Re: Re: address specifications

From: Neil McKee (neil_mckee@inmon.com)
Date: 01/17/03

  • Next message: Neil McKee: "Announce: version 2.2.32 (KeepFreeMBytes bugfix and RedHat 8.0 install)"

    Hello Tim,

    You are correct. Well spotted. The non-contiguous mask works when it is
    the only one in the list, but fails when a second one is added.

    This will be fixed for the 3.0 release. In the mean time the workaround
    is to make a separate query for each one (or craft your own filter
    string using two "addressMaskEqual" compares instead of one
    "addressMaskInList").

    (FYI, the problem occurs because a list of subnets is loaded into a
    special data structure for rapid searching (important if there are
    thousands of subnets). The search is longest-prefix-first and it
    assumes contiguous masks. A mask like "255.255.0.255" would be
    interpreted as a 32-bit (all 1s) mask because the lsb is 1).

    regards,
    neil

    Tim Durack wrote:
    > I have noticed what looks like an anomaly when using non-contiguous bit
    > masks for address specifications.
    >
    > If I write:
    >
    > destinationAddress=10.1.0.255/255.255.0.255
    >
    > I get all directed broadcasts in the 10.1.0.0 network.
    >
    > If I write:
    >
    > destinationAddress=10.1.0.255/255.255.0.255,10.2.0.255/255.255.0.255
    >
    > I get no data, even though 10.2.0.0 shows directed broadcasts.
    > The filter string looks correct, any ideas why this might happen?
    >
    > Thanks,
    >
    > Tim:>
    >
    > __________________________________________________
    > Do you Yahoo!?
    > Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
    > http://mailplus.yahoo.com
    >
    >

    -- 
    Neil McKee, InMon Corp.
    tel: +1 (415) 661-6343
    http://www.inmon.com
    


    This archive was generated by hypermail 2.1.4 : 01/17/03 PST