn't scale. Still, it can monitor millions of files.
For several millions of files, it will probably consume several hundreds
of megabytes (at least), and this will be unswappable memory.

 From what I remember (I read this is still the case), there is much
worse problem with inotify - it doesn't throttle. When queue overflow
occurs, events get dropped and for most use cases, this is a disaster.
No retry, no recovery from this.
Kernel does send notification but only about overflow happening, without
detailed info about each lost event.
This means that events could slip trough your fingers unnoticed.

Even inotify(7) man page warns about inconsistencies due to race 
conditions.

There are other potential issues as well. Man page says:
-----BEGIN-----
  If a filesystem is mounted on top of a monitored directory, no event is
  generated, and no events are generated for  objects  immediately  under
  the  new  mount  point.   If  the filesystem is subsequently unmounted,
  events will subsequently be generated for the directory and the objects
  it contains.
-----END-----


> I did a prototype of this feature years ago (no coding LLM existed yet) 
> and
> it wasn't worth the effort (the architecture was different, when fd 
> starts
> it spawns a special thread and registers required watches from the 
> config
> file, so incremental job can ask this thread what changed - it was a 
> never
> ending story of issues, race conditions, edge cases, etc.).
> 
> Today there is a modern and scalable fanotify API (or FSEvents in macOS
> or USN Change Journal in win) which should be used instead, again 
> IMVHO,
> YMMV.

Fanotify might be more scalable and reliable but it can still lose 
events.
No guaranteed completeness - the stream of events you receive is not
guaranteed to represent all filesystem activity that actually happened.

Its man page fanotify(7) states: "The event queue can overflow. In this 
case,
events are lost.".

And for what I understood, just like with inotify, you get a single 
overflow
notification, so you know that a loss occurred but you don't know what.

In short, there are many cases where fanotify is better than inotify but
it comes with some other limitations and design choices that should be
considered when using it.


Regards

-- 
Josip Deanovic

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Radosław K. <rad...@ko...>]

Hi Arno,

wt., 28 kwi 2026 o 22:36 Arno Lehmann via Bacula-users <
bac...@li...> napisał(a):

> > Commvault uses a similar technique. So that's how I know it is
> > beneficial and feasible. =)
>
> Well, Commvault has years of engineering and a rather large team of
> developers working on those things, which I like to understand as
> justifying my doubts :-)
>

Please keep in mind that all commercial vendors need to differentiate their
own product among competitors.
So, the race is: do you have this functionality or not?
When you will carefully study all the functionality available in different
products you will find some of them are simply useless but allow marketing
teams to shine. :)


> But as I said -- I'd really like to give your product a try. There are
> quite some aspects that the public documenttaion does not cover -- for
> example, what will happen if the Sentinel Plugin gets incomplete file
> change history, or gets swamped with change notifications? How will it
> work with uncommon file systems, and what happens if backups are too
> rare so the list of changed files overflows?
>

The idea of an "incremental accelerator" for common local filesystems is
great.
But the inotify api is not the best solution for this case IMVHO, YMMV. The
main issue is the scalability issue.
You want to use "incremental accelerator" for all your high volume storage,
but inotify functionality does not scale.
The more directories you want to inotify the more resources it requires. A
tradeoff, sure.
I did a prototype of this feature years ago (no coding LLM existed yet) and
it wasn't worth the effort (the architecture was different, when fd starts
it spawns a special thread and registers required watches from the config
file, so incremental job can ask this thread what changed - it was a never
ending story of issues, race conditions, edge cases, etc.).

