possible bug in getnext implementation

hi there
 
I have a MIB implementation that handles a table.
 
Calling snmpgetnext -v2c - c public localhost MyMIB::MyOID.1  to get
index no. 1 works fine.
It works fine for indexes up to 8.
Trying snmpgetnext -v2c - c public localhost MyMIB::MyOID.9 resulted in
skipping to the next column.
 
Debugging this I found that that the index is stored in the string
length value of the varbind - mib.c line 3770: var->val_len = uitmp;
function parse_one_oid_index()
In my iterator handler I'm overiding this value by calling 
snmp_set_var_value(put_index_data, (u_char *)strStationName.c_str(),
strStationName.length() );
which sets the var->val_len to 8 (my string is "Station1")
 
Now the original varbind has value length of 9 and the modified varbind
has 8. This causes table.c : netsnmp_check_getnext_reply() to return 0 (
snmp_oid_compare  fails) and does not fill the ti_info->results
 
As a result netsnmp_table_next_column - table_iterator.c line 704 is
called causing increment of column index
 
 
Questions:
why is the var->val_len used to store index data? Shouldn't the index
variable be used for this?
why does netsnmp_check_getnext_reply causes the name to change by
setting var->name[0] = (oid) var->val_len; in ( mib.c line 3512
build_oid_noalloc->build_oid_segment() ) ?
 
will be happy to provide more information
 
Rotem

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php

_______________________________________________
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