This is a discussion on delegation and blocked transactions within the SNMP Users forums, part of the Networking and Network Related category; This is a multi-part message in MIME format. --===============0687748957== Content-class: urn:content-classes:message Content-Type: multipart/alternative; ...
|
|||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||
|
This is a multi-part message in MIME format.
--===============0687748957== Content-class: urn:content-classes:message Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C8B213.F921D044" This is a multi-part message in MIME format. ------_=_NextPart_001_01C8B213.F921D044 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, Problem: our agent is in the middle between an SNMP Manager and another software block. We don't get to redesign either of those. In one particular use case, the manager sends a GET request, and we delegate that to the other software block. But, in order for the other block to give us the information needed to finish the GET transaction, we depend on a series of SETs from the manager, and they will always come after the GET is initiated. This results in a deadlock situation, where net-snmp agent needs to send out the GET response before processing the SETs, but the software block needs the SETs to be processed before it will answer the GET. I realize this net-snmp behavior is by design, and for good reasons. But, I need to find a workaround for it... Interested in any suggestions, back doors included. Is it possible to map certain OIDs to a different session (but sharing the same endpoints)? And if so, would the sessions behave independently, so that the presence of an outstanding GET in one session not inhibit the completion of the SETs in a different session? The solutions we're looking at right now are pretty ugly. I guess if the session notion doesn't work out, we're looking for the best way to fool net-snmp into thinking the request has been answered (we know one way to do this already, and it's definitely a backdoor), and then, when the needed info is available, send a copy of the original GET request back to net-snmp for us to handle in a non-delegated fashion. Oh, we're using 5.3.1. Thanks Peter ------_=_NextPart_001_01C8B213.F921D044 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Dus-ascii"> <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = 6.5.7652.24"> <TITLE>delegation and blocked transactions</TITLE> </HEAD> <BODY> <!-- Converted from text/rtf format --> <P><FONT SIZE=3D2 FACE=3D"Arial">Hi,</FONT> <BR> <FONT SIZE=3D2 = FACE=3D"Arial">Problem: our agent is in the middle between an SNMP = Manager and another software block. We don't get to redesign either of = those. In one particular use case, the manager sends a GET request, and = we delegate that to the other software block. But, in order for the = other block to give us the information needed to finish the GET = transaction, we depend on a series of SETs from the manager, and they = will always come after the GET is initiated. This results in a deadlock = situation, where net-snmp agent needs to send out the GET response = before processing the SETs, but the software block needs the SETs to be = processed before it will answer the GET.</FONT></P> <P> <FONT SIZE=3D2 = FACE=3D"Arial">I realize this net-snmp behavior is by design, and for = good reasons. But, I need to find a workaround for it...</FONT> </P> <P> <FONT SIZE=3D2 = FACE=3D"Arial">Interested in any suggestions, back doors included. Is it = possible to map certain OIDs to a different session (but sharing the = same endpoints)? And if so, would the sessions behave independently, so = that the presence of an outstanding GET in one session not inhibit the = completion of the SETs in a different session?</FONT></P> <P> <FONT SIZE=3D2 = FACE=3D"Arial">The solutions we're looking at right now are pretty ugly. = I guess if the session notion doesn't work out, we're looking for the = best way to fool net-snmp into thinking the request has been answered = (we know one way to do this already, and it's definitely a backdoor), = and then, when the needed info is available, send a copy of the original = GET request back to net-snmp for us to handle in a non-delegated = fashion.</FONT></P> <P> <FONT SIZE=3D2 = FACE=3D"Arial">Oh, we're using 5.3.1.</FONT> </P> <P><FONT SIZE=3D2 FACE=3D"Arial">Thanks</FONT> <BR><FONT SIZE=3D2 FACE=3D"Arial">Peter</FONT> </P> </BODY> </HTML> ------_=_NextPart_001_01C8B213.F921D044-- --===============0687748957== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757...un.com/javaone --===============0687748957== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Net-snmp-users mailing list Net-snmp-users@lists.sourceforge.net Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/...net-snmp-users --===============0687748957==-- |
![]() |
| Thread Tools | |
| Display Modes | |
|
|