Today there is a modern and scalable fanotify API (or FSEvents in macOS
or USN Change Journal in win) which should be used instead, again IMVHO,
YMMV.

best regards
-- 
Radosław Korzeniewski
rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Radosław K. <rad...@ko...>]

Hi,

wt., 28 kwi 2026 o 20:06 Heitor Faria <hei...@gm...> napisał(a):

> So I guess Bacula Systems product, Bacula Enterprise, violates AGPL too...
>

No, Bacula Systems owns the full copyright to the Bacula code and releases
it under the AGPLv3 as Bacula Community.
Any external contributors had to sign
https://www.bacula.org/en/FLA-bacula.en.pdf

R.
-- 
Radosław Korzeniewski
rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Arno L. <al...@it...>]

Hi Heitor, all,

I'm replying on-list to an email sent to me directly. I hope I'll not 
sanitize qoutes too much, removing inappropriate amounts of context. 
Heitor agreed to me replying here, by the way.

Am 28.04.2026 um 17:11 schrieb Heitor Faria:
...
>      > I'm still deciding about the licensing. Any insights? For now, it is
>      > just full copyright.
> 
>     This, frankly, sounds dubious... Full copyright is not telling me
>     anything. Doe you mean "all rights reserved"?
> 
> 
> The plugins are proprietary, all rights reserved, commercial license. I 
> am still deciding whether to release a community edition with a subset 
> of features. No decision yet.

Ok. The parallel discussion about licensing should already prove 
helpful. I must admit I'm surprised you were not aware of the can of 
worms you were opening, though -- you're working with FLOSS software 
long enough, and with a "Law degree with specialization in software 
licensing" this should never have happened.

> 
>      > I have all the documentation, which I'm publishing progressively.
>      > Please verify the AKA Sentinel Plugin Whitepaper: https://
>      > baculaenterprise.com.br/wp-content/uploads/2026/04/PodHeitor-
>     <http://baculaenterprise.com.br/wp-content/uploads/2026/04/PodHeitor->
>      > Incremental-Accelerator-Ransomware-Detection-and-Remediation-
>     Plugin.pdf
>      > <https://baculaenterprise.com.br/wp-content/uploads/2026/04/
>     PodHeitor- <https://baculaenterprise.com.br/wp-content/
>     uploads/2026/04/PodHeitor->
>      > Incremental-Accelerator-Ransomware-Detection-and-Remediation-
>     Plugin.pdf>
> 
>     In this particular case, I see a very specific solution, seems to be a
>     combination of incremental accelerator where you try to use the change
>     tracking to also react to bulk changes. Good idea, in a way, although I
>     doubt it will work reliably. However, that's not based on actual
>     knowledge as I have not tried your plugin/add-on.
> 
> 
> Commvault uses a similar technique. So that's how I know it is 
> beneficial and feasible. =)

Well, Commvault has years of engineering and a rather large team of 
developers working on those things, which I like to understand as 
justifying my doubts :-)

But as I said -- I'd really like to give your product a try. There are 
quite some aspects that the public documenttaion does not cover -- for 
example, what will happen if the Sentinel Plugin gets incomplete file 
change history, or gets swamped with change notifications? How will it 
work with uncommon file systems, and what happens if backups are too 
rare so the list of changed files overflows?

>      >     Right now, I read big claims and no details at all, your github
>      >     pages do
>      >     not show anything about those,
>      >
>      > Because they are still private repositories.
> 
>     Do you intend to make those visible?
> 
> 
> Possibly, at least partially. I am considering opening the plugin API 
> layer and keeping the backend logic proprietary — similar to what some 
> commercial Bacula add-ons do. Nothing committed yet.

It might be a good idea to discuss with other contributors how such an 
API should be structured and implemented; I don't think it's beneficial 
if we have a number of incompatible but open plugin APIs existing next 
to each other, when a common effort might lead to a better product and 
give you all more time to work on differentiating solutions. Perhaps 
even on solutions that complement each other, and don't directly compete.

