Regarding SET on Tables having External Index. - SNMP

This is a discussion on Regarding SET on Tables having External Index. - SNMP ; Hi, This is a basic question on how an agent should behave. Well, I'm having a table(say table B) which is having two indices, out of which one is external (index from table A). When I receive a set reqeuest ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: Regarding SET on Tables having External Index.

  1. Regarding SET on Tables having External Index.

    Hi,

    This is a basic question on how an agent should behave. Well, I'm
    having a table(say table B) which is having two indices, out of which
    one is external (index from table A). When I receive a set reqeuest on
    table B, (this SET request will actually insert a new row), should I
    check the availability of the instance in table A ? Can I add the new
    row in table B alone (just ignoring the table A) ? or should I throw
    any error in this case indicating the non-availability of the instance
    in table A ?

    Any RFC reference would be greatly appreciated.

    Regards,
    Siva.

  2. Re: Regarding SET on Tables having External Index.

    HI,

    What you decide to do is based on whether or not you want to
    enforce referential integrety. What would you do in the CLI?
    Does it "make sense" to create a row in table A (are there
    appropriate values that you can use for the columns in table A)?

    In most cases, the appropriate course of action is to fail
    the set request until the row in table A exists.

    On Thu, 4 Nov 2004, Siva wrote:
    > Hi,
    >
    > This is a basic question on how an agent should behave. Well, I'm
    > having a table(say table B) which is having two indices, out of which
    > one is external (index from table A). When I receive a set reqeuest on
    > table B, (this SET request will actually insert a new row), should I
    > check the availability of the instance in table A ? Can I add the new
    > row in table B alone (just ignoring the table A) ? or should I throw
    > any error in this case indicating the non-availability of the instance
    > in table A ?
    >
    > Any RFC reference would be greatly appreciated.
    >
    > Regards,
    > Siva.
    >

    Regards,
    /david t. perkins

  3. Re: Regarding SET on Tables having External Index.

    vs_lax@yahoo.com (Siva) wrote in message news:<41105e74.0411040108.6b7be8b9@posting.google.com>...
    > Hi,
    >
    > This is a basic question on how an agent should behave. Well, I'm
    > having a table(say table B) which is having two indices, out of which
    > one is external (index from table A). When I receive a set reqeuest on
    > table B, (this SET request will actually insert a new row), should I
    > check the availability of the instance in table A ? Can I add the new
    > row in table B alone (just ignoring the table A) ? or should I throw
    > any error in this case indicating the non-availability of the instance
    > in table A ?


    No Standard for this I believe. It is implementation specific and can
    be followed in the following ways:

    1. If the instance is not available in table A, then create it table A
    when the SET request comes for table B. That is, u shud also create a
    row for table A with that instance when a SET request issued for table
    B.

    2. Or throw an error according to the standard for SET request PDU
    failure ( RFC1905):


    "Otherwise, if the variable binding's name specifies a variable
    which does not exist but can not be created under the present
    circumstances (even though it could be created under other
    circumstances), then the value of the Response-PDU's error-status
    field is set to `inconsistentName', and the value of its error-
    index field is set to the index of the failed variable binding."

    The only condition is table B with external index should not contain
    the instance without having it table A.



    >
    > Any RFC reference would be greatly appreciated.


    No. Sorry, I am not aware.
    :-)


    Thanks,
    Karthik. N

  4. Re: Regarding SET on Tables having External Index.

    Hi David,

    thanx for the info.

    dperkins@snmpinfo.com wrote in message news:...
    > HI,
    >
    > What you decide to do is based on whether or not you want to
    > enforce referential integrety. What would you do in the CLI?


    > Does it "make sense" to create a row in table A (are there
    > appropriate values that you can use for the columns in table A)?


    BTW, I thought of inserting the row in table A with some default data
    for the columnar nodes depending on the type of nodes (and for the
    index nodes also. Anyway for the external indeces I will get the
    values from the SET request to table B)

    > In most cases, the appropriate course of action is to fail
    > the set request until the row in table A exists.


    Ok. I will verify this and throw appropriate error.

