Patch Fix List?

[Former Member]

Hello all,

I just downloaded the 7.6.03.14 patch from the service marketplace download area... we're probably going to apply this in the next day or so. I was wondering if there still existed a "patch fix list" somewhere? A year or so ago I discovered such a listing, back when the MySQL folks were still involved.

Specifically, ever since we applied the 7.6.03.08 patch two months ago, we've been plagued by what our research suggests is a precompiler problem. Every once in a while, one of our users will run an ABAP program that ends up dumping with SQL error -4000. Our research seems to indicate this is a bug in the precompiler that our 46D_EXT kernel is using (our dev_w* logs say we're using precompiler version 7.3.1.021).

We've found a couple of SAP notes that suggest we could set SHAREDSQL to "no" to get around these dumps, but we're trying to avoid doing that for fear of the performance hit we'd take. It would be great if such a "patch fix list" still existed somewhere to predict if applying this patch might help us.

Thanks,

~Fred

[markus_doehr2 · Accepted solution]

Check

806690 - Feature list of liveCache/MaxDB 7.6

You will find the PTS corrections mentioned under http://www.sapdb.org/webpts

Markus

[Former Member]

Thank you both for your responses. They were both quite helpful.

[lbreddemann]

Hi Fred,

the site mentioned by Markus Döhr is the correct place to check what changes had been implemented between two versions of MaxDB.

Please use the "My Release Notes" option to generate the difference reporting between any two releases.

The featurelist notes are only meant to provide a overview over the most important changes and do not take in account if a specific build had ever been released.

Concerning your "fear": with Kernel 4.6D (and with all Kernels that use the irecompiler instead of SQLDBC) it's recommended to switch off shared sql. See note #1070232 for details.

(You can just copy and past the note number into the search field in the upper left of this SDN site to view the note).

Since the precompiler already uses cached statements on session level there shouldn't be such a big performance issue if you don't use shared sql.

KR Lars