Dave: As is my typical behavior, I had noted in my posting on 20040723 that I am using RH9, Net-SNMP 5.0.9 (and all of the tools it contains), and in the two subsequent postings (dated 20040723 and 20040724), that I am working to build an AgentX extension to implement the 802.11 MIB. The problem with Perl is therefore contained within the 5.0.9 distribution, which I downloaded from the Net-SNMP project pages on SourceForge. Hence, I can only assume that the problem continues to exist in the 5.0.9 distribution. Now, you had mentioned in todays email that the development team thought that the problem is now solved. You noted that the problem was addressed on or about 20030819, so I should expect that the 5.0.9 distribution does not suffer from the noted problem with PerlIO_fprintf. After all, 5.0.9 was released on 20030925, over a month after the date you mention. Thus, I must ask, why is it that I see this problem in the 5.0.9 distribution? Also, even if the problem has been solved in some release, why is is not also detailed in a posting to the net-snmp-users list. For one, such a posting would serve as a form of documentation for the tool. For two, even though may users fail to review the content of the users list archives, this is not true for all users, and so those of us who do use the archives would be able to service our problems more often, and rely less often on direct communication with the developers.
Part of my complaint is directed at those other users who have solve the problem, owing to their terse (incomplete) solution suggestion. It is true that I tend to be a bit verbose, and while others may find my postings to be too extensive, it is also true that my postings are more informative than are those posted by others. I think that users who have previously solved the PerlIO_fprintf problem did so by upgrading the Net-SNMP distribution with a fresh distribution of Perl. The problem is, those users did not say so, at least explicitly. This lack of complete information is a problem, for it is sure to cause those who rely upon newsgroup postings to act without full knowledge. Hence, such postings are counter-productive to the effective use of Net-SNMP, to its wider adoption, and to the users of Net-SNMP. That is, such postings do more harm than good. As for the other problems, those with mib2c, these include the observation of the name ANONYMOUS#4 in both the .h and .c outputs of mib2c, and the failure of the produced code to compile with gcc. I suspect these are all related problems. Moreover, as far as I can tell, the code produced by mib2c in Net-SNMP v4.2 is identical with regard to the FindMethod statements. Please, see my earlier emails for the exact details. It is far easier to do that, than to wade through multiple statements of the problem. All related postings by me are no more than 45 days old. I will look to verify that the FindMethod statements in the v4.2 code include trailing semi-colons but, as I have already said, the .h file looks to be similarly constructed in v4.2 and v5.0.9 of Net-SNMP. William R. Buckley ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click _______________________________________________ Net-snmp-users mailing list [EMAIL PROTECTED] Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users
