< afsdb-reports Documentation | Russ Allbery > Software > Orphaned Software > AFS Reporting DB | Stanford::LSDB > |
Bump large access count report to 10 million
We have lots of volumes that regularly get over one million accesses.
Lines changed: +3 -3
Removed testing address.
Lines changed: +2 -3
Case-sensitve match for the "RW volumes relocated" report.
Lines changed: +3 -2
Added case sensitivity to a GROUP BY statement, where things such as group.SAY and group.say were considered the same.
Lines changed: +2 -2
Change to the "new changes to unreleased changes" metric, to make sure we get distinct rows.
Lines changed: +2 -2
Changed SQL statments to run sanely under MySQL.
Lines changed: +99 -66
Updated AFS reporting scripts for lsdb-new, with changed ID strings for cvs export for Debian packaging, code cleanup, and changes to work with newer Stanford::LSDB for MySQL.
Lines changed: +8 -6
Add a copyright statement at the top of the script as well.
Lines changed: +4 -1
Use the latest version printing code and add a COPYRIGHT AND LICENSE section.
Lines changed: +12 -2
Add a link to the distribution web page.
Lines changed: +6 -1
Minorly obscure the afs-reports e-mail address since this is going up on the web.
Lines changed: +5 -5
Modify the query for 1000 volumes to use the new partition_count view that includes the volume type and handle both RW and RO types.
Lines changed: +3 -3
Use the view name volume_count rather than double_volumes.
Lines changed: +2 -2
We don't care about additions and deletions of backup volumes, and we don't care about growing volumes of type other than RW.
Lines changed: +6 -3
Helpful to only grab those volumes from double_volumes where count > 1. Otherwise, you get every volume in the cell, which makes for a long report.
Lines changed: +3 -2
When selecting from partition_count, you can just select out the count; you don't need count(*).
Lines changed: +2 -2
Send mail to afs-reports at stanford.edu rather than afs-reports@lists.
Lines changed: +4 -4
Only report on volumes that grew by 200MB, not just 50MB. Report on any duplicated volumes and on any server/partition pairs with more than 1000 volumes on them.
Lines changed: +33 -15
Remove unnecessary error handling now that Stanford::LSDB always sets RaiseError. Use SQL parameters wherever possible rather than interpolating strings into the SQL block.
Lines changed: +3 -6
Add a report for volumes with over one million accesses, and remove the report about people who are over quota by more than 50MB.
Lines changed: +19 -20
use afs-reports at lists.stanford.edu
Lines changed: +3 -3
Make -h work more reliably and fix a documentation paste mistake.
Lines changed: +5 -5
Send the report mail from root, and only send mail if asked to send mail.
Lines changed: +10 -7
Add the ability to send mail to afs-reports, add command-line parsing, and add documentation.
Lines changed: +82 -1
Working version, still sending reports to stdout. It works much better now, but it doesn't print out as many nice headings.
Lines changed: +32 -44
Initial version, sending the report to stdout. We think that it works, but we're not really sure yet. But it prints out nice headings.
< afsdb-reports Documentation | Russ Allbery > Software > Orphaned Software > AFS Reporting DB | Stanford::LSDB > |