> 
>      >     your web site is full of statements I
>      >     find, politely expressed, doubtful, and your offers seem to
>      >     aggressively
>      >     target Bacula Systems and other large backup software.
>      >
>      > I wouldn't say aggressive, but competitive.
> 
>     Fair enough in some respect, although I must admit I don't see how your
>     Bacula Enterprise Appliance can be competitive. I also think your
>     references are to a large extent not yours, but from Bacula Systems.
>     The
>     same for all the plugins you mention.
> 
> 
>   We are progressively adjusting our website.

Progressively removing incorrect claims may prove to be a bit too 
relaxed. But of course it's your web site. On the other hand, it's also 
your credibility and, perhaps, your lawyer's bill.

>      >     Can you show us anything that would kind of prove you can
>     actually
>      >     deliver on those promises? -- After all, you're using this
>     mailing list
>      >     for your marketing, which I consider reasonable (ok, borderline
>      >     reasonable :-) as it's surely interesting for Bacula users,
>     but you're
>      >     not giving us anything to discuss. I want this to remain a
>      >     discussion list.
>      >
>      > I'm sorry about the tone and shallowness. Before you sent this, I
>     had
>      > already instructed my AI to provide more technical details on those
>      > launch statements.
> 
>     And now I'm really disappointed... does this imply I'm conversing with
>     an AI, not a person now?
> 
> 
> No - you are conversing with me, Heitor.

I'll trust you on this. I must admit i'm a bit dissatisfied with recent 
developments (not only here, not only with you, and not only on public 
email lists) but I will probably start filtering known AI users pretty 
aggressively in the near future, and I might even investigate how I can 
automatically send rude replies to people who think it's fair to send 
their thoughtlessly generated "contents" my way so I can then spend time 
and effort on fixing their slop. Anyway, that is a digression here. 
Sorry for that.

Anyway -- I'd still be interested in seeing what you have, not only 
because I'm curious, but also because I think that, if your claims are 
even remotely realistic, there may be some value to Bacula users, and 
I'm very much interested in that.

Cheers,

Arno


-- 
Arno Lehmann

IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Heitor F. <hei...@gm...>]

Olá,

On Tue, Apr 28, 2026 at 3:57 PM Lie wang <wan...@ou...> wrote:

> Heitor, você é conhecido no Brasil por ter enganado seus clientes. Você
> mesmo sabe o que aconteceu, ou já esqueceu tudo isso? Vamos esperar para
> ver quem mais, entre seus antigos clientes, se manifestará sobre o assunto.
> Um vigarista sempre será um vigarista e um mentiroso sempre será um
> mentiroso.
>

Enaganado meus clientes? Deixa de ser palhaça.
Eu fui afastado da empresa da qual fundei em 2014, por decisão judicial
promovida por uma pessoa a quem doei cotas societárias, e utilizou de
política para conseguir uma decisão absurda. Inclusive, falsamente acusado
de estar probido de dar suporte aos clientes.
Essa mesma pessoa, cessou suporte a todos os clientes, desviou recursos e
negócios da Bacula do Brasil Ltda (que inclusive possui um débito grande
junto á Bacula Systems SA) para a empresa Nomaland, inclusive incorrendo em
FRAUDE CONTRA CREDORES.
O mais incrível disso tudo, é a Bacula Systems SA apoiar uma FRAUDE CONTRA
CREDORES contra um DEVEDOR que deve a ela mesmo. Não havia de esperar nada
mesmo que toma golpe até de por engenharia social, onde clientes pagaram
por licenças através de dados bancários falsos.

Atte.

Re: [Bacula-users] [SPAM] Re: PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Arno L. <al...@it...>]

I propose you leave this part of the conversation off the list.

Not only due to the language.

Cheers,

Arno
-- 
Arno Lehmann

IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück

[Bacula-users] [SPAM] Re: PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Lie w. <wan...@ou...>]

Heitor, você é conhecido no Brasil por ter enganado seus clientes. Você mesmo sabe o que aconteceu, ou já esqueceu tudo isso? Vamos esperar para ver quem mais, entre seus antigos clientes, se manifestará sobre o assunto. Um vigarista sempre será um vigarista e um mentiroso sempre será um mentiroso.


