RE: More error-status codes on SNMPv1?
At 3/31/2000:08:49 AM, Beecher Adams wrote: Hi, This thread belongs on [EMAIL PROTECTED] and probably not on [EMAIL PROTECTED] Moreover, I haven't been following this thread real closely and I'm not sure who owns the following quoted material but: >The normal use of error-index is to indicate which varbind may have been the >source of the error. I suspect many agents don't get this fancy though and >don't make use of the field. Very wrong. Only a very broken agent would not use the error-index field to point to the varbind for which an error-status is being returned. >I'm proposing to use this field instead as an index to an error details table. This is a non-starter due to error in your suspicion above. Cordially, BobN ISO 9001 Registered Quality Supplier --- Bob Natale | ACE*COMM | 301-721-3000 [v] Dir, Net Mgmt Prod | 704 Quince Orchard Rd | 301-721-3001 [f] [EMAIL PROTECTED]| Gaithersburg MD 20878 | www.acecomm.com - Free downloads at www.winsnmp.com -
RE: More error-status codes on SNMPv1?
See below, -Original Message- From: Abhishek Bagchi [mailto:[EMAIL PROTECTED]] Sent: Friday, March 31, 2000 3:00 AM To: Beecher Adams; [EMAIL PROTECTED] Cc: Shivendra Kumar; [EMAIL PROTECTED]; Amit Srivastava Subject: Re: More error-status codes on SNMPv1? Hi, Thanks for your valuable comments.Please find my comments inline. Beecher Adams wrote: > Although the error-status codes may not be extended in v1, what you could do > is implement an error details table under your private mib. When you issue > one of the standard error codes, the manager could then go query the error > details table to find out more specific information on why the error > occurred. The error-index field which is part of the get-response PDU could > be used as the index into the error table. This allows an unambigous > correlation between the error PDU and the corresponding details in the > table. This seems to be a good idea.Do you mean to say that an entry in this table will be created whenever the agent implementation populates the error index field of the response PDU with a varbind index? Yes I feel the columner objects in this table can be the v1 specific error-status and the value of the columner object can indicate to the specific error condition . But, the questions that come to my mind are: 1.If the index to the table entry is the error-index in the response PDU , some other response PDU may have the same error-index value for a different varbind.So, there is a problem with unique indexing for the table entries. The solution to this could be to include the request ID also in the index information as the request ID is unique to any request PDU. The normal use of error-index is to indicate which varbind may have been the source of the error. I suspect many agents don't get this fancy though and don't make use of the field. I'm proposing to use this field instead as an index to an error details table. The table could be of a fixed size such as 50 entries that will wrap. It is the agents responsbility to increment the error-index for each new snmp error that occurs. This way there is no ambiguity issue. 2.Assuming that the above problem is solved how does the mgmt entity querry this table? The most obvious way seems to be a next operation as the stn doesn't know which particular object to query in that entry.What do you feel? The error details table is just a regular snmp table, in your enterprise branch. Having the index from error-index, a simple get operation can be performed. 3.Since every time the error-index is populated an entry is created the number of entries can swell up at times. So we have to devise a way of periodically deleting entries from this table.How do you think we decide upon this? Simply use a fixed size table that wraps, e.g. index = 1, then 2, then ..50, then 1, etc. There would never be more than 50 outstanding snmp errors, or again whatever number you choose. There could be a number of other questions that may come to one's mind but right now I'm not able to think beyond these. Your valuable opinions can surely open it up a lot more. Thanks > > > Beecher Adams > Metro OptiX > 2201 Avenue K > Plano, TX 75074 > Email: [EMAIL PROTECTED] > Phone: 214.299.4214 Fax: 214.299.4289 > > -Original Message- > From: Abhishek Bagchi [mailto:[EMAIL PROTECTED]] > Sent: Thursday, March 30, 2000 2:21 AM > To: [EMAIL PROTECTED] > Subject: More error-status codes on SNMPv1? > > Hi all, > > We are working on SNMP interfaces for a product named Flash600 ( ADX, > Layer2 switching, ATM) for FNC Inc. We are using SNMPv1 framework , but, > unfortunately SNMPv1 supports only following error-status codes: > > noError(0): no error in the requested PDU. > toobig(1): The get-response message is bigger than that the local > implementation can handle. > noSuchName(2): one of the requested objects does not match anything in > the relevant MIB view that can be returned. > badValue(3): The set-request asked the agent to write an inappropriate > value. > readOnly(4): A set-request tried to write a value that the operator is > not allowed to write.Either the access specified is > READ-ONLY or the the variable MIB definition does not permit write > access. > genErr(5): A variable cannot be retrieved for reasons outside the ones > listed above. > > This provides very little granularity for the User to decide what went > wrong. > > Is there any other way we can add more error-status codes without > violating v1 > compliance? > We don't want to move over to SNMPv2 ,but, still want to add more > error-status codes? > Can we add more error-status codes? If so, how? > > Thanks & regards, > Abhishek > > > Ab
Re: More error-status codes on SNMPv1?
