After a mass ingest of content into a fresh system (and I confirmed with 
fresh record creations), if the `checker-email` task runs without first 
properly running the `checksum` task, the outcome is alarmist and 
surprising and (I believe) incorrect.

Every item that is 'yet to be checked' is added to the report with this 
header message:

The following is a UN-CHECKED BITSTREAM REPORT report for2/1/16 [*note the 
> lack of space before the date*]
> To add these bitstreams to be checked run the checksum checker with the -u 
> option [*I believe this line to be incorrect*]


The main problem
There does not appear 
<https://wiki.duraspace.org/display/DSDOC5x/Validating+CheckSums+of+Bitstreams> 
to be a "-u" option for the `checker` routine. Attempting to run `dspace 
checker -u` generates a FATAL log message. There does exist a "-u" flag for 
the `checker-email` task, but it simply generates the email with the 
un-checked report by itself -- *not* what is described in the report. I 
believe a more accurate message would be communicate the instructions to 
run the checker.

I realize it's a little nebulous and tough to condense those instructions 
but the checker does need to hit those records one way or another and there 
is no "-u" flag to make it easier. Maybe it is sufficient to point to the 
documentation or mention running "checker -l" to make sure everything gets 
seen.

Lack of Success / All is Well Confirmation
Now that I have processed every item, the checksum checker sends no 
message. Is there an option to generate an "all-clear" kind of message with 
the `checker-email` task? An "email anyway" option? The dspace log file 
even seems silent on `checker-email` completion. If no email is going to be 
sent, I recommend leaving an INFO note to reassure anyone that looks for 
one. An INFO note might be nice either way such as "INFO sending checker 
email results to ___" and "INFO no need to send an email for the checksum 
checker results. All is well."

A Dream Combo - email results immediately.
It would be really neat to create a flag that allows you to run the checker 
immediately followed by the `checker-email` routine rather than having to 
schedule two tasks separately and hope the one completes before the second 
begins. That stacking might be easier in Linux Cron job management but I 
don't see a way to chain them together in Windows' task scheduler.

Bonus question
Finally, for my sanity, on a Windows server running DSpace 5, is it safe to 
run the checker as "dspace checker -l -p" and do the flags operate the same 
as "dspace checker -lp"? I intend to run it weekly over every item and have 
the pruning happen automatically.

-- 
You received this message because you are subscribed to the Google Groups 
"DSpace Technical Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/dspace-tech.
For more options, visit https://groups.google.com/d/optout.

Reply via email to