________________________________
From: Rob Gerber <ro...@cr...>
Sent: Tuesday, April 28, 2026 10:03 PM
To: Heitor Faria <hei...@gm...>
Cc: bacula-users <bac...@li...>
Subject: Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

I am not a lawyer, or an expert in AGPLv3. I will try to help, but please understand that I may make a mistake.

It is complicated, but no, I very much doubt that Bacula Enterprise is violating any licenses.

The right to license any code belongs to the entity or person who owns the copyright on that code.

Bacula Systems owns the copyright to the Bacula Enterprise code. They can release it under any license they wish. They have released some of the Bacula Enterprise code in Bacula CE under the AGPL. They can also choose to release Bacula Enterprise code under any other license they wish.

I think it might be more complicated for code to go backwards, from Bacula CE to Bacula Enterprise. For this reason, the Bacula CE project probably requires any CE contributors to transfer (assign) copyright of the contributed code to the Bacula CE project foundation. The Bacula CE project can then license the contributed code back to Bacula Systems, under any license they wish. I believe most Bacula CE developers are employees of Bacula Systems.

As the copyright holder of code you personally authored,  you are entitled to license it under any terms you wish. However, you wish to use your code as part of another software project, Bacula CE. Bacula CE's license (AGPLv3) explicitly requires that any code that that interacts "too closely" with bacula also be licensed AGPLv3. What does "too closely" mean? I am not certain, but I have heard that an AGPLv3 program cannot use libraries not licensed AGPLv3. Based on what Radoslaw and others are saying, I suspect plugins are also covered by this restriction.

I think there are probably limits to the "infectious" nature of the AGPLv3. Radoslaw's program only interacts via the bconsole interface, in the same way a human might. Maybe this is a clue to ways to work around this issue. Maybe a config file is not covered by the AGPLv3, for example. Read carefully what Radoslaw said about using MetaPlugin. Maybe it can help you.

Understanding the requirements of the licenses of the open source projects you wish to use is just as important as the code you write, maybe moreso.

You should be aware that AIs are very unskilled for legal tasks. You should not rely on an AI to assist with legal advice. You need to retain an experienced intellectual property lawyer.


Robert Gerber
402-237-8692
ro...@cr...<mailto:ro...@cr...>



On Tue, Apr 28, 2026, 1:07 PM Heitor Faria <hei...@gm...<mailto:hei...@gm...>> wrote:
So I guess Bacula Systems product, Bacula Enterprise, violates AGPL too...

Atenciosamente,

Heitor faria (Miami)
https://www.youtube.com/@podheitor
WhatsApp: +1 786-726-1749 | +55 61 98268-4220

Em ter., 28 de abr. de 2026, 13:05, Radosław Korzeniewski <rad...@ko...<mailto:rad...@ko...>> escreveu:
Hi,

wt., 28 kwi 2026 o 18:25 Heitor Faria <hei...@gm...<mailto:hei...@gm...>> napisał(a):
PVT. Not joking. Why dont you find is a valid solution?

b/c you do not have the right to change the license of the product you do not own. (caveat: I'm assuming you did not get a dual-license from Bacula Systems for your product, right?)
Your patch does not change the license, so it will be useless from the legal standpoint.

Sure, you can do whatever you want in your own lab. You can mix and match licenses, change the code, produce the new code, even proclaim independence for this mix. Nobody's care.
But when you want to publish or provide services to your customers, you need to comply with the license.

You can't build a commercial solution without publishing or providing services.
All you can do is to split/separate your product on the AGPLv3 and the proprietary parts.
As I wrote, I did something like this for IBAdmin - an advanced webgui which does not violate AGPLv3 b/c it is just exec('bconsole') to manage Bacula. No plugins, no problem.
So, to follow these principles you can use MetaPlugin infrastructure to provide two layer solutions - plugin and backend. In this case all you need is to provide opensource (AGPLv3) FD plugin based on MetaPlugin protocol.
This should work for application plugins like K8S available in Bacula or vSphere advertised by you.

