Harald_Heidinger_152

How to determine if the report of job should write its report to database or to the filesystem

Discussion created by Harald_Heidinger_152 on Nov 23, 2016
What is the question?

Within Jobs (Windows/Unix...) - Tab there is the possibility to save a report either

.) to Database - in that case, the report which is generated on the operating system for the job will be transferred to DB
.) to Filesystem - in that case, the report will be generated on the OS and the file will persist there until it is deleted.
.) at error only (with both possibilities or with none)

.) both or all 3 options, in that case all of the described will be done
.) none NOTE in that case, the reportfile which is generated during the job is active will be deleted immediately after the job has ended on the OS in any case. That means, the report will not be accessible.

When which function should be used?

.) to Database:
This should be used, if the report has either to be saved on Automation Engine DB for revision purposes, or if the content of the report should be parsed for further action (it is also possible to parse the reportfile on the OS site but it is more complicated. This would be the default setting.

.) to Filesystem:
This is advisable for e.g. larger reports to avoid the reports will fill up your database and will cause longer reorganization times due increased DB usage, or if you need to have the reports saved on the device the task was running on. In that case, the reportfiles will not considered by regular Automic Reorganization, you might have to take care yourself to have the reportfiles deleted on a regulary base. (you can create an OS Job which deletes the reportfiles with a timestamp older than xx)

NOTE:
If a report is too large to be transferred to Database (the settings can be defined within UC_HOSTCHAR_DEFAULT MAX_REPORT_SIZE and REPORT_BLKSIZE) it will throw a message on the report of the Automic Task:

U0029007 Not all blocks (Limit '120') have been transferred [...]

in that case, the report will be written to database until the specific limit, and the report will also be saved to the OS site to have the complete report available.

my colleague Iris_Muttenthaler_7233 has written a nice contribution to the forum describing this situation in Detail:

https://community.automic.com/discussion/7362/explanation-of-error-message-u0029007-not-all-blocks-limit-120-have-been-transferred

.) at error only
This is advisable if you dont need the report for further processing and if you dont need to see the report at all (on Filesystem or on Database) within Automic system in case the Job runs fine. If you plan to use this, please consider to contact your security and revision department if this is OK for them, note, if you dont want to save the report, it is not possible to be recovered.

.) none
Altough this is possible technically it is not suggested not to let the system write a report any time. You will never have a chance to react on problems or to handle details within the report.

Conclusion:
In fact, for Automic is does not make a difference which function you use, but please consider that some functions will not be available if you use the one or the other setting for your reports. You should consider if.
.) the report is needed for revision purposes
.) the report is needed only in case of error
.) the report should be parsed by a following process within Automation Engine

based on these decisions, you should be able to find out the best solution for your system and which option you can choose the best for you.

Outcomes