We are happy to announce to the NWN Community that the Avlis Team has developed a real-time Open Database Connectivity (ODBC) interface. Unlike previous systems, which utilize either tokens to hold information or creatively read the log file and re-compile it into the module, the Avlis Persistence System (APS) and the underlying NWN Extender (NWNX) read and write information from a database on the fly and make information stored in the database available in real-time. Supports Windows NT, 2k, and XP.
[Update v1.12: fixed a potential problem when NWN server crashed] [Update v1.21: added code to APS and NWNX that prevents potential startup problems] [Update v1.22: added support for italian, some statistics, SQL error messages, and some minor fixes.] [Update v1.23: NWNX licensed under the GPL, source code is included.]
A few remarks/questions: - when I compile the source code, there's nothing wrong (just a warning on link, about PSAPI.DLL), but my NWN_Extender.exe file is 'only' 52KB large instead of 60KB for yours (for the release version, in debug mode, it's much more). As I kept all options unchanged, I were just a bit surprised of such a big difference. - don't you think that the fact you start NWServer.exe in debug mode could significatively increase CPU usage ? I've seen several echanges about the fact the APS raised CPU usage of NWServer.exe. Couldn't it be the point ? Maybe that's the reason you tried to experiment another alternative (#define AttachVsDebug) ? FF
Posted by FastFrench at 2003-02-1805:01:00
Great, thanks :) As 1.28 is not allready available in French, I'll finally have a look on APS. Maybe I won't be able to live without it soon afterall... Thanks a lot Papillon, I'll keep you informed if I can find some good improvements out of it. But at least I will include my server monitoring codes.
Posted by Papillon at 2003-02-1716:01:00
Often requested, now available: As promised, the source code to NWNX is included in the current distribution of aps (v1.23). Now I'm pretty curious what will come out of it...
Posted by Silk at 2003-02-1709:52:00
@FastFrench: The new CopyObject function facilitates a Token based system. You're deluding yourself if you think that even a good token based system would make APS "not of much use anymore". FYI: We're currently running a beta version of NWNX2 on one of our servers. Papillon has really outdone himeself again.
Posted by FastFrench at 2003-02-1703:24:00
With V1.28 (copyobject function) I feel like such heavy tools like APS are not of much use anymore. Hopefully I won't wait for APS source code anymore... :)
Posted by Arcanius_NWH at 2003-02-1617:20:00
aye we are using the latest, we are trying hard to narrow down where it goes wrong but so far without luck. just dont work..... can you contact me on icq 3420449?
Posted by Darkhavin at 2003-02-1616:23:00
Arcanius, I have the same problem: Found 6 bytes for requests, 16384 bytes for responses. (stop) (start) Found 6 bytes for requests, 16384 bytes for responses. (rinse repeat 4-6 times) then: Found 1024 bytes for requests, 16384 bytes for responses. What's up? (I've been using 1.22)
Posted by Papillon at 2003-02-1616:15:00
Arcanius, are you using the current version ? There is a script in there (aps_ready) that makes sure no queries are executed before NWNX is ready. Btw, NWNX2 is on the way, which will take everything to a whole new level. We won't see problems like this anymore ;- ).
Posted by Arcanius_NWH at 2003-02-1614:23:00
Has there been any serious testing as to why the initialization stage sometimes reports 0 bytes for requests? My module size problem now seems to be less of that and more of certain areas breaking the init phase. With hundreds of areas this is quite a bit of a pain to test. Why would the request bytes reserve break and not the send bytes reserve?
Posted by Darkhavin at 2003-02-1211:16:00
Pelemele: The problem is, when they were created you used large tag names that were truncated because they exceeded the size limitation of the blueprint resref field. Check em.
Site