On the other hand, your "Incremental Accelerator, Ransomware Detection and Remediation Plugin" is already two layer solution (as I can understand from the overall architecture), so only the FD part needs to be AGPLv3 compliant and your "PodHeitor Sentinel Daemon" can use any license you want - as long as the boundary is unix socket/named pipes as described at the architecture.

Ask for advice from your local Intellectual Property lawyer to ensure your ideas are compliant and your business can survive.
I wish you a very well for your new business solution. Fingers crossed.

best regards
--
Radosław Korzeniewski
rad...@ko...<mailto:rad...@ko...>
_______________________________________________
Bacula-users mailing list
Bac...@li...<mailto:Bac...@li...>
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Rob G. <ro...@cr...>]

I am not a lawyer, or an expert in AGPLv3. I will try to help, but please
understand that I may make a mistake.

It is complicated, but no, I very much doubt that Bacula Enterprise is
violating any licenses.

The right to license any code belongs to the entity or person who owns the
copyright on that code.

Bacula Systems owns the copyright to the Bacula Enterprise code. They can
release it under any license they wish. They have released some of the
Bacula Enterprise code in Bacula CE under the AGPL. They can also choose to
release Bacula Enterprise code under any other license they wish.

I think it might be more complicated for code to go backwards, from Bacula
CE to Bacula Enterprise. For this reason, the Bacula CE project probably
requires any CE contributors to transfer (assign) copyright of the
contributed code to the Bacula CE project foundation. The Bacula CE project
can then license the contributed code back to Bacula Systems, under any
license they wish. I believe most Bacula CE developers are employees of
Bacula Systems.

As the copyright holder of code you personally authored,  you are entitled
to license it under any terms you wish. However, you wish to use your code
as part of another software project, Bacula CE. Bacula CE's license
(AGPLv3) explicitly requires that any code that that interacts "too
closely" with bacula also be licensed AGPLv3. What does "too closely" mean?
I am not certain, but I have heard that an AGPLv3 program cannot use
libraries not licensed AGPLv3. Based on what Radoslaw and others are
saying, I suspect plugins are also covered by this restriction.

I think there are probably limits to the "infectious" nature of the AGPLv3.
Radoslaw's program only interacts via the bconsole interface, in the same
way a human might. Maybe this is a clue to ways to work around this issue.
Maybe a config file is not covered by the AGPLv3, for example. Read
carefully what Radoslaw said about using MetaPlugin. Maybe it can help you.

Understanding the requirements of the licenses of the open source projects
you wish to use is just as important as the code you write, maybe moreso.

You should be aware that AIs are very unskilled for legal tasks. You should
not rely on an AI to assist with legal advice. You need to retain an
experienced intellectual property lawyer.


Robert Gerber
402-237-8692
ro...@cr...



On Tue, Apr 28, 2026, 1:07 PM Heitor Faria <hei...@gm...> wrote:

