Bug 6901 - Bad implementation of function dbwrap_change_[u]int32_atomic (in dbwrap_util.c) reduce to "Fatal Error: UID range full!!"
Summary: Bad implementation of function dbwrap_change_[u]int32_atomic (in dbwrap_util....
Status: RESOLVED FIXED
Alias: None
Product: Samba 3.4
Classification: Unclassified
Component: Winbind (show other bugs)
Version: unspecified
Hardware: Sparc Solaris
: P3 critical
Target Milestone: ---
Assignee: Karolin Seeger
QA Contact: Samba QA Contact
URL:
Keywords:
: 6884 (view as bug list)
Depends on:
Blocks:
 
Reported: 2009-11-18 02:09 UTC by Michael Isaev
Modified: 2013-02-18 13:30 UTC (History)
2 users (show)

See Also:

Attachments
Fix for v3-4-test (git patchset) (2.26 KB, patch)
2009-11-20 17:53 UTC, Michael Adam 		vl: review+
Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Michael Isaev 2009-11-18 02:09:11 UTC 
# uname -rpmi
5.10 sun4u sparc SUNW,UltraAX-i2

I have "clean" samba installation (configured and installed from sources as usual).
After logon samba to domain and first run winbindd I see in log.winbindd-idmap:
...
Fatal Error: UID range full!! (max: 20000)
...
Fatal Error: GID range full!! (max: 20000)
...

This errors because in function dbwrap_change_int32_atomic/dbwrap_change_uint32_atomic (file dbwrap_util.c) don't using "byteorder" for new stored values!

Why you using simple assigment:
data.dptr = (uint8 *)&val;
instead byteorder implementation (as in dbwrap_store_[u]int32):
SIVAL(&v_store, 0, v);

This error exist in all versions 3.4.x
Comment 1 Michael Adam 2009-11-20 17:52:26 UTC 
Thanks for your bug report!

(In reply to comment #0)
> # uname -rpmi
> 5.10 sun4u sparc SUNW,UltraAX-i2
> 
> This errors because in function
> dbwrap_change_int32_atomic/dbwrap_change_uint32_atomic (file dbwrap_util.c)
> don't using "byteorder" for new stored values!
> 
> Why you using simple assigment:
> data.dptr = (uint8 *)&val;
> instead byteorder implementation (as in dbwrap_store_[u]int32):
> SIVAL(&v_store, 0, v);

This is simply an oversight. Thanks for spotting this.
I have pushed a fix to branches master and v3-5-test.

Next, I'll attach the fix for v3-4-test for review.

Cheers - Michael
Comment 2 Michael Adam 2009-11-20 17:53:49 UTC 
Created attachment 4972 [details]
Fix for v3-4-test (git patchset)

I guess this should be in 3.4.4.
Comment 3 Michael Adam 2009-11-20 18:29:41 UTC 
*** Bug 6884 has been marked as a duplicate of this bug. ***
Comment 4 Volker Lendecke 2009-11-21 06:57:46 UTC 
Comment on attachment 4972 [details]
Fix for v3-4-test (git patchset)

Looks good, thanks.
Comment 5 Karolin Seeger 2009-11-23 02:01:27 UTC 
Pushed to v3-4-test.
Closing out bug report.

Thanks!