Help to create daily reports for backups of the last 24h

[tpesselier]

Hi

I am looking for help to create daily reports for backups of the last 24h, one report by domain with OC 7.1.8

(domaine,nodename,status,schedname,errorcode,level(incr/img),size,numberfile,numbernotbackupedfile,numberscannedfile,startedhour,finishhour,duration,idlewait,mediawaittime,total reductionratio)

Really not easy

 

[tpesselier]

I had the chance to talk WITH tsm developer and here is a query that works well.
The ideal is to sort the failed from above and it would be even better.
IBM may integrate it into the OC template

select domain_name as "Domain",
entity "Client",
(case when successful='YES' then 'success' else 'failed' end) as "Status",
coalesce(schedule_name, '-') as "Schedule",
coalesce(bytes_written, bytes, 0) as "Size",
affected as "Files",
failed as "Files Not Backed Up",
examined as "Files Scanned",
start_time as "Started",
end_time as "Finished",
timestampdiff(8, end_time-start_time) || ' hours ' || mod(timestampdiff(4, end_time-start_time), 60) || ' minutes' as "Duration (hour)",
(case when bytes_protected > 0 then floor(((coalesce(bytes_protected, 0)*1.0 - coalesce(bytes_written, 0))/bytes_protected)*100.0) else 0 end ) as "Total Reduction Ratio"
from summary_extended inner join nodes on entity=node_name
where activity='BACKUP' and activity_type='SESSION_END' and end_TIME>=(current_timestamp - 24 hours) and domain_name='domain_test'

 

[marclant]

That's a good query. Personally though, I don't care about successful backup, there's no action that needs to be taken with them. I'd filter it with SUCCESSFUL='NO' only because those are the ones that need to be addressed. But that's just a personal preference.

 

[tpesselier]

Hey maclan i think the same but my users want to see everything in their domain

I also proposed the idea of doing the same thing with TDP4VE but using the VMTAG to identify the owner of the vm because you can not identify a vm with a nodename and a domains ,i'm waiting for the answer


after I agree it's easier to highlight the failed especially if you have 1200 vm like me, but at home every owner wants to have the vision of these machines in their domain

I find that IBM is not consistent between its vision of detached domain administration and the way it is supervising

 

[tmaria]

I had the chance to talk WITH tsm developer and here is a query that works well.
The ideal is to sort the failed from above and it would be even better.
IBM may integrate it into the OC template

select domain_name as "Domain",
entity "Client",
(case when successful='YES' then 'success' else 'failed' end) as "Status",
coalesce(schedule_name, '-') as "Schedule",
coalesce(bytes_written, bytes, 0) as "Size",
affected as "Files",
failed as "Files Not Backed Up",
examined as "Files Scanned",
start_time as "Started",
end_time as "Finished",
timestampdiff(8, end_time-start_time) || ' hours ' || mod(timestampdiff(4, end_time-start_time), 60) || ' minutes' as "Duration (hour)",
(case when bytes_protected > 0 then floor(((coalesce(bytes_protected, 0)*1.0 - coalesce(bytes_written, 0))/bytes_protected)*100.0) else 0 end ) as "Total Reduction Ratio"
from summary_extended inner join nodes on entity=node_name
where activity='BACKUP' and activity_type='SESSION_END' and end_TIME>=(current_timestamp - 24 hours) and domain_name='domain_test'

I ran this query and almost is what I wanted. I would like to have the size as it is reported in the dsmschedule.log file, next to the "Total number of bytes transferred:" . If it is MB to be displayed as MB if it is GB then as GB. How can this be done?

 

[tpesselier]

Hi tmaria ,

I try to do it too but it's not easy with coelse , i keep you informed

 

[rpandey]

I had the chance to talk WITH tsm developer and here is a query that works well.
The ideal is to sort the failed from above and it would be even better.
IBM may integrate it into the OC template

select domain_name as "Domain",
entity "Client",
(case when successful='YES' then 'success' else 'failed' end) as "Status",
coalesce(schedule_name, '-') as "Schedule",
coalesce(bytes_written, bytes, 0) as "Size",
affected as "Files",
failed as "Files Not Backed Up",
examined as "Files Scanned",
start_time as "Started",
end_time as "Finished",
timestampdiff(8, end_time-start_time) || ' hours ' || mod(timestampdiff(4, end_time-start_time), 60) || ' minutes' as "Duration (hour)",
(case when bytes_protected > 0 then floor(((coalesce(bytes_protected, 0)*1.0 - coalesce(bytes_written, 0))/bytes_protected)*100.0) else 0 end ) as "Total Reduction Ratio"
from summary_extended inner join nodes on entity=node_name
where activity='BACKUP' and activity_type='SESSION_END' and end_TIME>=(current_timestamp - 24 hours) and domain_name='domain_test'

I can't make your command working tpesselier. I am getting

Output of command redirected to file '0'
ANS8001I Return code 3
.

 

[moon-buddy]

remove the space before and after ">" in: "(case when bytes_protected > 0"

Should be (case when bytes_protected>0

 

[rpandey]

That's a good query. Personally though, I don't care about successful backup, there's no action that needs to be taken with them. I'd filter it with SUCCESSFUL='NO' only because those are the ones that need to be addressed. But that's just a personal preference.

Hi
Do you have this script with SUCCESSFUL='NO' ?

 

[marclant]

Hi
Do you have this script with SUCCESSFUL='NO' ?

You can just add and SUCCESSFUL='NO' at the end of the where clause of the query.