> So I guess Bacula Systems product, Bacula Enterprise, violates AGPL too...
>
> Atenciosamente,
>
> Heitor faria (Miami)
> https://www.youtube.com/@podheitor
> WhatsApp: +1 786-726-1749 | +55 61 98268-4220
>
> Em ter., 28 de abr. de 2026, 13:05, Radosław Korzeniewski <
> rad...@ko...> escreveu:
>
>> Hi,
>>
>> wt., 28 kwi 2026 o 18:25 Heitor Faria <hei...@gm...> napisał(a):
>>
>>> PVT. Not joking. Why dont you find is a valid solution?
>>>
>>
>> b/c you do not have the right to change the license of the product you do
>> not own. (caveat: I'm assuming you did not get a dual-license from Bacula
>> Systems for your product, right?)
>> Your patch does not change the license, so it will be useless from the
>> legal standpoint.
>>
>> Sure, you can do whatever you want in your own lab. You can mix and match
>> licenses, change the code, produce the new code, even proclaim independence
>> for this mix. Nobody's care.
>> But when you want to publish or provide services to your customers, you
>> need to comply with the license.
>>
>> You can't build a commercial solution without publishing or providing
>> services.
>> All you can do is to split/separate your product on the AGPLv3 and the
>> proprietary parts.
>> As I wrote, I did something like this for IBAdmin - an advanced
>> webgui which does not violate AGPLv3 b/c it is just exec('bconsole') to
>> manage Bacula. No plugins, no problem.
>> So, to follow these principles you can use MetaPlugin infrastructure to
>> provide two layer solutions - plugin and backend. In this case all you need
>> is to provide opensource (AGPLv3) FD plugin based on MetaPlugin protocol.
>> This should work for application plugins like K8S available in Bacula or
>> vSphere advertised by you.
>>
>> On the other hand, your "Incremental Accelerator, Ransomware Detection
>> and Remediation Plugin" is already two layer solution (as I can understand
>> from the overall architecture), so only the FD part needs to be AGPLv3
>> compliant and your "PodHeitor Sentinel Daemon" can use any license you want
>> - as long as the boundary is unix socket/named pipes as described at the
>> architecture.
>>
>> Ask for advice from your local Intellectual Property lawyer to ensure
>> your ideas are compliant and your business can survive.
>> I wish you a very well for your new business solution. Fingers crossed.
>>
>> best regards
>> --
>> Radosław Korzeniewski
>> rad...@ko...
>>
> _______________________________________________
> Bacula-users mailing list
> Bac...@li...
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Heitor F. <hei...@gm...>]

So I guess Bacula Systems product, Bacula Enterprise, violates AGPL too...

Atenciosamente,

Heitor faria (Miami)
https://www.youtube.com/@podheitor
WhatsApp: +1 786-726-1749 | +55 61 98268-4220

Em ter., 28 de abr. de 2026, 13:05, Radosław Korzeniewski <
rad...@ko...> escreveu:

> Hi,
>
> wt., 28 kwi 2026 o 18:25 Heitor Faria <hei...@gm...> napisał(a):
>
>> PVT. Not joking. Why dont you find is a valid solution?
>>
>
> b/c you do not have the right to change the license of the product you do
> not own. (caveat: I'm assuming you did not get a dual-license from Bacula
> Systems for your product, right?)
> Your patch does not change the license, so it will be useless from the
> legal standpoint.
>
> Sure, you can do whatever you want in your own lab. You can mix and match
> licenses, change the code, produce the new code, even proclaim independence
> for this mix. Nobody's care.
> But when you want to publish or provide services to your customers, you
> need to comply with the license.
>
> You can't build a commercial solution without publishing or providing
> services.
> All you can do is to split/separate your product on the AGPLv3 and the
> proprietary parts.
> As I wrote, I did something like this for IBAdmin - an advanced
> webgui which does not violate AGPLv3 b/c it is just exec('bconsole') to
> manage Bacula. No plugins, no problem.
> So, to follow these principles you can use MetaPlugin infrastructure to
> provide two layer solutions - plugin and backend. In this case all you need
> is to provide opensource (AGPLv3) FD plugin based on MetaPlugin protocol.
> This should work for application plugins like K8S available in Bacula or
> vSphere advertised by you.
>
> On the other hand, your "Incremental Accelerator, Ransomware Detection
> and Remediation Plugin" is already two layer solution (as I can understand
> from the overall architecture), so only the FD part needs to be AGPLv3
> compliant and your "PodHeitor Sentinel Daemon" can use any license you want
> - as long as the boundary is unix socket/named pipes as described at the
> architecture.
>
> Ask for advice from your local Intellectual Property lawyer to ensure your
> ideas are compliant and your business can survive.
> I wish you a very well for your new business solution. Fingers crossed.
>
> best regards
> --
> Radosław Korzeniewski
> rad...@ko...
>

