Sorry for my previous (cryptic and errorful) reply I sent it from my phone while I was away, and it auto-corrected my text. :-) The long form of my reply is:
I have seen this type of behavior before. In one of my applications I have a table of (up to) 1000 entries. While building the table and when its only a few entries long, I can query at a rate of 5 per second, but at at few hundred entries it slows to 1 per second, and by the time I reach 1000 entries in the table, 'it' has slowed down to 1 query per multiple seconds. Unfortunately I haven't had the time (yet?) to investigate where the issue is: - in my code - in the way I use the agentX interface to Net-SNMP - in the agentX interface in Net-SNMP - in Net-SNMP itself. So yes, extending your timeouts is a workaround, but, IMHO... it shouldn't have to be that slow. On Wed, Jul 2, 2014 at 10:50 AM, sarath azad <saraths...@yahoo.co.in> wrote: > Hello, > > Thanks for the replies and inputs. > The problem was that the table being of very large size, SNMPWALK was taking > more time to finish. > When we configured the view to contain only a single column of the table, > and then doing SMNPWALK on that column of the table, we were getting the > data from that particular column. Once that column is done, SNMP-get-next > (internal to SNMPWALK) tries to get the next column data. But that column is > not there in the view, so it continues to search for next valid data present > in the view , till the end of the table. As the table is of very large size, > it takes very long time to return and in the mean time SNMPWALK query gets > timed out. > > Initially we tried with : "snmpwalk -t 10 ... " but we go same timeout > problem. But when tried with "snmpwalk -t 20 ..." it worked . > So there is no problem with our code, its the general behavior only. > > thanks and regards > S Sarath > > On Wednesday, June 25, 2014 11:28 PM, Fulko Hew <fulko....@gmail.com> wrote: > > Can't saw if this is yr problem but I have seen that it has linear? Delays > when dealing with large tables via agent. Perhaps the problem is inherent? > On Jun 24, 2014 5:28 PM, "sarath azad" <saraths...@yahoo.co.in> wrote: > > Hello, > > Could you please help in resolving the below problem? > > If we are configuring a view containing a single MIB table and do a SNMPWALK > on it, it is showing all the data on the table and at the end of the table > it is showing a message: > "No more variables left in this MIB View (It is past the end of the MIB > tree)" > > If this table is a small table with few entries it is showing above message. > And if it is a big table with some 4 thousand odd entries it is showing > "Timeout: No Response from <device - ip>". > > When looking at the snmp packets exchanged, For the query on the smaller > table, after the table's last entry is sent, another snmp get-next is sent > to the device with the last entity of the table. Now as the next entity is > outside the view (since our table is the last item in the view) it is > getting message that no more variables in the view. But if the table is big > with few thousands of entries, this last snmp get-next with the last index > is not responded. After a few retries we are getting Timeout message. > > Could you please help in resolving this issue, as why this timeout is > showing for large tables? > > thanks and regards > S Sarath ------------------------------------------------------------------------------ Open source business process management suite built on Java and Eclipse Turn processes into business applications with Bonita BPM Community Edition Quickly connect people, data, and systems into organized workflows Winner of BOSSIE, CODIE, OW2 and Gartner awards http://p.sf.net/sfu/Bonitasoft _______________________________________________ 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/listinfo/net-snmp-users
