jobquery custom report

Hi Carlos, thank you for the fast aswer.

I run the command

>jobquery -i input_jobquery.txt

where the file input_jobquery.txt has these lines:

. type: save job

show

print

I'll keep only the usefull fields, but I need to know if I can have the "command:" field in only one line instead of:

command: \

"savefs -s s183b001.it183.corpintra.net -c S194INFORMA01 -g DOMINO -p -l full\

-R -v -F NOTES:";

and filter only the  jobs runned in the last 24h using some kind of instruction like:

print start time<1338832804   ######## it doesn't work

I see that the first result is a 48h old save job.

I don't undestand how can I use mminfo to have a job schedule report... it seems that it provide a "media oriented" report.

Simone

Hi Simone,

I would suggest you to use mminfo instead, as depending on the configuration you have for the jobsdb retention and size, this information could be already deleted from the DB, and if you use jobquery you could be missing some information.

What exactly do you mean by job type?

What details would you exactly need so we can find the correct syntax?

Thank you.

Carlos.

Simone,

If your looking for all the savegroups which ran in the last 24 hours you can do this via mminfo quite easily.

You can then add further flags to specify a particular client, saveset etc into the query and report many things.

e.g.

mminfo -v -r client,name,ssid,totalsize,ssflags

mminfo -t "1 day ago" -r client,name,ssid,totalsize,ssflags

This would give everything from the last 24 hrs from the time the command was run.

You could also specify something like: -

mminfo -q client=, savetime="1 day ago" -r name,ssid,totalsize,volume,ssflags

which would give you all savesets for a particular client secured over the past 24 hrs

There are many variations you can choose to provide the info you require. You can also use the -ot flags after the query (-q) and reporting (-r) statements to order by time and the last lines reported will be the latest backups.

See the command line reference guide for all options and further syntax

Regards,

Bill Mason

Thank you William.

I'm sorry for my little Networker experience...

I'd like to obtain a schedule report with these information for each backup:

client name,

group,

schedule,

backup status,

backup command,

start time,

end time

I don't need to know the totalsize or the ssid information...

I found these information with the jobquery command, but I can't format the output as I need. Can I obtain the same information with mminfo?

Simone

Thank you very much Bill,

I want to create a report for our monitor team, so they can restart the failed backups directly with the backup command.

Simone

Hello, I have another question for you!

In the report obtained with mminfo I can't see the running backups nor the failed backup... I used this command:

>mminfo -t "1 day ago" -r client(30),group(20),level,ssflags,savetime(20),sscomp(20),name

Simone,

You should be able to tell if the backup is failed or running from the ssflags reported. From the command line reference, incomplete = failed, in-progress = running, f and v flags show backup is complete and success: -

ssflags - The save set flags summary, one or more characters in the set CvrENiRPKIFk, for continued, valid, purged (recoverable), eligible for recycling, NDMP generated, incomplete, raw(not for savesets backed up

using rawasm), snapshot, cover, in-progress and finished (ended), checkpoint restart enabled, respectively.

Personally I would not use the command line produced by NetWorker to restart the backup, much easier way is to restart the group the client belongs to as this will only backup anything which failed or was not complete when the group finished or was stopped. It is much safer and clearer then than starting manually outside of NetWorker and could reduce the amount of data you secure as we automatically do what is required and command line may do all data again.

Regards,

Bill Mason

I read the command reference and I found the same information you're talking about, but it seems that I can't find the failed backup nor the running backup, all the lines have vF ssflag (valid and Finished).

Maybe I should change the "-t" option ?

I finally found my failed backup with the vrEiF ssflag...

Thank you very much.

Simone

Actually, jobdb is ideal for this kind of report. mminfo won't report savefs failed backups for example - simply because they do not get into media database as they failed at savefs level.  I made a command for that which I call nsrcli, but I will reveal code after summer vacation on my blog... I need to develop few more functions inside.  With that said, I sort events by time, but I use last 24 only because I keep jobdb for 24 hours.  In NW8 this will change and I have to rewrite certain queries, but that's future.

Hello Hrvoje!

so you say that mminfo is not the best way to create a job report? which command do you use for the jobdb queries?

thank you very much

Simone

jobquery command. You can use mminfo, but it won't cover all you want. Ideally, you use both as both commands can and will deliver information which is interesting to backup admin.

Specific report can be achieved with specific query, but more specific shape you are after - more you have to shape it (using awk, sed or whatever).