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Radosław K. <rad...@ko...>]

Hi,

wt., 28 kwi 2026 o 18:25 Heitor Faria <hei...@gm...> napisał(a):

> PVT. Not joking. Why dont you find is a valid solution?
>

b/c you do not have the right to change the license of the product you do
not own. (caveat: I'm assuming you did not get a dual-license from Bacula
Systems for your product, right?)
Your patch does not change the license, so it will be useless from the
legal standpoint.

Sure, you can do whatever you want in your own lab. You can mix and match
licenses, change the code, produce the new code, even proclaim independence
for this mix. Nobody's care.
But when you want to publish or provide services to your customers, you
need to comply with the license.

You can't build a commercial solution without publishing or providing
services.
All you can do is to split/separate your product on the AGPLv3 and the
proprietary parts.
As I wrote, I did something like this for IBAdmin - an advanced
webgui which does not violate AGPLv3 b/c it is just exec('bconsole') to
manage Bacula. No plugins, no problem.
So, to follow these principles you can use MetaPlugin infrastructure to
provide two layer solutions - plugin and backend. In this case all you need
is to provide opensource (AGPLv3) FD plugin based on MetaPlugin protocol.
This should work for application plugins like K8S available in Bacula or
vSphere advertised by you.

On the other hand, your "Incremental Accelerator, Ransomware Detection and
Remediation Plugin" is already two layer solution (as I can understand from
the overall architecture), so only the FD part needs to be AGPLv3 compliant
and your "PodHeitor Sentinel Daemon" can use any license you want - as long
as the boundary is unix socket/named pipes as described at the architecture.

Ask for advice from your local Intellectual Property lawyer to ensure your
ideas are compliant and your business can survive.
I wish you a very well for your new business solution. Fingers crossed.

best regards
-- 
Radosław Korzeniewski
rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[A.F.M <a.f...@pr...>]

I am not an expert but I doubt that you can just relicense a software as you like.

Heitor Faria <hei...@gm...> schrieb am Dienstag, 28. April 2026 um 18:59:

> That trap is easy to solve. Fork the bacula-fd code in AGPL and change that plugin_license part. 😁
>
> Atenciosamente,
>
> Heitor faria (Miami)
> https://www.youtube.com/@podheitor
> WhatsApp: +1 786-726-1749 | +55 61 98268-4220
>
> Em ter., 28 de abr. de 2026, 11:52, Radosław Korzeniewski <rad...@ko...> escreveu:
>
>> Hello,
>>
>> wt., 28 kwi 2026 o 17:14 Heitor Faria <hei...@gm...> napisał(a):
>>
>>> Hello AFM,
>>>
>>> Thank you for writing, though I will be direct: this message contains several factual inaccuracies, and I would like to correct them for the list's benefit.
>>>
>>> On legal challenges: No legal concerns have been raised in this thread that apply to my software. The prior discussion was about the plugin_license string required by the Bacula FD API — a technical detail I have already addressed.
>>
>> This is not a technical detail, but a valid field describing the proper license for the whole product.
>>
>>> That is not a legal challenge; it is a configuration value.
>>
>> For the Bacula Community you can't legally load the plugin which is not AGPLv3 licensed. You can provide external tools as "full copyright", i.e. I prepared the proprietary (open-source version is available too) tool - IBAdmin which manages Bacula Community without any plugin, just with simple cli execs and sql queries.
>>
>>> The plugins are proprietary commercial software. The documentation is public. Pre-built packages (RPM, DEB, Windows installer) are available for evaluation. That has been consistent throughout this discussion.
>>
>> If you distribute this software, or provide services based on a modified version of it over a network, you have an obligation to provide the corresponding source code to the users, unless you have a separate license that permits otherwise.
>>
>> I hope this is clear.--
>>
>> Radosław Korzeniewski
>> rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Radosław K. <rad...@ko...>]

