Platform: Debian squeeze, samba 2:3.5.6~dfsg-3.
Running testparm yields the warning
rlimit_max: rlimit_max (1024) below minimum Windows limit (16384)
From  I see that this is perfectly harmless and normal
(thanks to Christian Perrier for a pointer) but the warning
message suggests a misconfiguration. The Debian bug log
mentioned above contains a patch to change the message to
rlimit_max: increasing rlimit_max (%d) to minimum Windows limit (%d)
which seems less alarming. That message is still not very clear
but it seems good enough to me.
A possible related problem is that the log level is allowing this
message to be shown while smb.conf does not contain any directives
about log level.
Ping? Is there any further information I can provide here?
To recap, in the default configuration (at least on Debian), testparm produces confusing output that makes debugging by novices unnecessarily difficult. There is a patch available to at least fix the message.
Jeremy, would you like to comment on this one?
Can you attach the debian patch please and I'll review to see if we can apply it.
Created attachment 6269 [details]
s3-param: Make "rlimit_max below minimum Windows limit" notification less scary
Thanks, Jeremy; attached. From <20100125175026.GC23333@samba1>:
"Does setting the fd limit higher result in better behavior, or does
the automatic increase have results that would be counterintuitive if
"If the former, maybe the default rlimit_max should be increased
(on Debian squeeze it seems to be 1024 fwiw). If the latter, I think
the message should say
"rlimit_max: rlimit_max (8192) increased to minimum Windows limit (16384)
"to be more informative and sound less like a warning. Like this,
Steve Langasek's reaction:
"At minimum, the debug message should be cleaned up to state *what* rlimit is
too low; the current message mentions nothing about file descriptors.
"Otherwise, I'm not overly concerned about this. Nowadays, the default log
level is '0', isn't it? So only users who have manually adjusted the
smb.conf 'log level' setting would be seeing this warning."
The default log level is not '0', however, at least on Debian.
Looks good to me for 3.5.next - applied to master and v3-6-test.
Re-assigning to Karolin for inclusion.
Pushed to v3-5-test.
Closing out bug report.
Thanks, Karolin. :)