    >
    > On Thu, 4 Nov 2004, Siva wrote:
    > > Hi,
    > >
    > > This is a basic question on how an agent should behave. Well, I'm
    > > having a table(say table B) which is having two indices, out of which
    > > one is external (index from table A). When I receive a set reqeuest on
    > > table B, (this SET request will actually insert a new row), should I
    > > check the availability of the instance in table A ? Can I add the new
    > > row in table B alone (just ignoring the table A) ? or should I throw
    > > any error in this case indicating the non-availability of the instance
    > > in table A ?
    > >
    > > Any RFC reference would be greatly appreciated.
    > >
    > > Regards,
    > > Siva.
    > >

    > Regards,
    > /david t. perkins


  5. Re: Regarding SET on Tables having External Index.

    Hi karthi, thanx.. and i decided to follow the second choice.

    easprasanna@yahoo.co.in (Karthikeyan.N) wrote in message news:<737324a4.0411042136.15e97029@posting.google.com>...
    > vs_lax@yahoo.com (Siva) wrote in message news:<41105e74.0411040108.6b7be8b9@posting.google.com>...
    > > Hi,
    > >
    > > This is a basic question on how an agent should behave. Well, I'm
    > > having a table(say table B) which is having two indices, out of which
    > > one is external (index from table A). When I receive a set reqeuest on
    > > table B, (this SET request will actually insert a new row), should I
    > > check the availability of the instance in table A ? Can I add the new
    > > row in table B alone (just ignoring the table A) ? or should I throw
    > > any error in this case indicating the non-availability of the instance
    > > in table A ?

    >
    > No Standard for this I believe. It is implementation specific and can
    > be followed in the following ways:
    >
    > 1. If the instance is not available in table A, then create it table A
    > when the SET request comes for table B. That is, u shud also create a
    > row for table A with that instance when a SET request issued for table
    > B.
    >
    > 2. Or throw an error according to the standard for SET request PDU
    > failure ( RFC1905):
    >
    >
    > "Otherwise, if the variable binding's name specifies a variable
    > which does not exist but can not be created under the present
    > circumstances (even though it could be created under other
    > circumstances), then the value of the Response-PDU's error-status
    > field is set to `inconsistentName', and the value of its error-
    > index field is set to the index of the failed variable binding."
    >
    > The only condition is table B with external index should not contain
    > the instance without having it table A.
    >
    >
    >
    > >
    > > Any RFC reference would be greatly appreciated.

    >
    > No. Sorry, I am not aware.
    > :-)
    >
    >
    > Thanks,
    > Karthik. N


  6. Re: Regarding SET on Tables having External Index.

    Hi
    that is why you usully see "augment" in index field. So that one has
    to check the entry for the index in the TableA , before actually
    creates the entry in TableB.This augmenting is by standard rfc.

    Regards
    KMJ



    vs_lax@yahoo.com (Siva) wrote in message news:<41105e74.0411152324.1356b34@posting.google.com>...
    > Hi karthi, thanx.. and i decided to follow the second choice.
    >
    > easprasanna@yahoo.co.in (Karthikeyan.N) wrote in message news:<737324a4.0411042136.15e97029@posting.google.com>...
    > > vs_lax@yahoo.com (Siva) wrote in message news:<41105e74.0411040108.6b7be8b9@posting.google.com>...
    > > > Hi,
    > > >
    > > > This is a basic question on how an agent should behave. Well, I'm
    > > > having a table(say table B) which is having two indices, out of which
    > > > one is external (index from table A). When I receive a set reqeuest on
    > > > table B, (this SET request will actually insert a new row), should I
    > > > check the availability of the instance in table A ? Can I add the new
    > > > row in table B alone (just ignoring the table A) ? or should I throw
    > > > any error in this case indicating the non-availability of the instance
    > > > in table A ?

    > >
    > > No Standard for this I believe. It is implementation specific and can
    > > be followed in the following ways:
    > >
    > > 1. If the instance is not available in table A, then create it table A
    > > when the SET request comes for table B. That is, u shud also create a
    > > row for table A with that instance when a SET request issued for table
    > > B.
    > >
    > > 2. Or throw an error according to the standard for SET request PDU
    > > failure ( RFC1905):
    > >
    > >
    > > "Otherwise, if the variable binding's name specifies a variable
    > > which does not exist but can not be created under the present
    > > circumstances (even though it could be created under other
    > > circumstances), then the value of the Response-PDU's error-status
    > > field is set to `inconsistentName', and the value of its error-
    > > index field is set to the index of the failed variable binding."
    > >
    > > The only condition is table B with external index should not contain
    > > the instance without having it table A.
    > >
    > >
    > >
    > > >
    > > > Any RFC reference would be greatly appreciated.

    > >
    > > No. Sorry, I am not aware.
    > > :-)
    > >
    > >
    > > Thanks,
    > > Karthik. N


+ Reply to Thread