Hi,

wt., 28 kwi 2026 o 17:59 Heitor Faria <hei...@gm...> napisał(a):

> That trap is easy to solve. Fork the bacula-fd code in AGPL and change
> that plugin_license part. 😁
>

Let's be serious. It is not working this way.

best regards
-- 
Radosław Korzeniewski
rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Radosław K. <rad...@ko...>]

Hello,

wt., 28 kwi 2026 o 17:14 Heitor Faria <hei...@gm...> napisał(a):

> Hello AFM,
>
> Thank you for writing, though I will be direct: this message contains
> several factual inaccuracies, and I would like to correct them for the
> list's benefit.
>
> On legal challenges: No legal concerns have been raised in this thread
> that apply to my software. The prior discussion was about the
> plugin_license string required by the Bacula FD API — a technical detail I
> have already addressed.
>

This is not a technical detail, but a valid field describing the proper
license for the whole product.


> That is not a legal challenge; it is a configuration value.
>

For the Bacula Community you can't legally load the plugin which is not
AGPLv3 licensed. You can provide external tools as "full copyright", i.e. I
prepared the proprietary (open-source version is available too) tool -
IBAdmin which manages Bacula Community without any plugin, just with simple
cli execs and sql queries.

The plugins are proprietary commercial software. The documentation is
> public. Pre-built packages (RPM, DEB, Windows installer) are available for
> evaluation. That has been consistent throughout this discussion.
>

If you distribute this software, or provide services based on a modified
version of it over a network, you have an obligation to provide the
corresponding source code to the users, unless you have a separate license
that permits otherwise.

I hope this is clear.
-- 
Radosław Korzeniewski
rad...@ko...

Re: [Bacula-users] PodHeitor Incremental Accelerator, Ransomware Detection and Remediation Plugin for Bacula

[Heitor F. <hei...@gm...>]

That trap is easy to solve. Fork the bacula-fd code in AGPL and change that
plugin_license part. 😁

Atenciosamente,

Heitor faria (Miami)
https://www.youtube.com/@podheitor
WhatsApp: +1 786-726-1749 | +55 61 98268-4220

Em ter., 28 de abr. de 2026, 11:52, Radosław Korzeniewski <
rad...@ko...> escreveu:

> Hello,
>
> wt., 28 kwi 2026 o 17:14 Heitor Faria <hei...@gm...> napisał(a):
>
>> Hello AFM,
>>
>> Thank you for writing, though I will be direct: this message contains
>> several factual inaccuracies, and I would like to correct them for the
>> list's benefit.
>>
>> On legal challenges: No legal concerns have been raised in this thread
>> that apply to my software. The prior discussion was about the
>> plugin_license string required by the Bacula FD API — a technical detail I
>> have already addressed.
>>
>
> This is not a technical detail, but a valid field describing the proper
> license for the whole product.
>
>
>> That is not a legal challenge; it is a configuration value.
>>
>
> For the Bacula Community you can't legally load the plugin which is not
> AGPLv3 licensed. You can provide external tools as "full copyright", i.e. I
> prepared the proprietary (open-source version is available too) tool -
> IBAdmin which manages Bacula Community without any plugin, just with simple
> cli execs and sql queries.
>
> The plugins are proprietary commercial software. The documentation is
>> public. Pre-built packages (RPM, DEB, Windows installer) are available for
>> evaluation. That has been consistent throughout this discussion.
>>
>
> If you distribute this software, or provide services based on a modified
> version of it over a network, you have an obligation to provide the
> corresponding source code to the users, unless you have a separate license
> that permits otherwise.
>
> I hope this is clear.
> --
> Radosław Korzeniewski
> rad...@ko...
>