bug or feature?

[Stefan Krüger]

Hello,

short question, if ein Start a Fullbackup manually, and this takes a lot of time (~20hours) and the next scheduled window is coming within this time, its a new client, it checks the database if there are previous jobs if no run a fullbackup. Bareos said I don't care about the running jobs, so I run also a Fullbackup of the same client ..

So my qustions is, is it normal that bareos only checks the database and not the running jobs? If not does exist a Bugreport for this?

Best regards

Stefan

[Florian Panzer - PLUSTECH]

Am 19.08.19 um 09:31 schrieb 'Stefan Krüger' via bareos-users:

> [...]

> So my qustions is, is it normal that bareos only checks the database and
> not the running jobs? If not does exist a Bugreport for this?

yes, AFAIK this is intended behavior.

--
Florian Panzer

-----------------------------------
PLUSTECH GmbH
Jäckstraße 35
96052 Bamberg
Telefon: +49 951 299 09 716
https://plustech.de/
Geschäftsführung: Florian Panzer
Amtsgericht Bamberg - HRB 9680
-----------------------------------
AWS Certified Solution Architect - Associate
AWS APN Consulting Partner
Mitglied in der Allianz für Cybersicherheit
des Bundesamtes für Sicherheit in der Informationstechnik

[Stefan Krüger]

And why? That makes no sense to me

[Andreas Rogge]

Am 19.08.19 um 10:41 schrieb 'Stefan Krüger' via bareos-users:

> And why? That makes no sense to me

The later backup needs to know the exit status of the former one to
decide what to do (i.e. upgrade to Full or not). How should Bareos know
wheter or not the still running job will exit successfully?

The only way is to not run the job twice which can be configured using
"allow duplicate jobs" and directives like that.

Best Regards,
Andreas
--
Andreas Rogge andrea...@bareos.com
Bareos GmbH & Co. KG Phone: +49 221-630693-86
http://www.bareos.com

Sitz der Gesellschaft: Köln | Amtsgericht Köln: HRA 29646
Komplementär: Bareos Verwaltungs-GmbH
Geschäftsführer: S. Dühr, M. Außendorf, J. Steffens, Philipp Storz

[Florian Panzer - PLUSTECH]

Am 19.08.19 um 10:52 schrieb Andreas Rogge:

> The later backup needs to know the exit status of the former one to
> decide what to do (i.e. upgrade to Full or not). How should Bareos
> know wheter or not the still running job will exit successfully?

Correct, plus: It is possible that the manually started job was modified
on purpose (different fileset selected, backing up to a different SD or
different volume pool or so) and therefore the manual backup task might
not fully "replace" the scheduled version of the task.

So, in this case, the behaviour is actually desireable.