Hi, Thanks for your valuable comments.Please find my comments inline. Beecher Adams wrote: > Although the error-status codes may not be extended in v1, what you could do > is implement an error details table under your private mib. When you issue > one of the standard error codes, the manager could then go query the error > details table to find out more specific information on why the error > occurred. The error-index field which is part of the get-response PDU could > be used as the index into the error table. This allows an unambigous > correlation between the error PDU and the corresponding details in the > table. This seems to be a good idea.Do you mean to say that an entry in this table will be created whenever the agent implementation populates the error index field of the response PDU with a varbind index?I feel the columner objects in this table can be the v1 specific error-status and the value of the columner object can indicate to the specific error condition . But, the questions that come to my mind are: 1.If the index to the table entry is the error-index in the response PDU , some other response PDU may have the same error-index value for a different varbind.So, there is a problem with unique indexing for the table entries. The solution to this could be to include the request ID also in the index information as the request ID is unique to any request PDU. 2.Assuming that the above problem is solved how does the mgmt entity querry this table? The most obvious way seems to be a next operation as the stn doesn't know which particular object to query in that entry.What do you feel? 3.Since every time the error-index is populated an entry is created the number of entries can swell up at times. So we have to devise a way of periodically deleting entries from this table.How do you think we decide upon this? There could be a number of other questions that may come to one's mind but right now I'm not able to think beyond these. Your valuable opinions can surely open it up a lot more. Thanks > > > Beecher Adams > Metro OptiX > 2201 Avenue K > Plano, TX 75074 > Email: [EMAIL PROTECTED] > Phone: 214.299.4214 Fax: 214.299.4289 > > -Original Message- > From: Abhishek Bagchi [mailto:[EMAIL PROTECTED]] > Sent: Thursday, March 30, 2000 2:21 AM > To: [EMAIL PROTECTED] > Subject: More error-status codes on SNMPv1? > > Hi all, > > We are working on SNMP interfaces for a product named Flash600 ( ADX, > Layer2 switching, ATM) for FNC Inc. We are using SNMPv1 framework , but, > unfortunately SNMPv1 supports only following error-status codes: > > noError(0): no error in the requested PDU. > toobig(1): The get-response message is bigger than that the local > implementation can handle. > noSuchName(2): one of the requested objects does not match anything in > the relevant MIB view that can be returned. > badValue(3): The set-request asked the agent to write an inappropriate > value. > readOnly(4): A set-request tried to write a value that the operator is > not allowed to write.Either the access specified is > READ-ONLY or the the variable MIB definition does not permit write > access. > genErr(5): A variable cannot be retrieved for reasons outside the ones > listed above. > > This provides very little granularity for the User to decide what went > wrong. > > Is there any other way we can add more error-status codes without > violating v1 > compliance? > We don't want to move over to SNMPv2 ,but, still want to add more > error-status codes? > Can we add more error-status codes? If so, how? > > Thanks & regards, > Abhishek > > > Abhishek Bagchi > Wipro Technologies-Telecom Solutions > #72,Electronics City,Bangalore-29, > India > Tel:91-80-8520408/0416 Ext-2108 > Fax:91-80-8520478 > --- > Applying Thought > -- Abhishek Bagchi Wipro Technologies-Telecom Solutions #72,Electronics City,Bangalore-29, India Tel:91-80-8520408/0416 Ext-2108 Fax:91-80-8520478 --- Applying Thought
Re: More error-status codes on SNMPv1?
Hi - > Message-ID: <[EMAIL PROTECTED]> > Date: Thu, 30 Mar 2000 13:51:09 +0530 > From: "Abhishek Bagchi" <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Subject: More error-status codes on SNMPv1? ... > Is there any other way we can add more error-status codes without > violating v1 > compliance? At the level of the protocol, no. > We don't want to move over to SNMPv2 ,but, still want to add more > error-status codes? > Can we add more error-status codes? If so, how? ... I would suggest using SNMPv3. Randy Presuhn [EMAIL PROTECTED] http://www.bmc.com/ Voice: +1 408 546-1006 BMC Software, Inc. 1-3141 2141 N. First Street Fax: +1 408 965-0359 San Jose, California 95131 USA Any relationship between my opinions and BMC's should be coincidental.
More error-status codes on SNMPv1?
Hi all, We are working on SNMP interfaces for a product named Flash600 ( ADX, Layer2 switching, ATM) for FNC Inc. We are using SNMPv1 framework , but, unfortunately SNMPv1 supports only following error-status codes: noError(0): no error in the requested PDU. toobig(1): The get-response message is bigger than that the local implementation can handle. noSuchName(2): one of the requested objects does not match anything in the relevant MIB view that can be returned. badValue(3): The set-request asked the agent to write an inappropriate value. readOnly(4): A set-request tried to write a value that the operator is not allowed to write.Either the access specified is READ-ONLY or the the variable MIB definition does not permit write access. genErr(5): A variable cannot be retrieved for reasons outside the ones listed above. This provides very little granularity for the User to decide what went wrong. Is there any other way we can add more error-status codes without violating v1 compliance? We don't want to move over to SNMPv2 ,but, still want to add more error-status codes? Can we add more error-status codes? If so, how? Thanks & regards, Abhishek Abhishek Bagchi Wipro Technologies-Telecom Solutions #72,Electronics City,Bangalore-29, India Tel:91-80-8520408/0416 Ext-2108 Fax:91-80-8520478 --- Applying Thought