Kern,
Actually, no. I have three separate SDs, with their own Device entry in
the bacula-dir.conf, running on separate hosts, etc. Of those, two are
the ones in question, and other than the name, they're identical. I
only included one example, but there are two like that, other than the name.
You're right, though. I should've made that more clear. I apologize
for that.
On top of that, I did have multiple devices per SD, but that's a
separate issue.
Lloyd
On 6/2/19 6:31 AM, Kern Sibbald wrote:
>
> Hello Lloyd,
>
> I suspect that you might be mixing up two concepts. One is storage
> daemons: to have more than one you must have two different (with
> different names or directories) bacula-sd.conf files. If you have
> only one, then you are dealing with a concept of Storage Devices
> rather than Storage Daemons. One storage daemon can have multiple
> storage devices.
>
> Best regards,
>
> Kern
>
> On 5/31/19 7:38 PM, Lloyd Brown wrote:
>> On 5/31/19 10:46 AM, Martin Simmons wrote:
>>> The "waiting on max Storage jobs" is caused by the Maximum Concurrent Jobs
>>> in
>>> the Storage section of bacula-dir.conf.
>>>
>>> It would be useful to see the output of "show storage" while jobs are
>>> waiting.
>>>
>>> __Martin
>> Martin,
>>
>> I'm not entirely certain that this is a good test case, but I tried to
>> re-create the symptoms, and captured the output of both "status dir" and
>> "status storage", along with some config extracts, all of which are
>> attached here. I was only able to get one job ("fslhome_c9") to be in
>> the "waiting on max Storage jobs" state, and it resolved a few minutes
>> later. Thus why I'm not sure it is the best test case. But it did have
>> that same error, and there were only 9-10 jobs running on that FD and SD
>> at the time, AFAICT.
>>
>> Note that the large number of devices on the SD, was a misguided idea
>> from several years ago, that didn't quite do what I wanted. But it also
>> didn't seem to break anything, so I haven't gotten around to cleaning it
>> up. We were all young and naive once, right? ;-)
>>
>> If this doesn't provide the insight you were looking for, I will try to
>> re-create it again, and capture the same info for you.
>>
>> Lloyd
>>
>>
>>
>>
>> _______________________________________________
>> Bacula-users mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/bacula-users
--
Lloyd Brown
HPC Systems Administrator
Office of Research Computing
Brigham Young University
http://marylou.byu.edu
_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users
