Bug #39910 | mysql_init() writes to another memory | ||
---|---|---|---|
Submitted: | 7 Oct 2008 13:47 | Modified: | 8 Oct 2008 17:28 |
Reporter: | Anatoly Lyutin | Email Updates: | |
Status: | Can't repeat | Impact on me: | |
Category: | MySQL Server: C API (client library) | Severity: | S2 (Serious) |
Version: | 4.1.21 | OS: | Linux (Red Hat) |
Assigned to: | CPU Architecture: | Any | |
Tags: | mysql_init() |
[7 Oct 2008 13:47]
Anatoly Lyutin
[7 Oct 2008 14:01]
Anatoly Lyutin
{ char buf[100]; sprintf( buf, "%d ", nCurFlag2[k] ); fwrite( buf, strlen(buf), 1, fd ); /* One fwrite */ } copy/paste error! Sorry!
[7 Oct 2008 20:04]
Sveta Smirnova
Thank you for the report. You complain about mysql_init, but print values of nCurFlag2[k]. Does this mean your example would work as expected if not initialize tCurMysql? Will error occur if skip Apache part and run test like: int main() { mod_init(); }; ? Could you also please try with current version 4.1.23 and current GA version 5.0.67?
[8 Oct 2008 8:07]
Anatoly Lyutin
>You complain about mysql_init, but print values of nCurFlag2[k]. Does this mean your >example would work as expected if not initialize tCurMysql? Yes. The main tips is that if after tCurMysql we have more than 2 array, the zero element damages only in the first array. :( >Will error occur if skip Apache part and run test like: >int main() { > mod_init(); >}; >? No, I have checked this. >Could you also please try with current version 4.1.23 and current GA version >5.0.67? I can not try at this version unfortunately, but in 5.0.51a it works without errors. Should I repost this issue into Apache bugzilla?
[8 Oct 2008 9:57]
Sveta Smirnova
Thank you fro the feedback. > Should I repost this issue into Apache bugzilla? Yes, it could be bug in Apache as well, although I additionally check at MySQL side.
[8 Oct 2008 10:16]
Anatoly Lyutin
Thanks you for your attention.
[8 Oct 2008 17:28]
Sveta Smirnova
Thank you for the feedback. Tried with current development 4.1 sources and could not repeat the problem. So I close the report as "Can't repeat". Feel free to reopen the report when you are able to try latest 4.1 and if bug is repeatable with it.