At 10/9/04 3:00 AM, Robert Kehl
Robert L Mathews schrieb:
It looks like only some of the patch got applied, so it didn't solve all the line wrapping issues. :-( I added a full description of the issue to the bug report (explaining why the other part of the patch is necessary) and reopened it; hope that's okay... Sorry to be such a pest.
Let me add that the patch is running smoothly on 1.3, at least for me. I'm auto-patching OTRS with several patches everytime I update it from CVS HEAD and did not encounter any problems so far since weeks.
Unfortunately, the patch in bug 526 still hasn't been fully applied to CVS, so the fix to do proper line wrapping still isn't working. Can one of the OTRS guys tell me what's holding this up? Is there anything I can do to make it more acceptable and likely to be applied? The patch is in danger of getting out of date because other changes are being made to those files. (The same goes for bug 518: OTRS is almost unusable with PostgreSQL as the backend with a large number of tickets, so I submitted a patch that fixes the schema to speed up OTRS dramatically and make OTRS work properly. But the patch appears to be languishing and is in danger of getting out of date so that it doesn't apply cleanly any more.) -- Robert L Mathews, Tiger Technologies http://www.tigertech.net/ "Ignorance more frequently begets confidence than does knowledge." -- Darwin
Robert L Mathews schrieb:
(The same goes for bug 518: OTRS is almost unusable with PostgreSQL as the backend with a large number of tickets, so I submitted a patch that fixes the schema to speed up OTRS dramatically and make OTRS work properly. But the patch appears to be languishing and is in danger of getting out of date so that it doesn't apply cleanly any more.)
ACK, ACK, ACK! When it comes to a few 1000 tickets, a PostgreSQL database is merely not usable anymore due to the schema used. I changed my schema without a problem the last days and am experiencing a great improvement in queriying the db. Durations of the queries are lowered to a third in general, at least for the cost-intensive ones. I highly recommend EVERYONE using a PostgreSQL database applying the patch - you WILL run into trouble after a time - believe me, I have seen it. The OTRS team may hereby feel urged to acknowledge the work of Robert L. Mathews and implement his patches. Additionally OTRS should provide a script to convert an existing PG database to a new, 'integered' version. While we are at it: What about a new module taking names/ids of SQL queries and a few params to perform instead of the text of real queries. Designed that way, we could be able to build high/efficient query modules based on the database used. Just my .02, Robert Kehl PS: If you feel you need assistance in tuning your PostgreSQL database the way described above, contact me via PM or via http://roknet.de/contact
participants (2)
-
Robert Kehl -
Robert L Mathews