All jobs must have an associated client with suitable MaximumConcurrentJobs,
even if they don't use a client.  It is a feature :-(

__Martin


>>>>> On Thu, 20 Apr 2023 21:13:00 +0200, Justin Case said:
> 
> This seemed to say that the director’s client limit MaximumConcurrentJobs = 1 
> was hit. WHen I changed it to MaximumConcurrentJobs = 5 I was able to run the 
> Admin job concurrently. 
> 
> I still don’t understand why this job would count against the client job 
> limit, as I thought that it is a job that does something on the SD, not on 
> the client / FD. Do you understand this?
> 
> > On 20. Apr 2023, at 20:55, Justin Case <jus7inc...@gmail.com> wrote:
> > 
> > Hi Martin,
> > 
> >> On 20. Apr 2023, at 20:38, Martin Simmons <mar...@lispworks.com> wrote:
> >> 
> >> What is the output of the "status dir" command when the Admin job is 
> >> waiting?
> >> 
> > it says for the Admin job: is waiting on max Client jobs
> > what does that mean?
> > 
> >> When you say "Both jobs have set AllowMixedPriority = yes." do you mean all
> >> jobs that are running at the time you want to Admin job to run?
> > 
> > For now I have  job running for testing, and the Admin job, and both have 
> > AllowMixedPriority = yes.
> > 
> >> 
> >>>>>>> On Thu, 20 Apr 2023 12:14:57 +0200, Justin Case said:
> >>> 
> >>> Greetings to all,
> >>> 
> >>> I have the simple Admin job "truncate-pools-all” (see further down) and I 
> >>> would like to be able to run it concurrently while some backup job 
> >>> “backup1" (see further down)  is running. Lets say backup jobs have 
> >>> Priority = 20.
> >>> The Runscript Console command has Priority = 10 and uses drive number 9, 
> >>> which is very likely not in use when the Admin job is started. The backup 
> >>> jobs usually use drive number 0. Both jobs have set AllowMixedPriority = 
> >>> yes.
> >>> While I can successfully run this command in bconsole concurrently when a 
> >>> backup job is already running, when starting the Admin job the Bacula 
> >>> queuing algorithm puts this Admin job in the queue and it is waiting 
> >>> until the currently running backup job has finished. My understanding was 
> >>> that this is normal behaviour when AllowMixedPriority = no (default). 
> >>> However, I have explicitly enabled AllowMixedPriority and still it does 
> >>> not work. The MaximumConcurrentJobs are 5 or 20 in different components, 
> >>> except for the SD file autochanger drives, there it is set to 1.
> >>> 
> >>> My first guess would be, that somehow the SD does not automagically make 
> >>> use of the available unoccupied drives of the autochanger (although the 
> >>> default behaviour should be AutoSelect = yes). So it tells the director 
> >>> that the drive is busy and then the director makes the Admin job wait.
> >>> But I could also be wrong, as I am not an expert on Bacula topics.
> >>> 
> >>> What would I need to change to get this to work as expected and described 
> >>> at the top of this mail?
> >>> 
> >>> Thanks for considering my question and have a good time,
> >>> J/C
> >>> 
> >>> 
> >>> Job {
> >>> Name = "truncate-pools-all"
> >>> Type = "Admin"
> >>> JobDefs = "default1"
> >>> Enabled = no
> >>> Runscript {
> >>> RunsOnClient = no
> >>> RunsWhen = "Before"
> >>> Console = "truncate volume allpools storage=unraid-tier1-storage drive=9"
> >>> }
> >>> Priority = 10
> >>> AllowDuplicateJobs = no
> >>> AllowMixedPriority = yes
> >>> }
> >>> 
> >>> JobDefs {
> >>> Name = "default1"
> >>> Type = "Backup"
> >>> Level = "Full"
> >>> Messages = "Standard"
> >>> Pool = "default1"
> >>> FullBackupPool = "default1"
> >>> IncrementalBackupPool = "default1"
> >>> Client = “machine1"
> >>> Fileset = "EmptyFileset"
> >>> MaxFullInterval = 2678400
> >>> SpoolAttributes = yes
> >>> Priority = 20
> >>> AllowIncompleteJobs = no
> >>> Accurate = yes
> >>> AllowDuplicateJobs = no
> >>> }
> >>> 
> >>> This is the backup job that is already running:
> >>> 
> >>> Job {
> >>> Name = “backup1"
> >>> Pool = “pool1"
> >>> FullBackupPool = “pool1"
> >>> IncrementalBackupPool = “pool1"
> >>> Fileset = “fs1"
> >>> Schedule = “schd1"
> >>> JobDefs = “default2"
> >>> Enabled = yes
> >>> AllowIncompleteJobs = no
> >>> AllowDuplicateJobs = no
> >>> AllowMixedPriority = yes
> >>> }
> >>> 
> >>> JobDefs {
> >>> Name = “default2"
> >>> Type = "Backup"
> >>> Level = "Full"
> >>> Messages = "Standard"
> >>> Pool = "default1"
> >>> Client = “machine1"
> >>> Fileset = "EmptyFileset"
> >>> Schedule = “sched2"
> >>> Priority = 20
> >>> Accurate = yes
> >>> }
> >>> 
> >>> 
> >>> 
> >>> _______________________________________________
> >>> Bacula-users mailing list
> >>> Bacula-users@lists.sourceforge.net
> >>> https://lists.sourceforge.net/lists/listinfo/bacula-users
> >>> 
> >> 
> > 
> 
> 


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

Reply via email to