On 17/10/06, Roland Stigge <[EMAIL PROTECTED]> wrote: > Another idea: Is the .index file really technically necessary? It looks > like always read and regenerated (and rewritten).
No - I don't believe that's true. The index file is only regenerated if it's older than the timestamp on the directory (i.e. the index file is potentially out of date). Otherwise it's read in. Having these indexes saves having to open *every* *single* MIB file whenever an SNMP application starts up. This is perfectly feasible, but could potentially be quite time-consuming - particularly if someone has a fairly full collection of vendor-specific MIB files (which could run into thousands!) > Why not > generate it on > the fly, keeping it in memory and eliminate .index files? They are > basically file lists of the directories, *And* the name of the MIB contained within each of those files. These may typically be the same as the name of the file - but that's just a convention. The Net-SNMP libraries are quite happy with a mismatch between the file name and the MIB name. So it's not just a matter of reading the contents of the directory - we'd also have to read the first line of each MIB file. Which involves opening every single file - even if the corresponding MIB isn't actually on the list to be read in. The index files avoid the need to do this every single time - it's an efficiency measure. Did you have a chance to look at the patch I've put together to address the FHS issue? Dave ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
