Exlibris' Oracle schema names are also derived from the Metalib version/socket number. The schema you have used (M43_*) needs to be modified in the same way as the perl path, else the script will not return output.
Of course, it just requires a replace pass (in my case, :s%/M43_/M42_/g) but it took me a while (and some oracle hacking) to work it out.
I have some additions i'd like to make to the script, some of which i can see you've prepared for (eg, start date/end date, some). Would you like to review the changes i make, and if so, how should i submit them to you?
Also, there seem to be some discrepancies in the Alerts and Eshelf Records.. will try to work out more and pass any info i discover on to you. (of course, i haven't tried 1.1 yet)
--Daniel
EDIT: i've added the following:
Save to textfile (optional)
I'm currently planning on adding the following:
Start/end dates from CLI
More Stats
Searches (e.g searches in quicksearch only, metasearch only)
Total Failed Searches
Failed Searches Per Database
EDIT: i've been looking through the OracleDB for ways to extract the stats i want, but it looks like i may have to infer some things.. will update when i have determined more.
Also, It never occurred to me to run the (X-8, X-9) clearing utils before the script to get more accurate results (Alerts, Eshelf).
Daniel, you can send me your version by email. I'll review it and get back to you if there are any issues or publish it here.
We run util x-9 every night, so on 1st of any month we can get complete stats for the previous month. Well, we couldn't bypass running it as the temp stuff would fill the system in a couple of days.
Let me know if you find out anything about the discrepancies and whether you can see them in MetaLib Management too.
4 Comments
comments.show.hide2007-12-13
Daniel Tosello
Ere,
Exlibris' Oracle schema names are also derived from the Metalib version/socket number. The schema you have used (M43_*) needs to be modified in the same way as the perl path, else the script will not return output.
Of course, it just requires a replace pass (in my case, :s%/M43_/M42_/g) but it took me a while (and some oracle hacking) to work it out.
Hope this helps
--Daniel
2007-12-13
Ere Maijala
Daniel,
thanks for the notice. I should have used the ORA_USER_PREFIX environment variable... Maybe in the next version
--Ere
Edit: v1.1 available now.
2008-01-24
Daniel Tosello
Great work, Ere, and thanks
I have some additions i'd like to make to the script, some of which i can see you've prepared for (eg, start date/end date, some). Would you like to review the changes i make, and if so, how should i submit them to you?
Also, there seem to be some discrepancies in the Alerts and Eshelf Records.. will try to work out more and pass any info i discover on to you. (of course, i haven't tried 1.1 yet)
--Daniel
EDIT: i've added the following:
I'm currently planning on adding the following:
EDIT: i've been looking through the OracleDB for ways to extract the stats i want, but it looks like i may have to infer some things.. will update when i have determined more.
Also, It never occurred to me to run the (X-8, X-9) clearing utils before the script to get more accurate results (Alerts, Eshelf).
2008-01-24
Ere Maijala
Daniel, you can send me your version by email. I'll review it and get back to you if there are any issues or publish it here.
We run util x-9 every night, so on 1st of any month we can get complete stats for the previous month. Well, we couldn't bypass running it as the temp stuff would fill the system in a couple of days.
Let me know if you find out anything about the discrepancies and whether you can see them in MetaLib Management too.
Thanks!
--Ere